Benko, Pal - The Benko Gambit - A Dynamic Winning Strategy for Black.pdf

THE

BEMKO GAWBIT PAL BENKO EXPLAINS A DYNAMIC WINNING STRAHGY FOR BLACK

by 1nternational Grandmaster PAL BENKO

SIDNEY FRIED, Publisher BURT HOCHBERG, Editor-in-Chief DENNIS FRIED, Assistant Editor

RUM PRESS a division of RHM Associates of Delaware, Inc.

220 Fifth Avenue, New York, N.Y. 10001

Copyright

© 1973

RHM CHESS PUBLISHING a division of RHM Associates of Delaware, Inc.

220 Fifth Avenue, New York, N.Y. 10001 All rights reserved. No portion of this book may be reproduced in any form, except for brief pas­ sages in a review, without the written consent of the publishers.

ISBN 0-89058-029-4 Library of Congress Catalog Card No: 75-18495

Printed in the United States of America

You ore invited to subscribe to Tre RHM Survey of

CURRENT CHESS OPENINGS som · e of the advantages of our Survey: You select for pur­ chase only the Openings sections in which you have an interest. (List of Openings on next page)

�

Each section will be up-to-the-minute, reasonably priced, published loose-leaf style in a most attractive presenta­ tion (binder supplied with first order in any sectioQ) and will include the following:

100 most recent grandmaster games in each of the Open­ ings you choose;

� Selected for their importance and significance from many hundreds of games in that Opening by a top international grandmaster known for his expertise in that area;

� All

games annotated by leading grandmasters, including many by the players themselves;

� Current analysis in depth of some of the most theoretically

significant games by grandmaster members of our Board of Contributing Editors, our Board having the following distinguished membership, listed alphabetically:

Board of Contributing Editors of the R.H.M. Survey of Current Chess Openings Svetozar Gligoric Vlastimil Hort Borislav lvkov Anatoly Karpov

Lubomir Kavalek Viktor Korchnoi Bent Larsen

Tigran Petrosian Lajos Portisch Boris Spassky

(continued on next page)

List of Openings (Covering all Important Variations in each Opening)

Sicilian

Pirc

King's Indian

Alekhine

Larsen's

Griinfeld

Ruy Lopez

Queen's Gambit

Dutch

Nimzo-lndian

Benoni

King's Gambit

English

French

Queen's Indian

Caro-Kann

Benko Gambit

Reti

(additions will be made)

All "Chess Opening" theory is in a perpetual stage of change, some lines being successfully challenged and discarded, other lines improved, new and promising lines being contin­ ually discovered as thousands of games are played in current grandmaster tournaments. Not only can your own game in your favorite Openings be greatly improved by study of the 100 current games in the Openings section selected by you, but you will gain new and valuable insights into the middle game play and end game play flowing naturally from each line through the individual game annotation and analysis-in-depth by the many world­ famed grandmasters who will be serving on our Board of Contributing Editors. The average cost of each full-size section containing all we have just described should be modest, but send no money­ only your name and address on a postcard-so that you will be entered as a subscriber to receive announcements and full descriptions of each Openings section as they become ready for shipment. There is no charge for entering this subscrip­ tion, and it puts you under no obligation. You later order only what you wish to order. But you can help us (and yourself) by listing on the postcard the 5 top choices of Openings you would like to see covered. This informal "straw poll" will guide us in the order of pub­ lication of individual Openings sections. We are now preparing publication of sections covering some of the most popular Openings and commencing work on all the rest, and to receive announcements of each section as it becomes available, merely send your full name and address on a postcard to: Dept. 2

R.H.M. SURVEY OF CURRENT CHESS OPENINGS 840 Willis Avenue, Albertson, New York 11507 In Europe write to R•H•M PRESS LTD. • P.O. Box 55 London N13, England 5BE

Table of Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

History of the Gambit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7

Games: Bronstein vs. Lundin . . . . . . . . . . . . . . . . . . . . . . . . .

7

Szabo vs. Lundin . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

Taimanov vs. Bronstein . . . . . . . . . . . . . . . . . . . . . . .

15

Theory o f the Gambit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

22

A: The Gambit Deferred . . . . . . . . . . . . . . . . . . . . . . . . . . . .

23

B : White Gives Up Castling . . . . . . . . . . . . . . . . . . . . . . . . . .

29

C: White Retains Castling . . . . . . . . . . . . . . . . . . . . . . . . . . .

42

D : Quiet Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

54

Fianchetto Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

59

F: Transpositions to the Gambit . . . . . . . . . . . . . . . . . . . . . .

74

G. The Gambit Declined

82

H: The G ambit Reversed

92

E:

Twenty Selected Games Played by Pal Benko . . . . . . . . . . . . . .

95

Lis t of Games . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

96

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Ou tline of Variations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143

THE

BEMKO GAWBIT PAL BENKO EXPLAINS A DYNAMIC WINNING STRATEGY FOR BLACK

THE

BENKO

GAMBIT

Introduction

In the chapters that follow we will examine thoroughly a modern, dynamic gambit which, after years of personal experience, I can heartily recommend to my readers. But before proceeding with the analysis and the complete games, let us take a look at the character­ istic position after the first few moves : 1 2 3 4 5

P-Q4 P-QB4 P-QS PxP PxP

N-KB3 P-B4 P-QN4! P-QR3 BxP

This is the usual starting position o f the so-called Benko Gambit. The question naturally arises here : What sort of compensation does Black have for the sacrificed pawn? 1) He has an advantage in development. Not only has Black overtaken White in development, he has developed effectively: h is B ishop on QR3 hinders White's efforts to carry out the natural developing move P-K4. In addition, a hidden advantage is the fact that Black's Queen Rook stands on a useful open file without hav­ ing had to move. 2) Black's pawn structure is compact but not inflexible. He has the possibilities of P-K3, P-KB4 and-after White's eventually necessary P-QN3-the move P-QB5, freeing QB4 for a piece, creat­ ing the possibility of exploiting Q6, and breaking the defense of White's Queenside. 3) The pawn sacrifice has removed an important White central 1

THE BENKO GAMBIT pawn-the QBP-which makes the natural White central break­ through (P-K4-K5) more difficult because his Queen pawn wiii be undefended. Even if such a breakthrough should become feasible, its potency will be less than usual because of the absence of the QB pawn. Furthermore, Black will be able to gain the advantage of a mobile pawn center by the possibility of P-K3 , whether White replies P X P or allows Black to play P X P. In the latter case, White's Queen pawn will be weakened still more because, having already advanced his King pawn and exchanged his QB pawn, there will be no pawns to lend it support. 4) It is a simple matter to achieve an active setup for the Black pieces in which they cooperate excellently. The two Rooks occupy­ ing the QR and QN files, supported by the "Indian" Bishop on KN2, exert great pressure against White's Queenside which will be in­ creased by the deployment of the Queen and Knights. It is con­ ceivable that Black will prevail either on the Queenside or on the King's wing, once having drawn White's pieces to defend the Queenside, or even in the center by means of a well-timed break­ through. 5) It is a psychological advantage that Black can at once seize the opportunity to direct the course of the game and that he can force White to play defensively for a long time. During this phase of the struggle it is quite easy for White to commit an error which may soon have serious repercussions. From all this it can be learned that White's game is difficult (!) and that he must play very patiently and cautiously, compared with which Black has the initiative and a positional advantage as wel l. White must secure his Queenside before he can think of active play and the carrying out of the P-K5 breakthrough or the realization of his extra pawn on the Queenside. The loss of either the QR pawn or the QN pawn will usually mean more than merely the restoration of the material balance, for after such a loss the neighboring pawn will be fatally weakened. Exchanges, normally favoring the player with a material superiority, must in this case be avoided by White both for reasons of defense and to maintain possibilities for counter­ attack. Summing up the above points, we may state that the gambit is easier to play with Black than with White. It is anticipated that this will hold true for a considerable time. Long before the era of gam2

THE BENKO

GAMBIT

bits came to an end, most of them were already being reduced in effectiveness by the development of chess technique. Most such gambits were based on the idea of a quick Kingside attack made possible by an advantage in development, which was obtained by the sacrifice of material. But when it was discovered that by means of appropriate exchanges and the return of sacrificed material Black could refute the attack and emerge with the better position, those gambits fell into decline and eventual disuse. However, because those technical means do not apply to the Benko Gambit, it has had a lengthy period of remarkable success. This gambit is not based on tactical tricks or traps, but on solid positional ideas. It is a completely modern gambit which creates its own laws and possibilities. I have been adopting the gambit regularly in professional tourna­ ment practice for about seven years. I have played it almost without exception whenever the opportunity presented itself, and against all kinds of opposition. And I have made no secret of it. The best of the games I have played with it were published in the pages of Chess Life & Review* with my comments. I have as much as announced publicly that anyone can play against me with an extra pawn within a few moves of the opening. Despite this challenge, in the last several tournaments in which I have partici­ pated during the preparation of this book I had hardly any oppor­ tunities to adopt my favorite gambit. My opponents preferred to avoid the gain of a pawn and adopted an even more modest opening than the regular Queen's Gambit. My results show a prevailing plus with Black. I have lost only two games, but neither one because of the opening. As White I h ave played against the gambit three times and I must confess that I felt most uncomfortable. I could win only one game with White, and then because I surprised my opponent with a new continuation. In the meantime, many grandmasters, including a world cham­ pion, have adopted this "hot," dynamic gambit, and it can there­ fore be found in most serious tournaments. For this reason every player-those who may wish to adopt it and those who may have to face it-must become familiar with its theory. Nevertheless, this gambit must be understood rather than merely

•

Published

monthly

by

the U.S. Chess Federation,

New York 12550.

3

479

Broadway,

Newburgh,

THE

BENKO

G A M B IT

memorized. Its consequences are felt throughout the middle game and even into the endgame; I have therefore considered it useful to give full games where characteristic motifs are demonstrated. Even in the analytical sections, actual games are given frequently to their conclusion, for the same reason. I would like to clarify what is meant by "Benko Gambit." Many often speak of it incorrectly as the Volga Gambit. That name itself makes little sense since I have never heard of a player with that name nor of a tournament on the river Volga. I have even dis­ cussed this matter with several Soviet grandmasters, but they too considered the origin of that name unclear. They did agree, how­ ever, that the name Volga Gambit refers only to the treatment of the pawn sacrifice with Black playing an early P-K 3 , which is rather similar to the old Blumenfeld Gambit. The Volga Gambit differs completely from the Benko Gambit. The former gives the game an entirely different character by blow­ ing up the center immediately, whereas in the Benko Gambit Black fianchettoes his King Bishop and attacks on the Queenside. Tai­ manov's book, Damengambit bis Holliindisch, published in German in 1970, and the Russian magazine Shakhmatny Biulletin (1971, #5) treat only the Volga Gambit (with P-K3 by Black). But that gambit is not considered to be totally correct by most masters and it very rarely occurs in modern tournaments. It is a pity to confuse these two openings whic h have com­ pletely different goals--even the pawn sacrifice is offered for dif­ ferent reasons. Because of this, the Volga Gambit is not dealt with in this book. In the beginning I tried to popularize this opening under the name Benoni Countergambit, but the name did not stick as players began to call it the Benko Gambit. I must add, of course, that I have never claimed to have been the first to adopt it; in the chapter dealing with the history of the gambit I explain its origins in inter­ national chess. There can be no doubt, however, that I have made it popular with my games and articles-due to them it has become an accepted opening in international practice since I developed the idea that White must take this gambit into consideration before deciding upon his opening move. For this reason, some literature on the gambit already exists. Finally, allow me to quote from the Swedish book on the "Volga Gambit" (a misnomer): 4

THE

BEN KO

GAMBIT

The Volga Gambit has had an absolute renaissance in the last few years. The Hungarian-born [a m istake-! was actually born in France but raised in Hungary. PB] Amer­ ican Grandmaster Pal Benko has made a lot of "public relations" for the Benoni Countergambit as he calls it. He has written many articles on it but the most important thing is that he himself often plays the variation.

5

THE

BENKO GAMBIT

History of the Gambit

It is hardly possible to state precisely who first adopted the gambit. Some Swedish sources mention that it first occurred there in the 1 920s and that tournaments especially for the gambit were organ­ ized later. The three internationally known masters of Sweden's past-Stahlberg, Stoltz and Lundin-also used it, although Stahlberg found it not to his taste either as White or as Black. Indeed, the Swedish players appear to have used the Kingside fianchetto, if not the modern order of moves. Of course, the gambit may have occurred elsewhere as well, but it was undoubtedly Lundin who in­ troduced it into international tournament practice. Nevertheless, it later fell into oblivion. The earliest examples of the gambit in serious competition are the games Bronstein-Lundin, Szabo-Lundin, and Taimanov-Bron­ stein, which are analyzed below. The iirst example in a serious international tournament was the game between the Soviet star David Bronstein and Erik Lundin at the Interzonal in Saltsj obaden 1 948. Its importance at that time is attributable to the fact that first place in that critical tournament was actually decided by it! Swedish International Master Erik Lundin (b. 1 904) played it against both Bronstein and Laszlo Szabo, the two tournament leaders. Bronstein was able to win, but Szabo, tied with Bronstein going into the final round, lost against Lundin ( and the gambit ! ) and had to be content with second place. These two games, of some historical importance, are worth seeing. They illus­ trate the state of theory as regards the gambit at that time.

Saltsjobaden 1948 White: David Bronstein Black: Erik Lundin 1 2 3 4 5

P-Q4 P-QB4 P-QS N-QB3 P-K4

N-KB3 P-B4 P-Q3 P-KN3 P-QN4!? 7

THE

BENKO

GAMBIT

As we can see, the most effective order of moves in the gambit had not yet been found-it seems Black wanted White to play P- K4 before offering the pawn with P- QN4. For some time this was considered necessary to avoid other lines in which White could advantageously avoid playing P- K4. Opening treatises which deal with special systems are inherently unable to keep up with the development of theory, so Pachman and Euwe, for example, still consider this the main line. This despite the fact that I pointed out long ago that this view is out of date, and despite the fact that I have proved in practice that Black should not fear those once- feared possibilities. 6 PxP 7 B-K2

B-KN2 P-QR3

Now Black is correct to wait for White's Bishop to move. so as to gain a tempo if White captures on QN5 or QR6. Of course, the waiting move 7 N- B3 would have been more exact. 0-0

8 N-B3 9 PxP

This capture is not at all necessary; 9 0-0 P X P 1 0 B X P is better. B XP

9 10 BxB

Nor is this capture urgent, as we will see when we analyze later games. More correct is 10 0-0, since the text makes possible the plan 10 . . . R X B 1 1 0-0 QN- Q2, followed by 1 2 . . . Q-R 1 and 1 3 . . . KR-N 1 , which is an active setup for Black. This can be prevented, though only temporarily, by 1 1 Q-K2 instead of 1 1 0-0. 10 1 1 0-0

NXB N-Q2

Black strives to hinder the central breakthrough by P-K5 . For this purpose the indirect defense 1 1 . . . N-B2 was to be taken into consideration, with pressure on Q4. 1 1 . . . R- N 1 is not to be recommended because of 1 2 P-K5 P X P 1 3 N x P N x P? 1 4 8

THE

BENKO

G A M BIT

Q x N Q X Q 15 N X Q B X N 16 N X Pch and N-B6 !

12 B-NS!

An uncomfortable move wh ich-as is often the case-gains a tempo by attacking the King pawn. In case of 1 2 . . . P-R3 both 1 3 B- B4 (preparing P-K5) and 13 B- R4, maintaining the pressure, are to be considered . A further driving of the Bishop by P-KN4 would be sensible if Black could then exchange his Knight for the B ishop by a later N- KN5 or N-KR4. But since this will not be possible here, P-KN4 would mean only a weakening of Black's castled position. Nevertheless, 12 . . . P-R3 is playable so as to force White to make that decision and to prevent the possibility of a later B-KR6. 12

0

0

0

0

0

R-Nl

0

This move, in <;onnection with the next Rook move, is already a compromise. This is the moment the initiative goes over to White. Ideally, this square belongs to the King Rook so that both Rooks may exert pressure on the Queenside . 13 14 15 16

KR-Kl Q-R4 N-B2

Q-Q2 QR-N l KR-Bl B-R6

After this Black has an unpleasant choice to make: if he allows his King Bishop to be exchanged h is Kingside will be weakened, 9

THE

BENKO

GAMBIT

but if he avoids the exchange the White Bishop will stand well and undisturbed on KR6, offering White chances for mating attacks on the eighth rank, as will be seen later in fhe game. B-B3 R-N6

16 17 P-QR3

The Rook has apparently been able to occupy the hole on QN6 but this is only temporary as the Rook will have to retreat with loss of tempo. For this reason it was better to prepare the doubling of Rooks by 17 . . . R-N2, since 18 P-QN4 was not playable because of 1 8 . . . QxRP. 18 19 20 21 22

Q-B2 N-Q2 N-B4! Q-R4! NxQ

KR-N1 R/ 6-N2 Q-R3 QxQ N-N4?

But this is a serious blunder which allows White to free h imself on the Queenside. 22 . . . R-R2 23 P-QN3 and now 23 . . . N-N4 or 23 . . . N-K4 would have offered prospects for a long resistance. 23 P-QN4!

White at once seizes the opportunity to eliminate the pressure on the QN file. This can be carried out as shown in the following lines: 23 . . . R-R2 24 PxP ! R X N 25 P-B6 N-B 1 26 P-B7 R-B 1 27 N-N6, or if 24 . . . PxP 25 P-K5 NxKP 26 N X N B X N 27 N X P N X P 2 8 R X Rch B X R 2 9 N-R6 !, winning a piece. 23 24 K-B1

by

N-QS B-N2

Black tries to exchange the Bishop as all counterplay is obviated the weakness of his first rank. But even this doesn't help. 25 B-K3 26 P x P 27 R X R

R-R1 RXR PxP 10

THE

BENKO GAMBIT

Accepting the piece sacrifice by 27 . . . R X N is also hope­ less, e.g. 28 P-B6 N-B l 29 N-N6 R-R2 30 N-Q7 ! P-B4 31 N X N K X N 3 2 R-N7 R X P 33 B X N B X B 34 P-B7, etc. 28 29 30 31 32 33 34

N/R4-N6 P-QR4 RXN NXR K·Kl K-Q3 K-B4!

R-N1 NxN RXR N-N6 B-B6 B-R4 Resigns

In his game with Szabo from the same tournament, Lundin im­ proves on the order of moves and is rewarded with success, although he has some cooperation from White. Saltsjobaden 1948 White: Laszlo Szabo Black: Erik Lundin 1 2 3 4 5 6 7

N-KB3 P-B4 P-QN4!? P-QR3 P-N3 BXP P-Q3?!

P-Q4 P-QB4 P-QS PXP PXP N-QB3 P-K4

It is interesting that Lundin, having improved on his game with Bronstein by playing this order of moves (apparently for the first time), which I now recommend, does not follow the purpose of the gambit which he could have done by exchanging on KB8, depriving White of the right to castle. Even though White would still be able to "castle by hand," this would mean a loss of time. Also, White's King could be later subjected to an attack. 8 BXB 9 N-B3

NxB B-N2

11

THE BENKO GAMBIT 10 0-0 11 B-B4

KN-Q2

We have arrived at a position from the previous game one tempo ahead, but here, instead of the more provocative B-N5 White has substituted B-B4. The text is good enough, but castling is not to be avoided sooner or later. White wishes to carry out the central break­ through at once. 11 12 Q-Kl 13 KR-B1

0-0 Q-B2

This is not a logical preparation for P-K5, which would be better served by 1 3 KR-K 1 , and on 1 3 KR-N 1 14 QR N l . -

13 14 QR-N1

KR-N1 B x N?!

A surprising decision, but it is not easy for Black to find a plan because of the unfortunate situation of his Knight on QR3 . It can be activated only via QB2, but if the Queen leaves its post White may advantageously push P-K5 . Black should probably have tried the regrouping by R-N2 and Q-N l . But this would have taken time.

15 R X B?

White insists on his connected pawns, but this move reduces his opening edge. On 1 5 P X B the pawn would be weak but White 12

THE

BENKO

GAMBIT

would get a free hand in the center and on the Kingside, for in­ stance 15 ... Q-R4 1 6 P-K5 R X R 17 R X R Q X BP (if 17 . .. P X P 1 8 N X P N x N 1 9 B X N P-B3 20 B X P !, etc . , with advantage for White) 1 8 P x P P x P 1 9 Q-K7, and White has a strong attack. Black would thus be condemned to defending himself with the passive 1 5 ... P-B3 which would be met by 1 6 P-KR4 with some discomfort for Black. 15 16 N-Q2

Q-R4!

Obviously White has trusted in this move, which seemingly pro­ tects QR2 and QB3 ( 1 6 . . . Q x P ? 1 7 R-QR3). Nevertheless, Black can take the QN pawn: 1 6 ... R x P ! ? 1 7 R X R Q x R, and the penetration of the Rook by 1 8 R-N7 is not as dangerous as it looks because of 1 8 . . . Q-B8ch. But Black above all intends to improve the position of his Queen Knight. 16 17 R-QR3 18 R X R 1 9 P-QR3

N-B2 Q-N3 RXR N-N4

Black's Knight is headed for Q5. The natural Knight move 20 N-B3, to defend Q4, would be met by 20 . . . N-B6. R-RS!

20 B-K3 21 R-QB1

But now 21 N-B4? could be answered by 21 . . . R X N 22 Q X R N x P. 21 22 B x N 23 N-B3

N-QS P x B!

White's situation is not easy, even though he has two connected passed pawns which only temporarily require defense. The analysis of this game, when it was first published, included a question mark appended to this move, instead of which 23 Q-Q 1 and 24 Q-N3 13

THE

BENKO GAMBIT

were recommended. But then Black would have had a favorable Exchange sacrifice: 23 Q-Q l Q x P ! 24 R-B8ch K-N2 25 Q X R QXN and Black threatens both mate and 2 6 . . . N-N 3 . I f 25 N-B4 Q-R7 26 Q x Pch N-K4. N-B3 R-R4 P-Q6!

23 24 Q-B2 25 N-Q2 26 Q-B7?

White stiii believes he has some advantage, being a pawn ahead, and he underestimates the strength of the Queen Pawn. It was high time to ensure the draw by 26 Q x P Q X P 27 R-N l . Besides the text move, incidentally, there was another interesting possibility: 26 Q-B8ch K-N2 27 N-34 Q-B4 28 N X R? Q X Q 29 R X Q P-Q7 and wins. On 27 Q-B3 R-B4 28 N-B4 R X N 29 Q X R Q X P, and again the Exchange sacrifice offers good chances. But now the tables are turned. 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

QxP QxN PxP K-N2 QxQ R X QP P-N4 PxP P-B4 K-B3 R-K4ch R-K7ch RXP RXP R-N1 P-BS K-B4 P-B6 R-K1

P-KS?! PxN P-KR4 Q-B3 RXQ R-B1 K-B1 R-R1 P-R4 K-K1 K-Q2 KXP P-RS P-R6 K-B4 K-QS K-B6 K-N7

White resigned, since Black will give up his Rook for the QR 14

THE

BENKO GAMBIT

pawn, after which White will be helpless against the KB pawn and KR pawn, both protected by the Black King. These two games are not entirely convincing, of course. Hence the pawn sacrifice did not become fashionable. As an active player myself at the time, I considered Lundin's sacrifice a daring but unjustified extravagance, the more so because he was unable to avoid last place in the tournament in any case. Interestingly, Bronstein himself, obviously recognizing the diffi­ culties for White, chose the gambit in an important international tournament and was rewarded with full success. This happened in the Candidates' Tournament at Zurich in 1 9 5 3 , in the very first round. Later, Bronstein wrote an excellent book on the tournament, which, I am sorry to say, has yet to be published in English. Here is that noteworthy game from Bronstein's book, with his full comments for the first time in English.*

Zurich 1953 White: Mark Taimanov Black: David Bronstein

It is difficult to be objective when annotating one's own games. Variations favoring the annotator always appear interesting-one talks about them willingly and in great detail. But variations which favor the opponent are generally not clear. One usually looks for (and usually finds) justification for one's own mistakes, while those of the opponent appear natural and therefore require no explana­ tion. I wish to state, before I begin to annotate my game from the first round, that Black had no advantage for a long time, almost until the very end.The reason White loses this game is psychologi­ cal: he fails to notice the critical point of the game, when it was time to think of a draw. As for the purely chess reasons, I shall try to clarify them in my notes. 1 P-Q4 2 P-QB4 •

N-KB3 P-B4

Translation courtesy Burt Hochberg.

15

THE

BENKO GAMBIT

3 P-QS 4 N-QB3 5 P-K4

P-KN3 P-Q3 P-QN4

What does Black hope to get for the sacrificed pawn? First, he wants to weaken the advance-guard of the White pawn chain, i .e. Q5, and then, after the inevitable P-QR3 followed by White's cap­ ture of the QR pawn, to obtain a good diagonal for the Queen Bishop (QR3-KB8), which would have fewer prospects on its original diagonal. Additional justification for the sacrifice are the two open lines which allow Black active play against the White QN pawn and QR pawn. The effectiveness of the Black Bishop on KN2 should not be omitted inasmuch as in this system Black intends to keep his King pawn on K2. The scope of the King Bishop's activity will thus be automatically increased. Another interesting strategic idea, also inherent in other systems of the King's Indian, is the development of the Queen Rook w ithout moving it. There are, of course, negative aspects of the sacrifice-the main one being the loss of a pawn. If White is gradually able to over­ come his difficulties he will then have a clear advantage in the endgame. For this reason, the variation was not used again in this tournament. I went into it, however, partly because I did not want to start the tournament with a long defense, to which Black is usually condemned. B-KN2 0-0 P-QR3

6 PXP 7 N-B3 8 B-K2

Until now one might have hoped to regain the pawn. Now it is a real sacrifice. 9 PXP 1 0 0-0

BxP

I once had a similar position, playing White against Lundin in Saltsjobaden 1 948. I exchanged Bishops and Black recaptured with the Knight, eventually transferring it to Q5 via B2 and N4. This maneuver seemed to be too slow for Black. Therefore, in this game I recaptured with the Rook, intending to develop the Queen 16

THE

BENKO GAMBIT

Knight via Q2-N3-R5 and exchange it for White's Queen Knight, so as to weaken the defense of his QR and QN pawns. Q-B2 QN-Q2 RXB

10 11 R-K1 12 B x B 1 3 Q-K2

Taimanov had originally intended P-K5 i n order t o break up Black's pawn position, but it cannot be played for tactical reasons. Instead of White's 1 1 th and 1 3th moves, B-KB4 and Q-Q2 would have offered greater prospects, after which the King Rook could have been placed on QB 1 and the Queen Rook on QN l , having in view the gradual movement of all the pieces and pawns away from the long diagonal. It would also have been advantageous to prepare the move P-QN4 and to start the actual realization of White's extra pawn. Incidentally, that is the way I played against Lundin and I still think it was the proper plan. KR-R1

13 14

P-KR3

Taimanov was not satisfied with the previously mentioned 1 4 P-K5 because of 1 4 . P x P 1 5 N x P N X N 1 6 Q x N Q x Q 1 7 R X Q K-B l . .

.

N-N3 N-K1

14 1 5 B-NS 16 B-Q2

17

THE

B E N KO

GAMBIT

1 6 P-K5 does not work because of 1 6 . . . P X P 1 7 N x P P-B3 . But retreating the Bishop was not advisable; better was 1 6 P-R3, offering a pawn for the purpose of getting rid of Black's Bishop, for instance 1 6 . . . B X N 1 7 P X B R x P 1 8 QR-B l . Black would still have been active on the Queenside but his King would have been surrounded by dangerous weaknesses on the dark squares. 16 17 N x N 1 8 B-B3

N-RS RXN BXB

I had n o desire t o capture o n R 7 because White then reaches his goal : 1 8 . . . R X P 1 9 R X R R X R 20 P-K5, and Black has no targets on the Queenside while White is getting dangerous play in the center. 19 P X B 20 Q-Q3

Q-R4 Q-R3

Black's superiority in the ending is based on the fact that the anchor pawn on K2 is easily defended as it is still in the rear, whereas White's pawns on QB3 and K4 are excellent targets for Black's Rooks. Let us examine the following variations to illustrate: 21 Q x Q R/ 1 x Q 22 R-K2 N-B3, or 22 P-K5 N-B2, or 22 QR-N 1 R X RP 23 R-N8 R-R 1 24 KR-N 1 R-R8. Taimanov does the right thing by not exchanging Queens on R6, and it would have been better to avoid the exchange later. 18

THE

BEN KO

21 Q-Q2 22 R X R 2 3 P-KS

GAMBIT R x RP QXR

White has either overestimated his chances i n the ending or has simply failed to consider Black's 24th move. 23 24 N x Q 2S RxP

QxQ PXP K-B1

Black has a weak pawn on QB4 and all White has to do now to reach a theoretical draw is to exchange Knights and give up both Queenside pawns for the Black pawn, as four pawns against three in a Rook ending is usually not enough to win, although it is no easy matter to draw. 26 N-N3

Grandmaster Taimanov is an optimist by nature. Having j ust about overcome his difficulties, does he really think he can win the Black pawn? Now the more fitting move for the ending would have been K-B 1 , minimizing the danger of losing. 26 27 N-BS 28 K-R2

P-BS R-R8ch N-B3!

Meeting the threat of N-Q7ch and threatening to attack White's QB pawn. It suddenly becomes clear that Black can successfully attack on the Kingside as well.The Rook at the opponent's rear is a power to contend with. N-Q2 R-R7

29 N-K4 30 R-NS 31 R-N4

White almost lost his Knight. If 3 1 K-N3 P-B4. If 3 1 K-N 1 R-K7 3 2 N-N3 R-K8ch 3 3 K-R2 P-B4. 19

THE

BENKO GAMBIT P-B4 N-N3 NxP

31 32 R-B4 33 N-NS 34 R-Q4

I f White takes the pawn o n his B 4 h e loses the KB pawn, while the attempt to win the Black KR pawn would be met as follows : 34 R X P R x P 35 N X Pch K-N2 36 N-N5 R X Pch 37 K X R N-K6ch, and Black wins the Knight ending. 34 35 36 37 38 39 40

N-N3 K-N2 P-R3 K-B2 N-RS NxP

R-Q8ch P-B4 N-K6ch N-Q4 R-QBS RXP

Were the Knights off the board the game would end i n but now the Black Knight succeeds in forking. 40 41 N-B3

a

draw,

N-Q4

White defends the King Bishop pawn, but . .. R X Pch R-KB7

41 42 K-R1 White resigned.

Even this game found no following. The gambit was rarely adopted in tournaments after this game, as it was still considered dubious. But its significance was proven in that it was demonstrated that Black does not have to start an attack at all costs, even a pawn down, but may calmly exchange pieces, aiming for the better ending. To quote again from the Swedish book, Volga Gambit: " It is indeed strange that no one took the variation seriously after this game, until Benko started to play it some years later." 20

THE

BEN KO

GAMBIT

The first game in which I used it was against Vukic in Saraj evo 1 967 (the game will be found in the Append ix), and I am still using it consistently and with consistent success. The gambit is still an effective weapon against the Queen Pawn opening today. For example, in the most recent chess Olympiad (Skopj e 1 972), the Benko Gambit scored ten wins, four draws, and no losses! This was the most successful opening variation in the huge tournament. More and more masters and grandmasters are using it to strike for the initiative with the Black pieces ; to mention only some of them: Spassky, Browne, Szabo, Bronstein, Bilek, Vasiukov, Gheorghiu, Parma, Tringov, Gaprindashvili, et al. They are constantly adding to the theoretical development of the gambit, finding improvements for both sides, and providing new examples of dynamic chess. It looks as though the Benko Gambit has found a permanent place in chess theory, not only in the past, but in the present and future as well, for as long as our wonderful game exists.

21

THE

BENKO GAMBIT

Theory of the Gambit Due to increasing interest, many games utilizing the theme of this gambit have been collected in recent years. Systematizing the analysis is not easy since many lines may transpose to each other. Finally I decided to divide the material into chapters according to the treatment of the diagonal from Black's QR3 to KB8, as follows: Black does not occupy the diagonal with his Queen Bishop at once, somewhat postponing the gambit (old variation) . . . . . . . . . . . . . . . . . . . . . . . . . . . Page B: White Gives Up Castling. Black occupies the diagonal at once with his Queen Bishop. White challenges it at the cost of his right to castle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page C: White Retains Castling. Black occupies the diagonal at once as in (B). White challenges it only after he ensures castling (8 N-Q2) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page D: Quiet Variation. White does not abandon the diagonal but keeps it for himself (5 P-K3) . . . . . . . . . . . . . . . . . . . . . . Page E: Fianchetto Variation. White gives up the diagonal and develops his King Bishop on KN2 . . . . . . . . . . . . . . . . . . . Page F : Transpositions. The Gambit is reached through other openings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page G: The Gambit Declined. White does not accept the offered pawn Page H ; Reversing Colors. The Gambit as adopted by White . . Page A: The Gambit Deferred.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

22

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

23

29

42 54 59 74 82 92

THE

BENKO

GAMBIT

A

The Gambit Deferred 1 P-Q4 N-KBJ 2 P-QB4 P-B4 3 P-QS P-QJ'?! 4 N-QBJ P-KNJ 5 P-K4 P-QN4!'? 6 P X P

I n the following pages w e will examine the possibilities when Black postpones the pawn sacrifice (as in Bronstein-Lundin, given earlier). In this case Black waits until White has played P-K4, in order to avoid the lines with White's P-K3 or P-KN3 . Although this order of moves is not the most effective for Black, one should nevertheless seek new possibilities to improve the old variations. In any case it is usef ul to examine them since one may get into these lines in other ways. 6

.

. . P-QRJ(!'?)

It is usual to postpone this move until White's King Bishop has declared itself . Black may thus try to gain a tempo by forcing the Bishop to move a second time in order to recapture on QN5 . The game Schiffer-Hug, Berlin 1 97 1 , continued : 7 P-B4!'? B-KN2 8 N-B3 0-0 9 P X P B X P 1 0 B X B R X B? (N X B is better) 1 1 0-0 Q-N3 1 2 Q-K2 QN-Q2 (Q-N2) 1 3 P-K5 (White stands better) 1 3 . . . P X P 1 4 P x P N-N5 1 5 B-B4 KR-R 1 1 6 QR-K 1 P-B5ch 1 7 K-R 1 Q-B4 1 8 P-KR3 N-R3 1 9 Q-Q2 ! P-N4 2 0 N x P R-KN3 2 1 NB-3 , Black resigns. White may try to avoid the loss of a tempo by capturing the QRP at once, but then Black must answer with some other develop23

THE

BENKO GAMBIT

ing move and recapture on QR3 only after White moves his King Bishop. Gustavson-Akvist, Vaxjo 1963, continued: 7 P x P 8-KN2 8 8-Q3 0-0 9 KN-K2? ! (9 N-B3 is more natural) B X P 1 0 B X B N X B 1 1 0-0 Q-B2 1 2 B-B4 KR-N 1 1 3 Q-Q2 R-N5 1 4 P-B3 QR-N 1 1 5 QR-N 1 Q-N2 (The game i s equal) 1 6 P-QR3? (P-QN3 !) R-N6 ! (Black now stands a little better. Had he taken the NP at once, the heavy pieces would have been exchanged with no advantage for Black. He therefore waits for a better opportunity) 1 7 KR-B 1 N-Q2 1 8 R-B2 N-N3 19 N-Q 1 N-R5 20 N/ 2-B3 N x N 21 N x N N-B2 22 B-K3 N-N4 23 N x N Q x N 24 B-B2 B X P. Black has a clear advantage and later won the game. Seters-Lundwall, Biel 1 970, continued: 8 N-83 0-0 9 B-K2 B x P 1 0 0-0 QN-Q2 1 1 N-Q2 N-N3?! 1 2 P-QR4 ! ? B X B 1 3 Q x B Q-Q2 (Q-B2 is also playable) 14 P-R5 KR-N 1 1 5 R-R3 N-B 1 1 6 N-B4 N-K 1 (Q-N2-R3 is too slow) 1 7 R-K 1 N-B2 1 8 P-R4 ! ? B x N? 1 9 R X B N-N4 20 R-KN3 N-Q5 2 1 Q-Q 1 Q-N4 22 R-QB3 N-N6 23 P-R5 ! (White stands better) N X P 24 N-R3 Q-Q2 25 P x P BP X P (25 . . . RP X P 26 Q-Q2 and Q-R6) 26 R-R3 P-K4 2 7 P X P e.p. Q x P 28 P-K5 P-Q4 29 R-KB 3 (White has a winning position) N-N3 30 R-B6 Q-Q2 3 1 P-K6 Q-KN2 32 R-B7 Q-Q5 33 Q-B3 N-B3 34 N-B2 Q-KR5 35 P-KN 3 Q-Q 1 36 B-R6 N-K2 3 7 R-N7ch, Black resigns. In the modern sequence of moves, 1 P-Q4 N-K83 2 P-Q84 P-84 3 P-Q5 P-QN4 4 P X P P-QR3 5 P X P, a similar variation to that under discussion may be reached if Black does not recapture the QRP at once, but instead continues 5 . . . P-N3.

24

THE

BENKO GAMBIT

Gligoric-Udovcic, Yugoslavian Championship 1 957, continued : 6 N-QB3, P-Q3 7 P-K4, B-KN2 8 N-B3, 0-0 9 P-KR3, KN-Q2

(Black waits for White to move his KB before retaking the QRP) 10 B-K2 B X P 1 1 0-0 Q-N3 1 2 R-K 1 R-B 1 1 3 Q-B2 B X B 14 R X B N-R3 1 5 B-B4 N-N5 1 6 Q-Q2 P-B5 ? ! 1 7 B-K3 N-B4 1 8 B-Q4 N/ 5Q6 19 Q-K 3 , with a slight advantage for White. White m ay also play a waiting game with 9 B-KN5 (N-Q2 is possible too). Kluger-Lundin, Amsterdam Olympiad 1 954, con­ tinued: 9 . . . KN-Q2 10 Q-Q2 R-K 1 (Fighting for the tempo) 1 1 B-K2 B X P 1 2 0-0 Q-R4 1 3 KR-K 1 B X B 1 4 Q X B ! ? N-R3 (The only move. On 1 4 . . . B X N 1 5 P X B Q x P, the plan is 1 6 P-K5 with a plus for White) 1 5 QR-B 1 N-B2 16 P-QR3 QR-N 1 1 7 B-B4 N-N4 1 8 N X N Q X N 19 P-K5 ! with a small edge for White, who eventually won in about 80 moves. Most often, Black postpones P-QR3 until White moves his KB. But he must then take into account the following possibilities. One unusual way of playing for White occurred in Lemaire­ Hoorde, Brugge 1 954: 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P x P P-N3 5 N-QB3 P-Q3 6 P-K4 B-KN2 7 B-K2 0-0 ( P-QR3 is good here) 8 B-Q2!? P-QR3 9 P-QR4 P X P 10 P x P. By capturing with the RP White has doubled his QN pawns and has thus given up the usually preferred connected pawns on the Queenside. How­ ever, he has also greatly reduced Black's possible play on that wing, as Black will be able to attack the White pawns only with the greatest difficulty. Black must now look for counter-chances via the center break P-K3. The game continued: 1 0 . . . R x R 1 1 Q X R B-N2 1 2 Q-R7 ? ! (The Queen must shortly withdraw, but White is trying to hinder, if only temporarily, Black's P-K3 center break, which is playable on 12 N-B3 or 12 N-R3) 12 . . . Q-B 1 1 3 N-R3 QN-Q2 14 0-0 Q-B2 1 5 Q-R2 R-R 1 16 Q-N 1 N-N3 (P-K3?!) 1 7 P-B4 KN-Q2 1 8 Q-K 1 , with a small advantage for White. Unless Black plays P-QR3 at once (around the 4th move) White is not compelled to capture it later. Instead, he may try to use his QN5 as a base for his pieces. This method will be examined in this section. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4!? 4 P x P P-N3 5 N-QB3 B-KN2 6 P-K4 P-Q3 7 N-B3 0-0 8 B-K2 P-QR3 (Diag.) 9 0-0

25

THE

BENKO

GAMBIT

Laursen-Roos, correspondence, continued: 9 . . P x P 10 B X P B-QR3 (N-R3 i s possible) 1 1 N-Q2 Q-B2 (N-K 1 i s also playable) .

1 2 Q-K2 Q-N2 1 3 P-QR4 N-R4 1 4 N-B4 N-Q2 1 5 Q-B2 N-K4 1 6 N X N B X N 1 7 B-R6 KR-N 1 , with a slight edge for Black. In a game Prokorovitch-Beresin, 1 958, there was played: 11 Q-K2 B X B 1 2 N x B QN-Q2 1 3 P-QR4 ! N-K 1 1 4 Q-B2 N-B2 1 5 N X N Q X N 1 6 B-Q2! KR-N 1 1 7 B-B3 and White had the edge. Similar is a game Erikson-Akvist, Eksjo 1 970, which continued: 9 . . . QN-Q2 10 P-KR3?! P X P 1 1 B X P B-QR3 12 B X B (Q-K2!?) R X B 1 3 Q-K2 Q-R 1 14 B-B4 R-N 1 15 P-K5? ! N-R4 1 6 B-R2 P X P 1 7 N X P N X N 1 8 B X N R x N P ! ? 1 9 Q-K3!? P-B3 20 B-R2 P-B4 2 1 Q X BP?! (B-K5 is good) R-B7 22 QR-B 1 R/ 3 X P 23 R-N 1 ? (Not 23 Q-K 3 ? R X N 24 R X R B X R 25 Q X B Q X P with advantage to Black, but 23 Q-B6 ! Q X Q 24 P X Q R X R 25 R X R B X N 26 P-B7 with an unclear position) 23 . . . R X N 24 Q x P R-B 1 25 R-N7 R X P ! 26 R-Q 1 R-Q7 ! 27 R X R? (27 R/ 1 -N 1 R/ 7-QB7) 27 . . . Q-R8ch 28 K-B2 B-Q5ch 29 K-K2 Q-R3ch 30 R-Q3 R-B7ch, White resigns. Cafferty-Akvist, Wijk aan Zee 1 97 1 : 10 R-K1 P X P 1 1 B X P B-QR3 1 2 B X B R X B 1 3 P-K5 !? N x KP 1 4 N x N P x N 1 5 R X P R-Q3 1 6 B-N5, with a slight advantage for White. Stupica-Tringov, Ljubljana 1 969: 10 B-NS?! P-R3 1 1 B-R4 (A questionable maneuver, as it allows Black to exchange the valuable Bishop by P-N4 and N-R4. 1 1 B-B4 was better) 1 1 . . . P X P 1 2 B X P B-QR3 1 3 B X B R X B 1 4 Q-K2 Q-R 1 1 5 P-K5?! (N-Q2) P X P 1 6 N X P N X N 1 7 Q X N R-N 1 1 8 Q X P (QR-N 1 ) N x P 1 9 Q X P? ( 1 9 N x N leads to only a minimal advantage for Black) 1 9 . . . N-BS (Now Black has a distinct advantage) 20 P-B3 R X NP 2 1 26

THE

BENKO

GAMBIT

QR-Q 1 B X N 22 K-R 1 (22 R-Q8ch Q X R 23 B X Q B-Q5ch 24 Q X B N- K7ch, etc.) 22 . . . R-QB3 2 3 Q-K7 N-K3 , White resigned. Averkin-Forintos, Budapest 1 969: 10 B-Q2 P X P 1 1 B X P B-QR3 12 Q-B2 (Q-K2) B X B 13 N x B N-N3 14 P-QR4 Q-Q2 15 P-QN3 P-B5 ! (The game is equal) 1 6 N-B3 P X P 1 7 Q x P KR-N 1 1 8 QR­ N 1 Q-B 1 1 9 Q-R2 N x KP ! 20 N X N R X P 2 1 N X P ! P X N 22 Q-N3 R-B5 23 Q-Q3 R-B4 24 KR-B 1 N-Q2, drawn. Starting from the last diagram, let us look at some examples with the older 9 P X P (instead of 9 0-0), as in Bronstein-Lundin and Taimanov-Bronstein (see History of the Gambit). Janosevic-Parma, Belgrade 1 968: 9 . . . B X P 10 0-0 Q-N3 1 1 N-Q2 QN-Q2 1 2 B X B (N-B4) Q X B 1 3 P-B4 KR-N 1 1 4 N-B3 P-B5 15 Q-K2 N-B4 16 P-K5 N-K 1 17 P-B5 QP X P (Black has a slight edge) 1 8 N X P N-B3 1 9 N-B6 R-N2 20 P X P BP X P 2 1 B-K3? ! (After 2 1 N X Pch K-R 1 , Black would have a counterattack on White's undeveloped Queenside. Nevertheless, that would have given White an opportunity to fish in troubled waters, for instance with 22 B-N5) 21 . . . N-Q6 22 B-Q4 P-K4 ! (Black is better) 23 R X N (23 N X P R-K 1 ) B X R 24 N-K4 B-N2 25 N-B5 N x N 26 B X N Q-N4 27 Q-K3 B-R3 28 Q-B2 R-KB2 29 P-QR4 R X P 30 R X R R X Q 3 1 R-R8ch R-B 1 32 B X R B X B, White resigned. Similar is Kluger-Forintos, Budapest 1 968: 10 . QN-Q2 1 1 B X B R X B 1 2 Q-K2 Q-R 1 1 3 P-KR3 N-N3 1 4 B-Q2 N-R5 1 5 P-QN3 N X N 1 6 B X N R-N 1 1 7 KR-Q 1 R/ 1 -N3 1 8 Q-B2 R-R6 (The chances are equal) 19 B-N2!? R X RP 20 P-K5 N-Q2 21 P-K6 P x P 22 B X B R X Q 23 R X Qch K x B 24 R-R7 P X P ! ? 25 R X N K-B3 2 6 R-K 1 P-K4 27 P-KN4 P-R3 2 8 P-R4 P-K5 2 9 N-R2 K-K4 (Black stands better) 30 R-K7ch K-B3 3 1 R-Q7 R-B6 32 R-K3 R-Q6 3 3 N-B 1 K-K4 34 R-K7ch K-B5 ? (K-Q5 is stronger). Black now has a small edge but was unable to win. Before going ahead with the analysis of the more modern ap­ proaches to this opening, as seen in section B et seq. , it is worth examining several other rarely seen lines. Instead of 8 B-K2, the waiting move 8 P-KR3?! was played in a few games. Bajec-Forintos, Ljubljana 1 969, continued: 8 . P-QR3 9 P X P QN-Q2 10 B-K2 B x P 1 1 0-0 N-N 3 1 2 B-KB4 B X B 1 3 Q X B N-R 5 ! 14 QR-Q 1 ? (QR-B 1 ) N-R4 ! 15 B-Q2 N X P (Black stands better) 1 6 R-B l N-R5 1 7 KR-Q 1 Q-Q2 1 8 P-K5 N x N 19 B x N N-B5 20 Q-B2 Q-N4 21 R-Q2 R-R6 22 R-K 1 K R-R l 23 K-R2 Q-B5 24 R-K4 Q X B 25 R X N B X P 26 N X B .

.

.

27

.

THE

BENKO

GAMBIT

Q X N 27 P-N3 R X RP 28 Q-Q3 R X R 29 Q X R R-R8, White resigned. In Pribyl-Gaprindashvili, Tbilisi 1 97 1 , the continuation was: 11 B X B R X B 12 0-0 Q-R4 !? (Q-R 1 is normal) 13 B-Q2 R-N 1 1 4 R-N 1 N-K 1 1 5 Q-K2 R/ 3-N3 1 6 P-R3? R x P! ( I f now 1 7 N-Q 1 Q-N4). Black has the advantage and eventually won. Other setups for White are not very promising, e.g. O'Sull ivan­ O'Kelly, Dublin 1956: 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4!? 4 P x P P-N3 5 N-QB3 B-KN2 6 P-K4 P-Q3 7 B-Q3 0-0 8 KN-K2 P-QR3 9 0-0 P X P 10 B X P B-QR3 1 1 Q-Q3 Q-N3 1 2

P-KR3 ? (P-QR4) B X B 1 3 Q X B Q x Q 1 4 N x Q N x KP (Black is better) 15 P-B3 N-KB3 16 N/ 5-B3 N-R3, and Black later won. On occasion, the pawn sacrifice by P-QN4 is postponed, but if the timing is wrong white may be able either to prevent it or decline it advantageously. A game Barcza-Filip, Zagreb 1955, went: 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-Q3 4 N-QB3 P-KN3 5 P-K4 B-N2 6 N-B3 0-0 7 B-K2 P-QR3 ? ! 8 0-0 (P-QR4) QN-Q2 9 P-KR3 ( P-QR4 ! ) P-QN4 ! ? 1 0 PXP PXP 1 1 BXP B-QR3 12 BXB ( Q-K2) R X B 1 3 Q-K2 Q-N 3 (Q-R 1 ) 1 4 N-Q2 KR-R 1 1 5 P-QR4(?!) N-K 1 1 6 P-R5 Q-N 1 ( R X P 1 7 N-B4) 1 7 N-B4 N-K4 (Black has a slight edge) 1 8 R-N 1 N x N 19 Q x N R X P 20 P-QN4 P X P, and Black's small advantage was not enough to win.

28

THE

BEN KO

G A M BIT

B

White Gives Up Castling 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4!? 4 P x P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 P-K4!? B X B 8 K X B P-N3

This variation is one of the most radical solutions to the problem posed by the gambit: White offers to exchange Bishops and thereby gives up the right to castle. This loss is not tragic, however, as he can castle "by hand," even though this costs extra time. Black will try to make use of the extra time needed by White to build quick counterplay on the Queenside.

Bl

For the aggressive 9 P-KN4, see Avram-Benko and Visier-Benko in the Appendix. A dubious idea was seen in Kchouk-Fuller, Skopje Olympiad 1 972: 9 P-B3?! B-N2 10 KN-K2 0-0 1 1 B-K3 QN-Q2 12 K-B2 N-K4 1 3 P-KN4? Q-R4 (P-K3) 14 Q-B2 KR-N l 1 5 QR-QN l N-B5 16 B-B l N-K l 1 7 P-QR4 Q-N3 18 N-N5 N-K4 19 B-K3 Q-R4 20 P-N3 P-B5 (Black's pressure on the Queenside gives him the advantage) 21 N/ 2-B3 P X P 22 R X P R-B I 23 R-QB l R-B5 24 Q-K2 R/ 1 - B l 25 B-Q2 N-B2 26 P-B4 N X N 27 R X N Q-R2ch 28 B-K3 Q-B2 29 N-R2 R X R 30 N X R N-B5 3 1 N-Q3 N X B 32 Q x N Q-B7ch 3 3 Q-K2 Q X P 34 N-N2? (N-Kl ) B-Q5ch 3 5 K-N2 Q-B7, and White resigned. 29

THE

BENKO G A M B IT

The game Formanek-Browne, Atlantic Open 1 970, continued: 9 KN-K2 B-Nl 10 N-B4 0-0 1 1 P-KN4? ! I am convinced that this wild move, even with the Knight standing on KB4, can bring White no satisfactory result as long as Black does not underestimate White's chances on the Kingside. The game continued: 1 1 . . . N-R3 ( 1 1 . . . P-R3 1 2 P-KR4 KN-Q2 was to be considered with the intention of keeping the Kingside closed, e.g. 1 3 P-R5 P-N4, or 1 3 P-N5 P-R4) 1 2 P-KR4 Q-R4 1 3 P-R5 KR-N 1 (Black strives for active counterplay. It is rather late for 1 3 . . . P-N4 because of 14 P-R6) 14 RP X P RP X P 1 5 P-N5 N-Q2 (The Knight will be able to defend better from KB 1 , but White is unable to create serious danger on the KR file as his QB and QR do not participate in the attack) 1 6 Q-N4 N-K4 1 7 Q-R3 N-QN5 1 8 K-N2 Q-R3 1 9 Q-R7ch K-B 1 20 R-R3 N-B7 2 1 R-N 1 N-K8ch 22 K-N3 N/ 8-B6 23 B-K3 Q-B 1 (White has been too slow to bring all his heavy pieces to the KR file, after which he would have threatened to mate with a Queen sacrifice. Now he is himself threatened by mate) 24 K-N2 Q-N5ch 25 R-N3 Q-R5 26 Q X Q N X Qch 27 K-R3 N/ R-B6 28 N / 4-K2 P-B 3 ? This is a questionable move which weakens K3 and Q3 and at the same time closes his Bishop's diagonal. White's threat against the Knight by K-N2 was only good enough to draw. Correct was 28 . . . R-N 5 ! , e.g. 29 P-R3 R-N6 30 K-N2 QR-N 1 and now 3 1 R X N ? loses to 3 1 . . . N X R 32 K X N B X N 33 N X B R X N . Black eventually lost this game, but mainly because o f time-pres­ sure mistakes. 9 P-KN3 B-Nl 10 K-Nl

30

THE

BENKO

GAMBIT

Rabar-Milic, Zagreb 1 9SS, continued: 10 . . . QN-Q2 1 1 P-B4?! (This move can make the P-KS breakthrough more effective, but it is also very committing as it exposes White's King to many possible dangers) 1 1 . . . 0-0 1 2 N-B3 Q-N3 1 3 R-K 1 P-BS?! (Correct is 13 . . . Q-N2 ! followed by N-N3) 14 R-K2 ! N-B4 1 S B-K3 . White has a slight advantage. Black must always consider the move P-QBS very seriously and carefully, as it gives White his Q4 square for a Bishop or for a Knight, which can then go to QB6. Also, the QBP itself may become weak. 10 . .

.

0-0

Malich-Ciocaltea, Vrnjacka Banja 1 972, continued: 1 1 P-B4?! N-R3 ! ? ( 1 1 . . . P-K3 ? ! 1 2 P x P P X P 1 3 N-B3 N-B3 1 4 R-K 1 N-K 1 [ 1 4 . . . P-Q4 is to be considered, on which 1 S P-KS, closing the game, or 1S N-KNS, may follow] 1 S B-K3 R-N 1 1 6 Q-Q2 Q-R4 17 QR-B 1 with a small edge for White, who won in Gerusel­ Schaufelberger, Luxembourg 1 97 1 ) 12 N-B3 Q-N 3 1 3 R-K 1 N-QNS!? 1 4 R-K2 (P-KS) Q-R3 1S P-QR4 KR-N 1 16 R-R3 Q-BS ! ? 1 7 N-Q2 Q-Q6 1 8 N-B 1 ? (N-B3 is equal) Q x Q 19 N x Q P-BS 20 N-B2 N-Q2 21 B-K3 N-Q6? (P-B4 ! gives Black a plus) 22 N x N P X N 23 R X P R X RP, drawn.

B2

After 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P x P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 P-K4 B X B 8 K X B P-N3, the most frequently played and analyzed move is 9 N-B3. The Knight advance, which seems to be the most natural move, is some­ times preceded by P-KN3 and K-N2, but this usually means only a transposition of moves. (However, some independen t variations may be reached by transposition-see B2a and B2b.) At this point we will discuss variations arising from the move order 9 N-B3 B-N2 10 P-KN3 0-0 11 K-N2, and now 1 1 . . . QN-Q2 1 2 R-K 1 Q-N 3 is the usual sequence. However, the game Kane-Vogt, Skopje Olympiad 1 972, deviated by 11 . . . Q-N3 12 R-K1 N-R3 1 3 R-K2 N-Q2 14 B-NS P-R3 1 S B-K3 ( l S B X KP KR-K 1 1 6 B-R4 P-N4 i s unclear) 1 S . . . KR-N 1 31

THE BENKO GAMBIT 1 6 Q-Q2 K-R2 1 7 R-QB 1 N-B2 1 8 P-KR4 N-N4 1 9 N x N Q X N 20 P-N3 R-N2 (White i s slightly better) 2 1 B-B4 R/ 2-R2 22 R-B2 Q-N2 23 Q-Q3 N-B3 24 QR-Q2 N-R4 25 B-K3 N-B3 26 B-B4 N-R4 27 B-K3, drawn.

Position after 12 R-Kl

After 11 . . . QN-Q2 12 R-Kl, instead of the usual 1 2 . . . Q-N 3, the game Kuzmin-Georgadze, USSR Semifinal 1 972, continued 1 2 . . . Q-R4 13 B-Q2 (Obvious, but not the best. 13 P-KR3 is neces­ sary here, as in similar positions) 1 3 . . . KR-N 1 14 Q-B2 N-N 5 ! (Black already stands a little better) 1 5 P-QR4? ! Q-N5 1 6 P-R3 N/ 5-K4 17 N X N B X N ! ( 1 7 . . . N x N 1 8 N-N5 Q-B5 and White neutralizes the long diagonal with 19 B-B3) 1 8 N-Q 1 Q-N6 1 9 Q X Q R X Q 20 R-K3 (20 B-B3 B X B 21 P X B N-N3 , o r 20 R-R2 N-N3, etc.) 20 . . . B X P (Black is much better) 21 R-R2 (2 1 R X R B X R 22 R-N7 N-B3 2 3 R X P R X P 2 4 N-B3 R-B5 with advantage for Black) 21 . . . R X R 22 N X R B-Q5 23 N-B4 P-B4 24 P X P P X P 25 B-N5 (P-R5) K-B2 26 P-B4? ! R-QN 1 ! 27 N-R5 (27 P-R5 R-N5 28 N-Q2 P-B5) 27 . . . P-R3 ! 28 B x RP (28 B-R4 B-B3) 28 . . . N-N3 29 N-B6 R-QR1 30 P-N4 N X QP 3 1 N x B P x N 32 P-R5 K-N3 33 B-N5 P-K4! (With even material , Black's two passed pawns decide) 34 P-R6 N-K6ch 35 K-N 1 P-K5 36 B-K7 P-Q6 37 B X P N-B7 38 P X Pch K-B2 39 R-N2 P-Q7 40 R-N7ch K-N 1 , White resigned. 32

THE BENKO GAMBIT

Position after 1 2 . . . Q-N3

For 1 3 QR-N 1 , see Donner-Benko in the Appendix. In Sahovic-Benko, Vrnjacka Banja 1 97 1 , 13 Q-B2 was played with the following continuation: 1 3 . . . Q-N2 !? 1 4 B-B4 N-N5 1 5 R-K2 (A typical Rook move, protecting the second rank) KR-N 1 1 6 R-QB 1 Q-R3 1 7 P-KR 3, and in this equal position White offered a draw. Black's pieces are harmoniously developed and after the possible trades after 1 7 . . . N/5-K4, Black, thanks to his strong Bishop, may even have the more comfortable game, despite his pawn minus, e.g. 1 8 N X N N X N 19 B X N B X B. 13 R-K2 KR-N1

Keres-Pohla, Parnu 1 97 1 , continued: 14 B-B4 N-N5 ! 1 5 R-B 1 Q-R4 1 6 B-N5 (P-QR3 !?) B X N ( A remarkably embarrassing move for White) 1 7 P x B P-B3 (Black has made White's Bishop inactive 33

THE BENKO

GAMBIT

and he controls the QN file, while White's Queenside has been weakened) 1 8 B-B4 Q-R5 19 Q-Q3 Q-R3 20 P-B4? ! (20 Q-Q2 Q-B5) R-N5 (Black's pressure on the Queenside holds the game in bal­ ance) 21 P-KR3 KN-K4 22 N x N N x N 23 B X N BP X P 24 P-KR4 R-R5 25 R/ 1 -B2 K-N2 26 P-R5 (KR-Q2) Q-B 1 27 R-N2 P x P 2 8 R-K3 R X RP 29 R X R R X R 3 0 Q-N 1 R-R5 3 1 R-KB3 R X P 32 R-B5 Q-K 1 3 3 R-N5ch K-R 1 34 Q-Q3 R-N5 3 5 Q-KB3 R-N 1 3 6 R X RP Q-N3 37 R-B5 K-N2 3 8 P-N4 R-N 1 39 K-R3 P-R3, drawn. In Gligoric-Browne, Rovinj-Zagreb 1 970, strange to tell, White willingly gave up his Bishop: 14 B-NS? P-R3 15 B X N B X B 1 6 R-B 1 P-B5 ! (White has not found the time for P-QN3 and Black's Bishop is therefore even stronger than in other lines) 1 7 R/ 1 -B2 R-N2 1 8 Q-B 1 K-R2 (K-N2 is even better). Black has a slight edge but the game was eventually drawn. As we have seen in earlier examples, Black was able to success­ fully activate his King Knight by means of the maneuver N-N5-K4. In his game with Browne above, Gligoric attempted to avoid that possibility but without a satisfactory result. In a later game from the same tournament, Gligoric-Udovcic, he improved the line by 14 P-KR3 R-N2? (Better is Q-R3 as in the next example) 15 R-B2 N-K 1 16 B-K3 Q-R4 (Black seems to be afraid of the complications after 1 6 . . . N-B2 1 7 P-QN4) 1 7 Q-K2 QR-N 1 1 8 R-K 1 N-B2 1 9 B-B 1 N-R3 2 0 P-R3 N-B2 2 1 N-Q2. White has the advantage as Black has lost too much time with his complicated maneuvering. White later won. F. Portisch-Gaprindashvili, Tbilisi 1 97 1 , reached the position after 14 P-KR3 by transposition. The continuation was: 14 . . . Q-R3 1 5 P-K5?! (This action seems a little premature and is there­ fore easy to meet. Both players, by the way, have lost a tempo­ R-K2 and Q-R3) 1 5 . . . N X KP 1 6 N x N P X N 1 7 R X P Q-N2 1 8 Q-B3 R-Q 1 1 9 R-K2 N x P (The position has been equalized) 20 N x N Q x N?! (20 . . . R X N 21 R X P Q x R 22 Q X R R-Q 1 23 Q-B3 Q-K8 24 B-N5 Q X R 25 B X R Q x RP is a level position) 21 Q X Q RXQ 22 RXP (White now has an edge) 22 . . . R-Q8 23 R-K2 P-R4?! (A better King escape is provided by 23 . . . P-B4) 24 P-KR4 B-R3 25 R-B2 R-R4? (A drawish Rook ending is reached by 25 . . . B X B) 26 P-B4? (White wins at once by 26 P-QN4! since 26 . . . P X P is met by 27 R-B8ch and if 27 . . . K-R2 28 B-N2, or 27 . . . B-B 1 28 R X Bch, etc.) After the text, the game was eventually drawn. 34

THE

BENKO

GAMBIT

B2a

By p laying the White moves in a different order, it is possible to arr ive at positions different from those discussed above . After 1

P-Q4 N-KB3 2 P-Q84 P-84 3 P-QS P-QN4 4 P X P P-QR3 5 P X P 8 X P 6 N-Q83 P-Q3 7 P-K4 B x 8 8 K x 8 P-N3 9 P-KN3 8-N2 10 K-N2 QN-Q2 1 1 N-B3 0-0, White may play P-KR3 at this

point or later.

Position after 1 1 . . . 0-0

In the game Cobo-Vasiukov, Polanica Zdroj 1 972, White played 12 P-KR3 at once and the continuation was: 12 . . . Q-N3 1 3

R-K 1 KR-N 1 1 4 R-QN 1 Q-R3 1 5 Q-B2 R-N2 1 6 B-N5 N-K l !? (Surprisingly, 1 7 B X P is met by 1 7 . . . P-83 !?, and the Bishop can't get out) 17 B-B4? ! P-B5 1 8 B-K3 ( 1 8 P-K5 N-B4) 1 8 . . . N-K4 1 9 N X N B X N 20 KR-Q 1 N-B3 2 1 B-Q4 (Too many Bishop moves) 2 1 . . . N-Q2 22 P-R3 ?! (P-N3) B X B 23 R X B N-K4 (The position is equal) 24 N-Q 1 ? Q-N3 (But now Black has the edge) 25 Q-B3 (25 R-Q2 R X P) 25 . . . Q-B4 ! 26 R-Q2 R-N6 27 Q-Q4 (Losing the Exchange, but 27 Q-B2 N-B6 28 R-K2 N-Q5, or 28 Q-B 1 N-Q6 is just as bad) 27 . . . Q X Q 28 R X Q N-B6 29 R X P N-Q7 30 R/ 1 -B l N x R 3 1 R x N R-Q6 3 2 N-K3? (32 N-B3 R-Q7 3 3 R-N4, with Black having only a small edge) 32 . . . R-N l , and White resigned. The game Popov-Tringov, Varna 1 972, continued 12 . . . Q-R4 13 Q-82 N-N3 14 N-Q2?! (R-K 1 ) KN-Q2 1 5 N-N3 Q-R3 1 6 R-Q l N-R5 ! (Gaining the advantage, as the removal of White's Queen Knight increases Black's pressure on the Queenside) 17 N-Q2?! 35

THE

BENKO

GAMBIT

N x N I 8 P X N Q-K7 I 9 P-R3 N-K4 (Black wins with I 9 . . . P-B4 ! and if 20 P X P R X BP 2 1 R-B I QR-KB I . or if 20 R-R2 N-B3 ! 2 I P x P N x P, etc.) 20 R-R2 P-B5 2 I N-B I Q-B6ch 2 2 K-N I N-Q6 23 Q-K2 Q x Q 24 R X Q QR-N I , and Black had only a slight ad­ vantage. The game was later drawn. In a game Malich-Ciocaltea, Skopje Olympiad I 972, the con­ tinuation was: 13 R-K1 KR-N1 14 R-K2 N-K1?! 15 B-N5 P-R3 ! I 6 B-B4 ( 1 6 B X KP P-B3) I 6 . . . P-N4 ! ? I 7 B-B I (B-K3) N-B2 1 8 P-KR4 ! P-N5 I 9 N-R2 N-K4 20 R-B2 P-R4 2 I P-B3 P X Pch 22 N x P N-N5 (The position is unclear. White should try 23 Q-K2 and N-R2) 23 N-N5 ? ! R X P ! 24 R x R (24 B x R N-K6ch) 24 . . . B X N 25 Q-B3 P-B3 26 Q-B5 P X N 27 Q x Pch B-N2 28 R/ 1 -N I R-KB I 29 Q X RP Q-K8 3 0 Q x N Q-B8ch 3 I K-R2 R-B7ch 32 RXR Q x Rch 3 3 K-R I Q-B8ch, and the game was drawn by perpetual check. A less effective method for Black was seen in Knaak-Damjanovic, Sombor 1 972, which continued: 15 . . . B x N?! (Too committing. Better is I5 . . . P-R3 !, as in the previous game, or 15 . . . K-B I ! ?) I 6 P X N Q X BP? ( 1 6 . . . P-B3 ! is called for to keep the position closed; if I7 B-Q2 N-N3 or 17 . . . Q-R5 is not bad) I7 B X P (White gets a clear edge on I7 R-B I Q-N 5 I 8 B x P P-B3 I9 P-K5) 1 7 . . . N-K4 (Wh ite is better after 1 7 . . . P-B3 1 8 P-K5 QP X P 1 9 P-Q6, or 18 . . . BP X P 19 &K3 Q-N 7 20 N-N5) 18 R-K3 Q-N7 1 9 B-R4 N x N 2 0 K x N ! ? N-B3 (20 . . . R X P 2 I R X R Q X R 22 P-K5 and if 22 . . . P x P? 23 P-Q6 wins for White, but 20 . . . P-B3 2 1 P-N4 R X P 2 2 R x R Q X R gives Black a slight edge) 2 1 Q-QB 1 Q-K4 22 Q-B 3 N-Q2, and although Black stands a little better, White eventually won. Instead of the immediate I2 P-KR3, the game Trifunovic-Kozo­ mara, Sarajevo 1963, continued : 12 R-K1 N-N5!? 13 P-KR3 KN­ K4 1 4 N X N N x N 1 5 P-B4 N-Q2 1 6 Q-B2 Q-N 3 and Black achieved equality. A game Popov-Vasiukov, Varna 1 97 I , continued 13 R-K2 (in­ stead of 13 P-KR 3) 13 . . . Q-R4!? (Q-N 3 ) 14 B-N5 (Trying to punish the "premature" Knight sortie, but this has little effect) 14 . . . KR-K1 15 Q-B2 P-R3 I6 B-Q2 Q-R3 (The game is equal) 1 7 R-Q I KR-N I I 8 B-B I R-N2 I 9 P-KR3 KN-K4 20 N x N N X N 2 1 P-B4 N-Q2. A t this point White gave back the pawn with 22 P-K5? ! P x P 23 P-B5 , but the advantage is now Black's. White has already played his most constructive moves and a good plan for him 36

THE

BEN KO

GAMBIT

is hard to find. Instead of 15 Q-B2, the game Bagirov-Zilberman, USSR 1 972, co ntinued 15 R-B 1 R-R2 16 R/ 1 -B2!? (A pessimistic defense? !) N-N 3 17 P-KR3 N-K4 18 N x N B X N 1 9 Q-Q3 Q-N5 20 B-Q2 R-N2?! (Sharper and more logical is 20 . . . R-N 1 2 1 N-N 5 ? P-B 5 ! ) 2 1 P-N 3 P-B5 22 P X P Q X P 23 Q X Q N X Q 2 4 B-B 1 R-R 1 (White stands only slightly better; the extra pawn is not worth very much in view of the active Black pieces) 25 N-Q 1 N-R6 26 B X N R X B 27 N-K3 R-R5 2 8 R-B4 R/ 2-R2 2 9 R X R R X R 3 0 K-B3 K-N2 (Or 30 . . . R-R6 3 1 R-Q2 P-B4 with equality) 3 1 R-B2 P-R4 32 N-B4 B-Q5 33 K-K2 R-N5 34 P-KR4 P-B4. The game is even and was eventually drawn. From the last diagram, the moves 12 R-K1 Q-B2 were played in Holm-Toran, Skopje Olympiad 1 972, which continued : 1 3 R-K2 KR-N 1 1 4 R-B2 Q-N 2 15 P-KR3 Q-R3 16 B-N5 P-RJ 17 B-K3 N-K 1 18 Q-Q2 K-R2 19 R-Q 1 R-N5 ! (The chances are equal) 20 P-R3 ? ! R-N6 ! 21 Q-B 1 Rl l -N 1 22 N-Q2 R/ 6-N2 23 P-B4 ! ? (P-N 3) Q-Q6 24 N-B 1 ? Q X Rch 25 QXQ RXP 26 R-Q2 R X Q. Black has the advantage and finally won. Returning to the last diagram, instead of 12 R-K 1 , Mecking­ Szabo, Buenos Aires 1 970, continued: 12 N-Q2?! (To prevent Black's N-N5) N-N3 ! ? 1 3 Q-K2 R-K 1 ? (KN-Q2) 1 4 N-B4 N X N 1 5 Q x N Q-N 3 1 6 R-Q l !? (R-K 1 ) Q-N5 1 7 Q-K2 N-Q2 1 8 P-QR3 Q-N 6 1 9 P-QR4! The position is equal but Black later won. Gerusel-Pedersen, West Germany 1 97 1 , continued: 12 Q-K2 Q-R4 1 3 B-B4 N-N5 ! 1 4 N-Q2 KN-K4 1 5 B-K3 ?! (KR-QB 1 ) KR­ N l ! 1 6 P-B4? White insists on trying to win a piece, perhaps ex­ pecting only 1 6 . . . R x P, which could be advantageously met by 1 7 N-Q 1 N-N5 1 8 N X R N X Bch 1 9 Q X N B X N 20 N-B4. The better move is 1 6 QR-QN l . The game proceeded: 1 6 . . . N-N 5 ! 17 Q X N (Also 1 7 N-N3 N x Bch 18 QxN Q-N5 is advantageous for Black) 17 . . . R X P 1 8 N-K2 R X N 19 B X R Q X B 20 QR-Q 1 Q-B7 2 1 Q-B3 R x P 22 KR-K 1 P-B5 (White has a hopeless position despite being the Exchange ahead. Black's Knight is threatening to advance via QB4) 23 R-QB 1 Q-R5 24 K-R 1 B-N7 25 N-B3 Q-N6 26 Q-Q 1 R-R4 27 R-B2 B X N 28 QR-K2 R-R8 29 Q X R B X Q, White resigned.

37

THE

BENKO

GAMBIT

B2b

Another idea, not based on the N-N5-K4 maneuver, was tried after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P X P P-QR3 5 P x P B X P 6 N-QB3 P-Q3 7 P-K4 B X B 8 K X B P-N3 9 P-KN3 B-N2 10 K-N2, and now 10 . . . 0-0 (instead of QN-Q2 as in the previous section) 11 N-B3.

The game Gligoric-Diez del Corral, Siegen Olympiad 1 970, con­ tinued: 11 . . . Q-N3 12 R-K1 KN-Q2 (Black is obviously fearful of P-K5 and avoids the usual development QN-Q2, which would transpose into the usual lines) 1 3 R-K2 N-R3 14 B-B4 (Interesting is 1 4 B-N5 where White can get four pawns for a piece after 1 4 . . . P-R3?! 1 5 B X KP R-K l 1 6 B-R4 P-N4 1 7 N X P or 1 7 B X P, but his prospects would be unclear) 1 4 . . . KR-N 1 1 5 R-B 1 N-B2 1 6 R/ 1 -B2 B X N ! ? 17 P X B P-B3 1 8 P-KR4 N-N4 19 B-B 1 N-R6 20 R-N2 Q-R3 2 1 R X Rch R X R 22 B X N (If White is compelled to give up this Bishop, he may not expect too much in the way of results) 22 . . . Q X B 23 R-B2 Q-R5 24 Q-K2 R-N8 (The activity of the Black pieces is full compensation for the minus pawn. The chances are equal) 25 P-B4 P-R4 26 Q-Q2 K-B2 27 Q-B3 R-N5 28 N-Q2 N-K4 29 P-QR3 R-N 1 30 R-B 1 , drawn. White's pieces are tied to the defense of the weak squares on the Queenside, especially QR3 and QN4. The only possibility for activity lies in P-B4, but this would weaken the Kingside and give Black's Knight a fine place on KN5 . However, Black is unable to strengthen his position either, since all invasion points are protected. As we have seen, none of the well-known grandmasters were able to find success against their lesser-known opponents, who fought without undue fear.

B3

It has been shown by previous examples that White generally finds it useful to play P-KR3 for several reasons. Trying to kill two birds with one stone, it may be reasonable to use that fact to provide shelter for the King on KR2. If one merely counts tempos, it will be seen that the maneuver P-KR3 followed by K-N 1 (the King has already taken the Bishop on KB 1 ) and K-R2 uses one more than P-KN3 followed by K-N2. But although the King does 38

THE

BENKO

GAMBIT

not necessarily stand better on KR2 than on KN2, the move P-KR3 does not weaken the KR 1 -QR8 diagonal, as does P-KN3 , and is less committal. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P X P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 P-K4!? B X B 8 K X B P-N3 9 N-B3 B-N2 10 P-KR3!? 0-0 11 K-Nl!?

The parent game for this line is Gligoric-Browne, Skopje 1 970, which continued: 11 . . . QN-Q2 12 K-R2 Q-R4 (Q-N 3-see below) 13 R-K1 KR-N1 14 Q-B2 N-N 3 (The maneuver N-K 1 -B2 seems better) 1 5 N-Q l ! Q-R3 (N-R5) 1 6 B-Q2 Q-B5?! 1 7 N-B3 ( 1 7 Q x Q N X Q 1 8 B-B3 i s simpler and gives White a slight edge) 1 7 . . . KN-Q2 1 8 P-QN 3 Q-R3 19 P-QR4 N-QB 1 (P-B5) 20 QR-N 1 N-R2 21 N-Q 1 R-N2 22 N-K3 B-R3 23 N-N4 B X B 24 N X B (Q X B) P-R4 25 N-K3 N-QB 1 26 N/ 3-B4 N/ 1 -N3 27 P-R5 ? ! N X N 28 N x N N-K4 ! ? 29 N X N P X N 30 Q X BP Q x P. White has only a minimal advantage and the game was later drawn. In Hort-Jimenez, Palma Interzonal 1 970, the continuation was: 14 R-K2 N-K 1 15 R-B2 N-B2 1 6 B-Q2 N-N4 17 P-QR4 N X N 1 8 B X N B X B 1 9 P X B N-K4! 20 N X N P X N 2 1 Q-Q3 R-N6 (Black dominates the QN file and his heavy pieces are well placed; therefore White's extra pawn gives him only an insignificant advantage) 22 Q-N3 Q-B2 23 P-R4 P-R4 24 P-R5 R-R3 25 R/ 2-R2 P-B5 26 Q-K3 R-N4 27 Q-K2 K-N2 28 R-R4 R-B4 29 P-N3 K-R2 30 K-N 1 K-N2 3 1 K-N2 K-R2 32 R/ 1-R2 K-N2 (Black merely waits) 3 3 Q-B 1 Q-B 1 34 R-N4 Q-N5 3 5 R X P R/ 4 X RP 36 R X R R X R 37 Q-Q3 R-R8 3 8 Q-B2 Q-Q2 ( 3 8 . . . P-N4 39 P-B3) 39 R-N4 Q-R2 40 Q-N2(?) R-R7 41 R-N7 Q-R5 42 Q-N 1 (42 Q-N4? Q-B7) 42 . . . Q-B5, drawn (43 R-N4 Q X BP). 39

THE

BENKO GAMBIT

Instead of 1 2 . . . Q-R4, the move 12 . . . Q-N3 was chosen in Ogaard-Gheorghiu , Zonal Tournament in Finland 1 972: 13 R-K 1 N-K 1 (KR-N l ) 1 4 R-K2 N-K4?! 1 5 N X N B X Nch 1 6 K-R 1 N-B2 17 B-N5 KR-K1 18 R-B 1 R-R2 19 Q-Q3 Q-R3 20 Q-Q2, draw. White's advantage is tiny. In a few games Black chose another way of developing his Queen Knight. Schaufelberger-Parma, Luxembourg 1 97 1 , continued 11 . . . N-R3!? (instead of 1 1 . . . QN-Q2) 12 K-R2 Q-N3 13 N-Q2 ( 1 3 R-K 1 P-B 5 ! ?) 1 3 . . . N-Q2 1 4 Q-K2 N-N5 ! 1 5 N-B4 Q-R3 1 6 R-K 1 N-K4 1 7 N x N B X Nch 1 8 P-N3 (P-B4) KR-N 1 (Black has a slight advantage) 1 9 Q x Q R X Q 20 R-K2 N-Q6 2 1 R-Q2 N-K8 22 K-R 1 R-N 5 23 R-K2 N-Q6 24 K-N 2 R/ 3-N3 25 N-Q 1 R-R5 (White's Queenside is under strong pressure and he has no hope for any counterplay) 26 N-B3 R-B5 27 P-QR4 K-N 2 (Black can regain the pawn at any time, of course, but he has time for anything) 28 P-R5 R-R3 29 P-B4 B-Q5 30 P-K5 N X B 3 1 R x N R X P (Black has a clear advantage) 32 R/ 1 -B2 R-R3 33 K-B3 R-N 3 34 P-KN4 R-N6 35 P X P P x P. White overstepped the time limit, but he had a bad position anyway. A similar development (but with White's King Knight developed to K2) was seen in Enklaar-Ree, 1 972 Holland Final: 9 KN-K2 (in­ stead of N-B3) B-N2 1 0 P-KR3 0-0 11 K-N 1 Q-N3 (Q-R4!?) 12 R-N 1 N-R3 13 B-K3 KR-N 1 1 4 K-R2 (A typical King maneuver) N-Q2 15 Q-Q2 Q-R4 (This is clearly a lost tempo, but Black plans the maneuver N-B2-N4 which could not be played at once because of 1 6 P-QN4) 16 B-N5 Q-B2? ! (The Queen cannot find her proper place; Q-Q 1 is better. After the text, 17 B X P is of course a blunder because the B ishop is trapped after 1 7 . . . P-B3) 1 7 KR-QB1 (P-QN3 is better) P-B5 ! ? 1 8 B-K3 N-N5 1 9 N-Q4 N X RP ? ! (More prudent i s 1 9 . . . N-K4 20 P-R3 N/ 5-Q6 2 1 R-B2 N-B4 with equal chances) 20 N X N R X N 21 R-B2 (White has a small plus) N-B4? 22 R x P N X P 23 Q-B l N-B4 24 P-QN4 Q-N2 25 N-B6 and White won. Another alternative on the 1 3 th move (after 9 N-B3 B-N2 1 0 P-KR3 0-0 1 1 K-N 1 N-R3 1 2 K-R2 Q-N 3) was seen i n Forintos­ Vukic, Vrnjacka Banja 1 973, which continued: 13 R-K1 N-Q2 1 4 R-K2 N-N5 ! 1 5 B-N5 KR-K1 1 6 P-R3 ? ! Q-R3 1 7 Q-Q2 N-Q6 (Black has the edge) 1 8 B-R6 B-R l 19 N-K l P-B5 ! 20 N X N P X N 2 1 R/ 2-K 1 (2 1 R-K3 N-K4 2 2 P-QN3 KR-N l 2 3 P-B4 R X P ! wins) 21 . . . N-B4 22 QR-Q 1 Q-B5 23 B-K3 B-K4ch? (KR-N l ) 24 K-N 1 40

THE

BENKO

GAMBIT

N X P? ! (KR-N 1 ! gives Black the advantage) 25 N X N Q x N 26 Q X P Q X Q 27 R x Q B X P. Black still has a small edge, but White eventually won. The system of development with Black's Q-R4 and his Queen Kn ight on R3 was again tried in Smejkal-Jansa, Luhacovice 1 97 1 , which also saw White develop his Rook o n K l . From the preceding diagram, the continuation was 11 . . . N-R3 12 K-R2 N-Q2 13 R-K1 Q-R4 14 B-NS KR-N l 15 R-K2 Q-Q 1 ( 1 5 . . . R X NP? 1 6 R X R B X N 1 7 B-Q2) 1 6 R-B 1 R-N2 1 7 K-N 1 N-K4 1 8 N x N B X N 1 9 P-QN3 Q-R4 (White stands better) 2 0 N-R4 N-N5 (Better is Q-Q 1 followed by N-B2-N4) 2 1 R-Q2 Q-N4 22 Q-B 1 Q X Qch 23 K X Q P-B4 24 P x P P X P 25 R-B4 B-Q5 26 B-K3 N X QP (B -K4) 27 B X B P X B 28 R/ 2 x P. White has the advantage and l ater won. The attempt by Black to blow up the White center at once was tried in O'Kelly-Honfi, Budapest 1 972. Instead of 1 1 . . . QN-Q2 or 1 1 . . . N-R3, Honfi played 11 . . . P-K3?!, and the game con­ tinued: 12 P X P P X P 1 3 P-K5 ! (Destroying the Black center and reaching a better endgame) 1 3 . . . P X P 14 Q x Q R X Q 1 5 B-N5 N-B3 16 K-R2 KR-N 1 (P-K5) 17 B X N B X B 1 8 N-K4 B-K2 1 9 P-QN3 N-Q5? (P-B 5 ? ! was a better chance) 2 0 N x KP R-KB 1 (Threatening R-B5) 2 1 N-Q3 N-B7 22 QR-B 1 R X RP 23 KR-Q 1 ! (23 N-B3 B-Q3ch 24 P-N3 N-N5 25 N X R N x N/ 6) 23 . . . N-N5 24 N/ 3 X P B X N 25 R X B R-K7 26 R-K5 R/ 1 X P 27 R-Q8ch K-B2 28 N X R R X R 29 R-Q7ch K-K 1 (29 . . . K-N 1 30 N-N4) 30 R X P R-K6 3 1 R-KN7 R X P 32 R X P and White won.

41

THE

BENKO GAMBIT c

White Retains Castling Compared to Variation B, White tries a more cautious approach. As in the previous variation, he intends to exchange his King Bishop to neutralize the KB 1 -QR3 diagonal and establish P-K4, but he wishes to avoid at the same time losing the right to castle. For this purpose, he inserts the maneuver N-KB3-Q2, so that after he plays P-K4 he will be able to recapture the Bishop on KB l with his Knight. After some initial successes, this line became very popular. Later, however, ways were found for Black to improve ; at present it is considered that Black has full counterplay in this line. 1 P-Q4 N-KB3 2 P-Q84 P-84 3 P-QS P-QN4 4 P x P P-QR3 5 P X P 8 X P 6 N-Q83 P-QJ 7 N-83 P-NJ 8 N-Q2 8-KN2 9 P-K4

The seminal game for this variation is Taimanov-Benko, Wijk aan Zee 1970, which will be found in the Appendix.

C1 9

.

.

.

B X B 10 N X 8 0-0 1 1 N-KJ

This move illustrates the principal disadvantage for Black when he initiates the exchange of Bishops on his 9th move: White's Knight on K3 is very actively posted. After the Taimanov-Benko game (won by White), Najdorf­ Garcia, Siegen Olympiad 1970, continued: 11 . . . QN-Q2 12 0-0 42

THE

BENKO

GAMBIT

Q-Bl 1 3 B-Q2 KR-N l 1 4 Q-B2 Q-N2 1 5 P-QN 3 N-K l 1 6 QR-N l

N- B2 1 7 P-QR4 (White has crossed Black's plan of N-N4 at the right moment and now stands slightly better) Q-R3 1 8 QN-Q l Q-N2 1 9 N-B3 Q-R3 (Having prevented 1 9 B-B3 by the threat of R X P, Black is satisfied to repeat the position) 20 N-B4 N-N3 2 1 N X N R X N 2 2 KR-B 1 B-Q5 23 B-R6 QR-N 1 2 4 N-R2 (Th is is hardly an ideal spot for a Knight but it prevents P-B5 which would now be answered by P-QN4) 24 . . . N-K 1 25 R-Q 1 N-B3 26 P-R3 Q-N2 27 R-Q3 B-K4 (Threatening 28 . . . N x KP or 28 . . . P-B5 and if 29 Q X P R-QB I . The position is about equal) 28 R-KB3 N-Q2 29 B-Q2 B-Q5 30 N-B l , N-K4 3 1 R-N3 R-R3 32 P-N4? ! (Realizing the difficulties, White loses patience. Necessary is 32 P-R5, although after 32 . . . P-R4 or Q-N4 Black still has good counter-chances for his pawn) 32 . . . Q-R2 33 N-N3 R X RP 34 N X B P X N, drawn. In a game against Doroskevich in the USSR 1 97 1 , ex-World Champion Spassky played 12 . . . Q-N3 instead of 12 . . . Q-B2. The game continued: 13 R-N l Q-N5 ! ? (An original Queen maneuver so early in the game) 14 P-QR3 Q-Q5 1 5 N-B2 Q-B5 16 N-K3 (White has slightly better chances) 16 . . . Q-Q5 17 N-B2, drawn. White might have tried 1 7 Q-B2 in order to force Black to reveal his plan ( 1 7 . . . P-B5 ?!). Karasev-Platonov, USSR Championship 1 97 1 , continued: 13 Q·Bl KR-N1 14 R-N 1 N-K1 15 P-QN3!? N-B2 (The move 1 5 P-QN3 could have been better tested by 1 5 . . . B X N ! 1 6 Q X B R X P with equality. G iving u p the Bishop is not a t all dangerous for Black as has already been shown) 1 6 N-B4 Q-R3 1 7 P-QR4 N-K4 (White has an edge, thanks in part to the sad position of Black's Knight on QB2) 1 8 N X N B x N 1 9 B-R6 R-N2 20 KR-B 1 QR-N 1 2 1 N-Q l ! (Preventing P-B5 and aiming to occupy QB4 with the Knight. White has a clear advantage) 21 . . . B-N2 (P-K3 ! ?) 22 B X B K X B 23 N-K3 Q-R4? 24 N-B4 Q-R3 25 Q-B3ch K-N 1 26 N-R5 (White is winning now) R-N 3 27 N-B6 R-K 1 28 P-QN4 P-K3 29 NP X P R X R 30 R X R Q X P 3 1 BP X P, Black resigned. The game Bukic-Vukic from the Yugoslavian Championship 1972 followed the same course until White's 1 5 th move. Instead of 1 5 P-QN3, Bukic tried 15 B·Ql, which appears more reliable than giving back the pawn. The game continued: 1 5 . . . Q-R3 ? ! 1 6 P-QN3 N-B2 (No pawn is won by 1 6 . . . B X N because of 1 7 Q x B Q x P? 1 8 R-R l ! and wins. For this reason, 1 5 . . . N-B2 at 43

THE

B EN KO

once is better than 1 5 . . .

GAMBIT

Q-R3 . Now the Knight will again be

m isplaced on QB2. Also. 1 6 . . . N-K4 is better than the text) 1 7 P-QR4 (With the better game for White) N-N3 1 8 K R-B 1 R-N 2 1 9 N/ B-Q 1 P-K3 ! ? 20 N-B4 N x N 2 1 P x N P X P ? 22 R X R Q X R 2 3 KP x P N-R3 (Q-R3) 2 4 N-B 3 . White has a clear advantage and finally won. After 1 P-Q4 N-K83 2 P-Q84 P-84 3 P-QS P-QN4 4 PxP P-QR3 5 PxP 8 X P 6 N-Q83 P-Q3 7 N-83 P-N3 8 N-Q2 8-KN2 9 P-K4 BX8 10 N X 8, the move 10 . . . Q-R4!? has been tried.

Th is is an original attempt to reach an endgame, which creates a difficulty for Wh ite because of the necessity to protect the KP. The game Kazilevis-Beyer, Siegen

Olympiad

1 9 70, continue d :

1 1 N-Q2?! (Thus t h e Knight is denied K3 . Of course 1 1 Q-B2 i s a blunder because of 1 1 . . . N x KP) 11 . . Q-R3!? 1 2 Q-K2 Q X Qc h 1 3 K x Q QN-Q2 1 4 N-B4 (With an edge for Wh ite) 0-0 .

1 5 P-KR 3 ? ! K R-N 1

(But now the chances are equal)

1 6 B-Q2

N-N3 1 7 N x N ( 1 7 N-K 3 ? N-R5) 1 7 . . . R x N 1 8 P-QN3 R-N5 1 9 P-B3 N-Q2 20 K R-QN 1 ? (KR-Q B 1 ) P-B5 2 1 R-QB 1 P X P 22 P x P QR-N 1

2 3 QR-N 1

(No better is 2 3 R-R7 N-B4 24 N-R4

N x N P) 2 3 . . . N-B4 24 N-Q 1 , drawn .

A

similar Queen m aneuver,

change Queens, 1 97 1 :

was

seen in

but without the

Voksalova-E.

intention to ex­

Bilek, B alatonszeplak

11 N-Q2?! QN-Q2!? (Q-R3 !?) 1 2 0-0 0-0 1 3 N-B4? ! Q-R3

14 Q-K2 KR-N 1 (White's Knight is in an unpleasant p i n . Black has the edge, but the game O'Kelly-Damj anovic, Montilla 1 97 2 , ended in

a

draw after 1 5 R-K 1 R-N 5 ) 15 B-Q2 N-N 3 16 P-QN 3 N X N

1 7 P X N R-N 7 ( 1 7 . . . R-N 5 ? 1 8 N-N5 and 1 9 N-B7) 1 8 P-QR3 ? (Necessary is 1 8 QR-N 1 , but not 1 8 Q-Q 3 ? N-N5 with an obvious 44

THE

BENKO

GAMB IT

advan tage) 1 8 . . . N-Q2 1 9 N-N5 Q-R5 20 QR-N 1 Q-B7 2 1 R X R B X R. Black has a winning positio n . White played better in a g a m e Toth-Dr. Foldi, B u dapest 1 9 7 2 :

1 1 B-Q2 Q-R 3 ! ? 1 2 Q - K 2 0-0 1 3 N-K3 Q N - Q 2 1 4 Q x Q R X Q 1 5 K-Q 1 ! ? (The King hurries to protect the Queenside) 1 5 . . . R-B 1 (Preparing P-B 5 . I f 1 6 N-B4 to prevent it, 1 6 . . . N - N 5 is strong) 1 6 K-B2 P-B5 1 7 P-B3 N-K4 1 8 K R-QN 1 P-K3 1 9 N-N5 ? ( 1 9 P-QN4 is correct) P X P 20 N X P/ 5 N X N 2 1 P x N R-B4 22 N-B3 N-Q2 (Perhaps 22 . . . R/ 4-R4 is more active, but Black is a little better now in any case) 23 R-Q 1 R/ 4-R4 24 P-QR3 N-N3 2 5 N-R2 ! ? N x P ! 26 R-K 1 B-K4? (Equality is reached after 2 6

. . .

R-R5 ! 27 R-K8ch B-B 1 2 8 B-R6 R-R 1 29 R/ 1 -K 1 N-B2 ! ) 27 B X R R X B 2 8 N-B3 N x N (N-B5) 29 P x N B X R P 3 0 P-R4. White has a clear advantage and later won. The maneuver Q-R4 is possible at other points too, but it offers essentially no difference from the variations discussed earlier i n this section. An example is Litmanovicz-E. Bilek, Vrnjacka B anj a 1 97 2 : 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P X P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 N-Q2 B-KN2 9 P-K4 B X B 10 N x B 0-0 1 1 N-K3 Q-R4 (instead o f 1 1 . . . QN-Q2, but this is merely a transposition) 12 0-0 QN-Q2 13 Q-B2 Q-R3 (A deviation from Karasev-Platonov, above . This Queen move is also possible after B lack's Q-N 3 is some lines) 14 P-B3 ! ? (White prepares P-QN 3 and B-N2 which is not p l ay able at once because of 1 4 . . . N X KP) 14 . . . N-N3 15 P-QN 3 KN-Q2 1 6 B-N2 KR-B 1 ! 1 7 N / B-Q 1 (A questionable maneuver since Black was not threatening P-B5 be­ cause o f the answer P-QN4) 17

. . . P-B3 1 8 R-B2 P-B5 19 N x P

N x N 2 0 P x N R X P 2 1 Q-K2 N-B4 2 2 N-K3 R-R 5 . The chances are equal, but Black later won.

C2 Before this variation was very old, B l ack succeeded in improving the line by castling-after all, he is not forced to play B X B . This new approach has grown in popularity and is now almost con­ sidered the Main Line. White's best play here is the maneuver N-KB3 -Q2-QB4-K3 , as analyzed a little further on. To begin with, however, let u s look at 45

THE

BENKO GAMBIT

some other attempts. If the reader studies this book with particular interest in the Black side of the gambit, he should of course fa­ miliarize himself w ith inferior White l ines and the methods of refuting them. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P x P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 N-Q2 B-KN2 9 P-K4 0-0(!)

In Toran-Benko, Malaga 1 970, the continuation was 10 B-K2?! P-K3 ? ! (Instead of this double-edged break, 1 0 . . . Q-R4 1 1 0-0 KN-Q2 ! seems better. See Benko-Chellstorp in the Appendix) 1 1 B X B !? N X B 1 2 0-0 P X P 1 3 P X P R-K 1 1 4 N-B4 N-K5 1 5 B-Q2 R-N 1 , w ith an edge for White, although the game was drawn. Once Black has played P-K 3 , the pressure exerted by a White Knight on QB4 against the QP is very unpleasant. 10 B X B!?

This move helps Black develop, but it has proven to be better than 10 B-K2. The game Pytel-Peterson, Arhus 1 97 1 , continued: 10 . . . R x B?! 11 0-0 (Here 1 1 Q-K2 ! is to be seriously considered, temporarily preventing Black's following maneuver, at the same time proceed­ ing with development) 11 . . . QN-Q2 12 N-B4 N-N3 13 N-K3 Q-R1?! (Q-Q2 looks better, after which 14 P-QR4 would be a blunder because of KR-R 1 ) 14 P-QR4 (To prevent N-R5) 14 . . . KN-Q2 15 B-Q2 B X N?! (A committing move. Instead, 1 5 . . . P-BS 1 6 Q-K2 KR-B 1 , intending to regain the pawn without giving 46

THE

BENKO

GAMBIT

up the strong Bishop, was possible) 1 6 B X B N X RP (Had Black realized what White was planning, he would have played 1 6 . . . R X P and after 1 7 P-QN3 R X R 1 8 B X R Q-R7 1 9 P-B4 White has only a small edge) 1 7 Q-Q3 ! R-R2 1 8 N-B5 N-K4? ( 1 8 . . . R-K 1 is forced; then 1 9 N-R6ch K-B 1 20 Q-B3 P-B 3 leaves White with only slightly better chances) 19 R X N ! (With a winning posi­ tion) 19 . . . R X R 20 N X Pch K-N2 2 1 P-B4 P-B3 22 P X N BP X P 2 3 R-K 1 R-R8 24 Q-N3 R X Rch ( Q-R3 ) 2 5 Q X R Q-R3 2 6 P-R3 Q-Q6 27 N-B6 K-N 1 (27 . . . R-B5 2 8 N X P ) 28 B-Q2 Q-QN6 29 B-R6 R-K 1 30 Q-B2 Q-N2 3 1 Q-B6 Q-Q2 32 K-R2 Q-QB2 3 3 B-B 8, Black resigned. Following this line, the game Vasnetsky-Gaprindashvili, Vrnjacka Banj a 1 97 1 , varied on move 1 5 . Instead of 15 B-Q2, White tried 15 Q-B2, P-B5! 16 B-Q2 N-B4 (Black already has a slight advantage) 1 7 R-R3 R-B 1 1 8 N-N4 N-N6 1 9 B-B4 N x RP 20 N XN R X N 2 1 R X R Q X R 22 B-N5 (22 P-K5 P-R4) R-B2 23 B-R6 B-R 1 24 N-K3 Q-R7 25 N-Q 1 P-B6 (Black has a win) 26 B-K3 N-Q5 27 B X N B X B 28 P-R4 P X P 29 Q X R P-N8=Q 30 Q-B8ch K-N2 3 1 Q-N4 Q-Q6, White resigned. 10 . . . N x B 11 0-0 N-Q2

This Knight is ready to go to QN3 either to exchange or drive away White's Knight should it be played to QB4. The game Ghitescu-Benko, Siegen Olympiad 1 970, continued 12 Q K2 see the Appendix for the full game. More usual is 12 N-B4 N-N3, and Tatai-Browne, Malaga 1 970, continued 13 Q-K2?! (The exchange of Knights facilitates Black's counterplay on the Queenside; see the next example) 13 . . . N X N 1 4 Q XN Q-N3 1 5 N-Q 1 (This retreat is not very promising, but worse is 1 5 N-R4? Q-N5 16 Q-B2 N-B2 17 N-B3 KR-N 1 with an advantage for Black although White eventually won in Ivkov­ Browne, Zagreb 1 970) 1 5 . . . N-N5 1 6 P-QR3 Q-R3 17 N-K3 N-Q6. Black has a slight edge, but the game was later drawn. Black's task is simplified if White exchanges Knights on his 1 3th move. The game Weber-Delgado, Skopje Olympiad 1 972, is a case in point: 13 N x N (N-K3 !) Q x N 14 Q-K2 KR-N 1 15 K-R 1 N-B2 16 P-B4 N-N4 17 P-K5 N-Q5 1 8 Q-KB2 P-B4 19 P X P e.p. B X P 20 R-QN 1 N-B4 2 1 B-Q2 Q-N2 2 2 Q-B3 Q-Q2 2 3 KR-K l N-Q5 24 Q-Q3 Q-B4! 25 Q x Q P X Q (Black is clearly better) 26 P-KR3 ? -

47

-

THE

BENKO

and Black won. Much stronger for White after N-B4 N-N 3 , i s 1 3 N-K3!

GAMBIT

N-N4

10 B X B N x B 1 1

0-0 N-Q2 1 2

This is how the game O'Kelly-Toran (Olot 1 9 70) continued. Black's buildup on the QN file is hindered to a certain extent by the Knight on QN 3 . This line may be considered the principal variation in this section-its results have strongly favored White until quite recently, when improvements have been found for Bl ack . 13 . . . Q-Q2 14 P-QR4!?

Although this move weakens the Queenside, it does cast into doubt Black's control of certain Queenside squares. In addition, White's Queen Rook may be activated via QR3 , as will be seen. O'Kelly-Toran continued: 14 . . . Q-N2 15 R-R3 KR-N 1 16 P-R4!? (White tries to make use of the fact that Black's pieces are absent from the Kingside and he launches an attack there) 16 . N-Q2 17 P-KRS N-B3? (Black must have been afraid of Q-N4, but now it turns out that his entire strategy has been faulty since his Knight returns to its original developed square, having expended four tempi to accomplish nothing of importance. 1 7 . . . Q - N 3 was necessary immediately) 1 8 P X P RP X P 1 9 P - B 4 (White has a clear advantage) Q-N5 20 P-K5 N-R4 (Black's Knight stands badly here, but 20 . . . N-K5 would have been met by 21 Q-B 3 N x N 22 P X N Q-N2 2 3 P-B5) 2 1 Q-B3 P X P (The threat was P -KN4) 2 2 P X P Q-R5 (B X P i s also bad. White has a winning attack) 2 3 Q x P ch .

.

K-R2 24 N-B 5 ! P x N 2 5 Q x Pch K-R 1 26 N-K4 R-KB 1 27 N-B 6 ! N X N 2 8 P x N B-R3 2 9 R-R3 Q-Q5 ch 3 0 K-R2 R X P ( 3 0 48

. . .

THE

BENKO

GAMBIT

Q X BP 3 1 B X B) 3 1 Q-N5, Black resigned. Voksalova-Gaprindashvili, Vrnjacka Banja 1 97 1 , continued: 17 . . . Q-NS (instead of N-B3) 18 P X P BP X P? (RP X P) 19 Q-N4 N-B 1 20 P-B4!? B X N 2 1 P X B? ! (R X B) Q x KP 22 P-B5 Q X Q 23 N x Q P X P 2 4 N-R6ch K-R 1 2 5 N X P N-N3 2 6 B-N5, w ith a slight edge for White. For more on the 1 3 . . . Q-Q2, see Vranesic-Benko in the Ap­ pendix. From the last diagram, the move 13 . . . Q-B1?! was tried in Vranesic-Diez del Corral, Siegen Olympiad 1 970, with the idea of developing the Queen on QR3 and leaving Q2 free for the Knight in case White plays P-QR4-5. The game continued: 14 P-QR4!? N-NS 15 R-R3 Q-R3 16 N-NS Q-N2 17 Q-N3 P-B4? (Black tries to introduce complications on the other side, since 1 7 . . . N X RP would be met by 1 8 N x P P X N 1 9 R X N R X R 20 Q X R N-Q6 21 N-B4 w ith advantage to White) 1 8 B-Q2 (But now White has the advantage anyway) P-KB5 ( 1 8 . . . P X P 19 B X N, etc.) 1 9 N-Q 1 N X RP 2 0 R X N Q X N 2 1 R X R R X R 2 2 N-B3 Q-Q6 23 B X P R-N 1 (B-Q5) 24 B-N 5 ! P-R3? 25 B X KP N-B3 26 Q-R4 N X B 2 7 R-Q 1 (The point of White's winning combination-Black's Queen is trapped) 27 . . . P-B5 28 R X Q P X R 29 N-Q 1 R-B 1 30 Q-N3 P-Q7 3 1 K-B 1 P-N4 3 2 K-K2 N-N3 3 3 K X P B-Q5 34 Q-KR3 R-K 1 3 5 QXP K-B2 3 6 Q x P, Black resigned. In a game Doroskevich-Georgadze, USSR 1 972, the continuation was 17 P-QN3 (instead of 17 Q-N3) P-B4 1 8 P X P P X P 19 Q-R5 Q-Q2 20 R-Q 1 P-KB5 21 N-B4 N X N 22 P X N N-B7 23 R-R3 Q-B4 24 Q-K2 R-B2 (N-Q5 !) 25 R-Q2 (White gets a powerful position by 25 B X P ! and if 25 . . . R X P 26 R-KB3) 25 . . . N-N 5 26 R-R3 Q-N8, and White had only a small advantage and eventu­ ally won. The idea of . . . Q-B 1 was used with another plan in Fitzgerald­ Browne, Washington 1 970: 13 . . . P-BS!? 14 Q-K2 Q-B1 (Black h as occupied his QB5 while his Q5, which he has abandoned, is temporarily unavailable to White) 15 N-B2? ! ( 1 5 B-Q2 is answered by N-N5 with an unclear position) 1 5 . . . N-B4 16 B-K3 N / 3-R5 17 N-Q4 N x N 1 8 P x N N x P 19 N-B6 (White has offered too much to achieve this Knight position, which is not of such great value because now that his KP-the support of his QP-has fallen, the QP can be easily undermined by a timely . . . P-K3) 19 . . . N X QBP 20 Q x P Q-Q2 2 1 P-QR4 KR-B 1 22 QR-B 1 (22 KR-B 1 49

THE

BENKO

GAMBIT

N X QP) P-K3 ! 2 3 R X N P X P 24 N-K7ch K-B l ! 25 Q X Rch R X Q 2 6 R X Rch K X N 2 7 KR-B 1 Q X P, and Black has the better chances. The move . . . P-B5 is worth studying in this context. The move 13 . . . Q-B2!? has also been played. For the time being, QB2 is probably the best place for the Queen, since the way is clear for the KR to reach the Queensidc and Q2 is left vacant for possible occupation by the Knight. The game Soos-Jakobsen, Stockholm 1 972, continued: 14 B-Q2 (P-QR4 !?) KR-N 1 1 5 Q-K2 P-B5 ! 1 6 KR-B 1 (P-QR4 !?) N-B4 17 N x P? ( 1 7 QR-N 1 N-Q6 is unclear, but 17 . . . B X N is dubious) 1 7 . . . N x N 1 8 Q x N R X NP 1 9 B-K3 Q-N2 2 0 B X N?! (20 P-B3) P X B (Black i s better) 2 1 Q X P? R-B 1 (And now he is winning) 22 Q-K3 R X N 23 R X R B X R 2 4 R-Q 1 R-N8, White resigned. Instead of trying to challenge the White Knight by . . . N-N 3 , another idea was tried in Doroskevich-Georgadze, USSR 1 972 (not to be confused with another game between these two players given earlier). In the position after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P x P P-QR3 5 P x P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 N-Q2 B-KN2 9 P-K4 0-0 10 B X B N X B 11 0-0 N-Q2 12 N-B4,

Georgadze played 1 2 . . . N-B2 !? 1 3 Q-K2 Q-N 1 14 B-Q2 (B-N5) N-N4 (The Knight is headed for Q5. If 15 N x N Q X N 16 KR-K l , then 1 6 . . . R-R5 i s unpleasant for White) 1 5 KR-B 1 N-Q5 1 6 Q-Q3 N-N3 1 7 N-K3 Q-N2 1 8 QR-N l Q-R3 1 9 Q X Q R X Q 20 K-B 1 (20 P-QN4 is better. The King does not stand well on KB 1 ) 2 0 . . . P-B4 ! 2 1 P-QN3 (To meet the possible P-KB5 o r N-B5. 2 1 P X P would b e answered b y N X B P and the Wh ite QP would become fatally weak. Black has the edge) 2 1 . . . P X P 22 P-QN4 P x P 23 RXP N-B4 24 RXP B X N 25 RXB RXP 26 R-Q3 N-Q2 27 N X N R X N 2 8 B-K3 N-B3 2 9 R X P N X P 30 R-Q7 N X Bch 3 1 R X N R / 4 X Pch. Black has a clear advantage and later won.

C3

Finally Black discovered how to cross White's plan of N-KB 3 Q2 X B by playing a n early Q-R4. In this case White has a choice of insisting on exchanging B ishops by P-K4 and being obliged to recapture with the King, or of transposing into the Fianchetto Variation (see E), temporarily giving up P-K4. If White wishes to play P-K4 and s till be able to recapture with the Knight, he will -

50

THE

BENKO

GAMBIT

have to play P-B3(?!) first. However, this move does not belong in this system. There is not much material on this variation; consequently we are unable to provide a definitive analysis at this time. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P x P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 N-Q2 Q-R4!?

As a matter of fact, the move Q-R4 was adopted by Black in Donner-Browne, Amsterdam 1 97 1 , but as early as the 7th move, i.e. 7 N-B3 Q-R4. Its independent significance lies, however, only in that White is prevented from playing an immediate 8 P-K4 B X B 9 K x B. That game continued: 8 N-Q2 (B-Q2!?) QN-Q2 9 P-KN 3 P-N3 1 0 P-K4 B X B (Routinely played. Better is 1 0 . . . B-KN2 to make use of White's inaccurate order of moves) 1 1 K X B B-N2 (And so we have arrived at a normal position, similar to those analyzed in section B) 1 2 N-B4 Q-N5 1 3 Q-K2 N-N3 1 4 N X N Q X N 1 5 K-N2 0-0 1 6 QR-N l KR-N l 1 7 B-Q2 N-K l 1 8 P-N3 N-B2 1 9 P-QR4 (White has a little edge) R-N2 20 KR-QB l QR-N l 2 1 Q-Q3 Q-R2 22 Q-B2 Q-R3 23 N-Q l Q-K7 24 N-K3 N-R3 25 R-K 1 Q-R4 26 N-B4 R-KB l 27 N-R5 R-B2 28 Q-B4 N-N 1 29 P-QN4 (White is now definitely better) P-B4 30 P-B4? (30 P-K5 ! and if 30 . . . B X P 3 1 R X B and 3 2 P-Q6ch wins, or 30 . . . N-Q2 3 1 P X P P X P 32 N-B6) 30 . . . P X KP 3 1 Q x KP (3 1 R X P!?) P X P 32 R X P N-R3 33 R-B4 N-B4 34 Q-B3 Q x Qch 35 K x Q R-B4 36 N-N3 R X P 37 B-K3 R-Q6 38 N X N R X N 39 R X R P X R 40 K-K4 R-R6 41 B X P R X Pch 42 K-B3 B-B3, drawn. To return to the diagram position, after 8 . . . Q-R4, the game Kraidman-Browne, Netanya 1 97 1 , continued 9 P-K4 B X B 10 K X B 51

THE

BENKO GAMBIT

( 1 0 N X B? N x KP) B-N2 (Possible here is 1 0 . . . Q-R3ch 1 1 Q-K2, but on 1 1 . . . Q X Qch 12 K x Q, a position occurs that is similar to one reached after 10 . . . Q-R4 in Variation C 1 , but here White is a tempo behind. It is worth comparing Q-R4 in C 1 with the present variation) 11 P-KN3 QN-Q2 (Here, too, Q-R3ch is playable, since P-KN3 would then turn out to be a wasted tempo. 1 1 N-B4 would have been more exact) 12 K-N2 0-0 (Q-R3) 1 3 Q-B2 (More natural is 1 3 N-B4 which could lead to the continuation of the previous game) 1 3 . . . Q-R3 ( 1 3 . . . KR-N 1 14 P-N 3? ! N x QP 1 5 N x N B X R 1 6 N X Pch K-B 1 1 7 N-B6 Q X P 1 8 Q X Q R X Q 1 9 N X R N x N 2 0 N-B4 B-Q5 2 1 R-B 1 K-K2 2 2 B-B4 R-R3, with approxi­ mate equality) 14 P-QR4 ( 1 4 P-N3? N-K4 15 B-N2 Q-Q6 ! 1 6 KR-QB 1 B-R3 ! 1 7 P-B4 N/ 3-NS and wins) 1 4 . . . KR-N 1 1 5 R-R3 R-N5 1 6 N-N5 N-K 1 1 7 P-N3 N-N3 1 8 R-K 1 P-B 5 ! (The game is equal) 19 N x BP N X N 20 P X N R-B 1 2 1 Q-Q3 R/ 1 X P (R/ 5 X P) 2 2 N-B3 Q-B 1 (If Black wishes t o win, the plan N-B3-Q2 is better) 23 B-Q2, drawn. Marschalak-Wach, Wistla 197 1 , continued : 12 N-B4 Q-R3 1 3 Q-K2 0-0 1 4 K-N 2 KR-N 1 1 5 B-Q2 N-N3 1 6 P-N 3 KN-Q2? ! (If 1 6 . . . N X N ! ?, we get into the game Voksalova-E. Bilek in section C l . Now it can be seen that there is a great similarity between the two lines and how easily they can transpose) 17 N x N R X N 1 8 Q x Q R/ 3 X Q 1 9 P-QR4 P-B5 ! (Breaking u p White's pawn con­ figuration and providing a fine post for the Black Knight. The sacrifice is only temporary. The move P-B5 is typical in some variations of the gambit-see also the previous game) 20 P x P N-N3 (The chances are equal) 21 KR-QN 1 N X BP 22 B-K 1 N-K4 23 R-R2 N-Q6 24 B-Q2 N-B4 25 P-B3 P-B4 (Black has time for everything) 26 P x P P X P 27 R-QN4, drawn. A very similar position was reached by transposition in the game Kuzmin-Stein, USSR Championship 1972: 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P x P P-QR3 5 P X P P-N3 6 N-QB3 B X P 7 N-B3 B-KN2 8 N-Q2 P-Q3 (8 . . . Q-R4!) 9 P-K4 Q-R4!? (Black's plan is to go into an early endgame) 10 B X B Q X B 1 1 Q-K2 0-0

(Careless. Q x Qch is more consistent) 1 2 N-N5 ! QN-Q2 1 3 P-QR4 KR-B 1 14 N-B4 N-K 1 15 R-R3 N-B2 16 N X N R X N 17 B-Q2 (White has a slight edge) 17 . . . N-N 3 1 8 P-N3 R-N2? 19 N-R5 ! R-B2 20 Q x Q R X Q 2 1 K-K2 P-BS 22 N-B6 (Now White has a clear superiority) 22 . . . P x P 23 R x P P-B4 24 P-R5, and White won. 52

THE

BENKO

GAMBIT

Instead of 9 P-K4 in the diagram position, the move 9 P-KN3(?) has been tried in Szabo-Bronstein, USSR 1 97 1 . Of course, that move means forsaking P-K4, and I am of the opinion that it will not find many followers. White in fact transposes into the Fianchetto Variation (section E), with the difference that he has played N-Q2, which is not a part of the fianchetto system. However, Black's Q-R4 fits into it quite well and is often seen. The game continued: 9 . . . QN-Q2 1 0 B-R3(?) B-KN2 1 1 0-0 0-0 1 2 Q-B2 P-BS (Eyeing the QP. Now on 1 3 P-K4 Black's Q6 would present itself for a Black Knight) 1 3 R-Q 1 N-N3 (Black stands slightly better) 1 4 B-N2 (So the plan B-R3 did not work, as the protection of QS is forced) 1 4 . . . B-N2? ! (White's tempo loss i s met b y Black's. Better i s 1 4 . . . KR-B 1 ) 1 5 P-QN4!? P X P e.p.?! ( 1 5 . . . Q X NP!? is still a little better for Black although after 16 R-N 1 White would obtain some counterplay on the QN file) 16 N x P Q-R3 17 P-K4 KR-B 1 1 8 B-B 1 Q-R2 1 9 N-Q4 Q-R4 20 B-Q2 Q-B4 2 1 N-N3 Q-R6, drawn.

53

THE BENKO

GAMBIT

D Quiet Variation (5 P-K3?!) 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P-K3?!

This is a quiet treatment of the gambit which, however, will usually mean the loss of a tempo for White since he w ill most often play P-K4 later. The significance of this method is also de­ pendent to a certain extent on the current valuation of the Volga Gambit (5 . . . P-K3), since against that opening White's 5 P-K3 is today considered the safest line. But against 5 P-K3, the Benko Gambit fianchetto, 5 . . . P-N3, can be comfortably adopted. 5 . . . P-N3

Instead of the fianchetto, 5 . . . P X P is premature because of 6 B x P Q-R4ch 7 N-B3 N-K5 8 N-K2. If Black intends to capture w ith his RP, it is best to wait until White's KB has already moved so that White will be forced to lose a tempo in recapturing. 6 N-QB3 P-Q3

Instead of this, 6 . . B-KN2 7 P x P 0-0 8 N-B3 was adopted in Kaufman-Benko (see the game in the Appendix), and after 8 . P-Q3 the game transposed into the regular line. However, White may consider entering the unclear complications of 8 P-Q6. After 8 . . P-Q3, the game Benko-DeFotis, U.S. Championship .

.

.

54

.

THE BENKO

GAMBIT

1 972, continued: 9 P-K4 Q-R4 !? 1 0 B-Q2?! (A suggestion by Portisch after our game in Palma 1 97 1 . See the game in the Ap­ pendix) 10 . . . B X P 1 1 B X B?! (On the more natural 1 1 B-K2 there may follow 11 . . . Q-N5 ! with an unpleasant attack on White's QNP and KP, e.g. 1 2 Q-B2 B X B 1 3 K X B? ! with an edge for Black, or 1 2 B X B N X B 1 3 Q-B2 Q-B5 with a definite advantage for Black, but better is 1 3 Q-K2 and 0-0) 1 1 . . Q X B 1 2 Q-K2 KN-Q2 ! 1 3 P-QR4 !? ( 1 3 B-N5 is better) 1 3 . . . Q x Qch 14 K X Q N-R3 (Black has a small advantage) 1 5 QR-QN 1 KR-N 1 (To be considered was 1 5 . . . N-B2, or even 1 5 . . . P-B4 with complica­ tions) 16 N-QN5 N-B3 17 K-K3 N-N5ch 1 8 K-K2, drawn. The repetition of moves is not to be avoided, since on 1 8 . . . N-B3 the attempt 1 9 K-Q3 is dangerous because of 1 9 . . . P-B5ch 20 K x P N x KP o r the immediate . . . R-B 1 ch. .

Position after 6 . . . P-Q3

We will now look at the following main possibilities : 7 N-B3, 7 P-K4 and 7 P X P, as well as a few other less significant lines. 7 N-83 B-KN2 8 P-K4 0-0 9 P X P Q-R4!? 10 N-Q2 B X P 1 1 B X B Q X B 1 2 Q-K2 QN-Q2 was played in the game Farago-Benko, Vrnjacka Banj a 1 970. The move of the QN is tamer than 1 2 . . . KN-Q2 as in the previous example (Benko-DeFotis) and leads only to equality. The game continued: 1 3 Q x Q R X Q 14 N-B4 R-N 1 1 5 P-B3 N-K 1 1 6 K-Q2 N-B2 1 7 K-Q3 N-K4ch 1 8 N X N B X N 1 9 K-B2 P-B4! (The chances are equal) 2 0 P-KR4 N-K l 2 1 P-R5 BP x P 22 RP X P RP X P 23 N X P N-B2 24 R-Q l P-B5 (Black's aggressive position now gives him better chances) 25 N-B3 B X N 26 P X B R-N4 27 P-R4 R/ 4-R4 (Perhaps 27 . . . R X QP ! ? offered 55

THE BENKO

GAMBIT

better chances for initiative) 28 R-Q4 R X RP 29 R X R R X R 30 K-N2 (Once again the position is equal) 30 . . . K-B2 31 B-R6 R-R4 32 R-B4ch K-K 1 33 R X P N x P, drawn. In Johansson-Kvarby, Sunbyholm 1 97 1 , the continuation was 8 . . QN-Q2 9 B-K3(?) 0-0 1 0 B-K2 P X P 1 1 B XNP B-QR3 1 2 B X B R X B 1 3 0-0 Q-N3 1 4 Q-B2 R-N 1 1 5 KR-N 1 N-N5 1 6 B-B4 N/ 5-K4 (P-B5) 17 N-Q2 Q-N5 1 8 P-KR3 P-B5. Black has a slight advantage and later won. In Erwick-Lundwall, Wijk aan Zee 1 970, White reconciled him­ self to the loss of a tempo by 8 B-K2, and there followed : 8 . . 0-0 9 0-0 P X P 1 0 B X P B-QR3 (N-R3) 1 1 B X B R X B? ! (N X B is better) 1 2 Q-K2 ! Q-R4 1 3 N-Q2 QN-Q2 1 4 N-B4 Q-B2 1 5 P-K4 KR-R 1 1 6 R-N 1 N-N3 1 7 P-QR3 KN-Q2 (Black should have played . . . N X N before now) 1 8 N-N5 Q-N2 1 9 P-QN3 N-K4 20 N-K3 , with White better. The game Danner-Toran, Vienna 1 972, continued : 8 P X P 0-0 9 B-K2 B X P 1 0 0-0 Q-N 3 1 1 P-K4 QN-Q2 1 2 R-N 1 N-N5 (KR-N l ) 1 3 B-N5 ! KR-N 1 14 N-Q2 ( 1 4 B X P P-B3) 14 . . . N/ 5 -K4 1 5 P-QR3 N-KB 3 ? (N-Q6) 1 6 P-R3 P-R3 1 7 B-K3 N-K 1 1 8 P-B4 B X B 1 9 Q X B, with the better game for White. Returning to the last diagram, instead of 7 N-B3, the immediate advance 7 P-K4 was tried in Kchouk-Toran, Siegen 1 970, which continued: 7 . . . B-KN2 8 P x P 0-0 9 P-B3(?) N X RP 10 B-QB4 N-Q2 ! 1 1 KN-K2 N-K4 (White's KB does not find a satisfactory position in this setup) 1 2 B-N3 P-BS 1 3 B-B2 Q-N3 ! (Black already has a clear advantage) 14 N-R4 Q-R2 1 5 P-QR3 B-Q2 1 6 N/ 2-B3 N-B2 17 K-K2?! Q-R3 1 8 B-K3 B X N 19 N X B ( 1 9 B X B N-Q6 is obviously good for Black) 19 . . . P-B6ch 20 K-B2 P x P 21 N x P KR-N 1 22 N-Q3 N-BS 23 R-R2 N x B 24 K X N N-N4 (Black is winning) 25 Q-K 1 N-B6 26 R-R 1 Q-BS 27 K-B2 N X Pch 28 P X N B X R 2 9 Q X B Q X Bch 3 0 K-K3 R-N6, White resigned. Another dubious try by White was seen in Hamman-Gheorghiu, Skopje Olympiad 1 972: 9 P-B4?! Q-R4 1 0 B-Q2 B X P 1 1 B X B N X B 1 2 N-B3 N-QN5 1 3 0-0 P-BS (Black has a n edge) 1 4 K-R 1 N-Q6 1 5 Q-K2 !? (If 1 5 P-QN3 N-KNS wins) 1 5 . . . KR-N 1 1 6 QR-N 1 R X P 1 7 R X R N X R 1 8 N-Q4 ! N-Q6 (Black still has the preferable position, but the game is a long way from being over) 1 9 N-B6 Q-B2 20 P-KS N-K 1 2 1 P X P N x QP 22 N x Pch (Q X P was a little better) 2 2 . . . K-B 1 2 3 N-B6 R-K 1 24 Q-B3 Q-N 3 25 N-Q 1 N-K5 26 B-R5 Q-N4 27 P-BS ?! (N-K3) 27 . . . Q X P (Now .

.

56

THE

BENKO

GAMBIT

Black is definitely on top) 28 P X P P X P 29 N-Q8 P-B4 30 B-N6 B-B3 3 1 N-K3 Q-Q2 32 P-N4 P-B5 33 N x P B X N 34 B X B Q X B 3 5 K-N2 Q-Q5 3 6 R-Q 1 N/ 5-B7 37 N-Q6 Q X N 3 8 Q x N / 2 Q-B3ch, and White resigned. The immediate 7 P X P (from the last diagram) was played in Ree-Bellon, Caorle 1 972. The game continued: 7 . . . B-KN2 8 B-N5ch KN-Q2 9 KN-K2 0-0 1 0 0-0 B X P?! (Better is 1 0 . . . N X P in connection with . . . N-B2) 1 1 B X B N x B 1 2 P-K4 Q-N 3 1 3 R-N 1 N-B2 1 4 B-N5 KR-N 1 1 5 Q-Q2 ( 1 5 B x P P-B3) 1 5 . . . K-B 1 1 6 B-R6 N-K4. Black stood better and eventually won. For more on 7 P X P B-KN2, see the game Kuypers-Benko in the Appendix. After 7 P X P, Black tried 7 . . B X P(?) in a game Laurine-Green, Eskilstuna 1 97 1 . The continuation was: 8 B X B N x B 9 KN-K2 B-N2 1 0 0-0 0-0 1 1 Q-Q2? N-QN5? 12 P-QR3 Q-Q2 1 3 R-N 1 N-R3 1 4 P-QN4 (White stands better) 1 4 . . . P X P 1 5 P X P KR-N 1 1 6 P-N5 N-B4 1 7 P-B3 N-K 1 1 8 N-Q4 (Now White is winning) 1 8 . . . R-N3 1 9 B-N 2 N-B2 20 R-R 1 QR-N 1 2 1 R-R7 B X N 22 Q X B P-K4 23 P X P e.p. P X P 2 4 Q-B6, Black resigned. Any reasonable reply is answered by 25 N-Q5 . Grandmaster Reshevsky, in his game against Browne in Skopje 1 970, played 7 P-N6?!, but the return of the pawn in this way allows Black to get a comfortable game with no particular difficulty. The game continued: 7 . . . QN-Q2 8 P-QR4 Q X P 9 P-R5 Q-B2?! (Q-N2 is better) 1 0 N-B3 B-KN2 1 1 P-R3 0-0 1 2 B-B4 R-N 1 13 0-0 N-K 1 1 4 Q-K2 Q-N2 15 B-Q2 N-B2 1 6 Q-Q3 N-K4 (Black's position is slightly preferable) 1 7 N x N B X N 1 8 P-B4 B-N2 1 9 QR-N 1 B-Q2 20 K-R 1 B-N4 2 1 P-QN3 P-K3 22 P-K4 P X P 2 3 P X P B-Q2 2 4 P-B5 ! ? B X P 2 5 R X B P X R 2 6 Q X P Q-B 1 27 Q-R5 P-B4 28 R-KB 1 Q-K 1 29 Q-R4 Q-N 3 30 N-K2 B-B3 3 1 Q-B2. Black later won, but the present position is quite unclear. For 7 B-B4?!, see the game Blumin-Benko in the Appendix. After 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P-K3, a slight transposition occurred v i a 5 . . P-N3 6 N-QB3 B-KN2. The game Holm-Yisier, Kapfenberg 1 970, con­ tinued with the awkward 7 KN-K2? 0-0 8 N-B4 P-Q3 9 B-B4 QN-Q2 10 P-QR4 N-N3 1 1 B-K2 P X P 12 B X P B-QR3 13 B X B R X B 1 4 P-K4 Q-Q2 15 0-0 KR-R 1 (Black i s clearly better) 1 6 P-K5 P x P 1 7 N-Q3 Q-82 1 8 Q-83 N X RP 1 9 N x N P-K5 ! (Not 1 9 . . . R X N? 20 P-Q6 P X P 2 1 Q x Rch ! R X Q 22 R X Rch B-B 1 .

.

57

THE

BENKO

GAMBIT

23 B-R6 N-Q2 24 KR-R 1 , etc.) 20 Q-B4 Q x Q 2 1 N X Q R X N 2 2 R X R R X R 2 3 B-K3 N-Q2 2 4 R-N 1 B X P 2 S P-B3 R-NS 26 P X P B-QS 27 R-K1 R-N 8 ! and Black won, of course. Instead of 6 N-QB 3, the game Biro-Dr. Foldy, Budapest 1 973, went 6 N-KB3 P-Q3 7 P x P B-KN 2 8 B-N Sch QN-Q2? ! (KN-Q2 !?) 9 0-0 0-0 10 N-B3 Q-R4 !? 1 1 Q-K2?! B X P 12 B X B Q X B 1 3 Q x Q R X Q (Black has sl ightly better chances due to his active Rooks and powerful KB) 1 4 R-N 1 ( 1 4 B-Q2 N-N3-BS) 14 . . . R-N 1 1 S P-QN 3 N-K1 1 6 B-N 2 B x N 1 7 B x B R x R P (Having won his pawn back while keeping his positional edge, Black stands clearly better) 18 P-QN4 R-B7 19 B-Q2 P X P 20 R X P R X R 2 1 B X R N/ 2-B3 22 N-Q4 R-BS 2 3 B-Q2 N-KS ? ! (23 . . . N x P 24 R-B 1 ) 24 B-RS N/ 1 -B3 2S P-B3 N-B4 26 R-Q 1 N x P (Black has a winning position) 27 N-BS P x N 28 R X N N-N2? (P-BS) 29 R X BP P-K4. Black now has only a tiny advantage, and the game was drawn eventually. Another try for White was seen in Korchnoi-Calvo at Palma de Majorca 1 972, which went: 5 . . . P-Q3 (instead of the immediate 5 . . . P-N 3) 6 N-QB3 P-N3, and now a new idea in this position, 7 P-QR4?!, which is interesting in comparison with older methods examined in Variation A. The continuation was 7 . . . B-N2 8 Q-N3 P X P (Is this necessary because of the 9 P-N6 threat? Possible is 8 . . . QN-Q2 !?) 9 B X Pch B-Q2 (More promising is 9 . . . KN-Q2 !? to be followed by N-R3-B2 and B-QR3) 1 0 R-R3 0-0 (White's idea was 1 0 . . . B X B 1 1 P X B) 1 1 KN-K2 N-R3 1 2 0-0 N-B2 1 3 B-B4 R-N 1 1 4 Q-B2 (White is slightly better) B-B 1 1 5 P-QN3 B-QR3 1 6 B-Q2 Q-B 1 1 7 Q-Q3 Q-N2 1 8 P-K4 N-Q2 1 9 P-B4 N-N3 2 0 N-N S ! ? B X N 2 1 P X B R-R 1 (2 1 . . . N-B l ! 22 R-RS N-R2 or 22 . . . Q-N3 is unclear) 22 R X R R X R 23 B-B3 B X B 24 N X B R-R6 25 P-KS . White stands better and Korchnoi finally won after 54 moves. We can credit his victory, however, to his fighting abilities rather than to his opening novelty.

58

THE

BENKO

GAMBIT

E

Fianchetto Variation 1 P-Q4 N-KB3 2 P-Q84 P-84 3 P-QS P-QN4 4 P x P P-QR3 5 P X P 8 X P 6 N-Q83 P-Q3 7 N-83 P-N3 8 P-KN3 8-KN2 9 8-N2 0-0 10 0-0 QN-Q2

The fianchetto of the White King Bishop is a frequent choice. developing his Bishop in this way he of course abandons the diagonal KB 1 -QR6 to his opponent's Queen Bishop, but he is able to castle without difficulty. Tournament experience has shown that in this variation too, Black obtains enough counterplay to com­ pensate for the gambit pawn. As in other variations, Black will occupy the QN and QR files with his Rooks, and by the Knight maneuvers N-N3-B5 and N-K l -B2-N4 he will try to exert pressure against White's Queenside with the invaluable aid of the powerful King Bishop. White's KB, on the other hand , frequently plays only a m inor passive role or eventually is forced to return to KB 1 to be exchanged for Black's QB. The most frequently played moves for White here are 1 1 Q-82, 11 R-N1 and 11 R-K1. These three moves may be played in one or another sequence or combination and the variations may trans­ pose b ack and forth. But the Black replies are what give the varia­ tions their individual character. There are three Queen developing moves for Black to consider: 11 . . . Q-R4, 11 . . . Q-N3 and 11 . . . Q-B2. It has not yet been determined which of the three is "best, " since each has certain advantages and disadvantages compared with the others. The effiBy

59

THE

BENKO

GAMBIT

cacy of one or the other move m ay therefore be said to depe nd o n B l ack's mood.

E1 1 1 . . . Q-R4 seems superficially t o b e the most aggressive. Its disadva ntage, however, is that the Queen can be m o re easily at­ tacked here than o n other squares, for example a fter White plays B-Q2, when B l ack m u s t be wary of sudden moves by White's Queen Knight. An u nsuccessful approach for White was seen i n Podgaetz-Peev,

11 R-N 1 Q-R4 12 Q-B2 KR-N 1 13 R-K1?! ( 1 3 R-Q 1 , protecting the QP, is better) 1 3 . . . Varna 1 972 (continue from last d i agram) :

P-R3 ? ! 1 4 B-R3 B-B5 ! 1 5 N-Q2 N-N3 1 6 N x B (If 1 6 P-K4 N - K l and White has problems w i t h his center pawns) 1 6 R-Q 1 N-K l . . .

. . . N xN 1 7

1 8 B-Q7 (If 1 8 B-Q2 N X B 1 9 R X B R X P ! , etc . ) 1 8

B X N 1 9 P X B R X R 2 0 Q X R N-B 3 . Black i s a little better

and eventually won.

1 1 Q-B2 Q-R4 12 R-N 1 was played in Padevsky-B rowne, Sara­ 1 9 7 0 , which continued: 12 . . . KR-N 1 1 3 B-Q2 N -N 3 1 4

jevo

P-N 3 Q-R6 (Black has equal ized) 1 5 B-B 1 Q-R4 1 6 B-Q2 Q-R6 1 7 B-B 1 , drawn b y repetition o f moves. A n improvement for Black was seen in Forintos-Browne, Skopje Olympiad 1 9 7 2 :

12 . . . N-N3! 1 3 R-Q 1 N-B5 14 N-Q2 N-Q2 1 5

B-R3? (Driving th e Knight to a better square. I t was better to simplify by N X N) 1 5

. . .

N / 2-N3 1 6 P-R3 N x N 1 7 P-QN4(?)

N x R ! (Black now gets a clear superiority) 1 8 Q x N (If 1 8 P x Q N x N 1 � P X N N x R 2 0 Q x N KR-N 1 w ith the better of it) 1 8 . . . N X P 1 9 N x N P X P 2 0 P X P Q-Q 1 2 1 B-N 5 R-K l 22 P-N 5 B-N2 and B lack had a m aterial and positional superiority although he eventually lost the game.

In was:

Perevoznik-E. Bilek, Balatonszepl a:-:

1 9 7 1 , the continu ation

11 Q-B2 Q-R4 12 R-Q1 KR-N 1 (N-N 3 is probably better) 13

B-Q2 N-N3 achieved)

16

14 P-N 3 N-K 1 N-QR4 Q-N4

15 17

P-K4 P-BS NxN PxP

(Equality has been 18

QxP QxN

19

QR-N 1 Q-B2 2 0 Q-K3 Q-B7 2 1 QR-B 1 Q X RP 2 2 N-Q4 Q-R5 , drawn.

11 R-N 1 Q-R4 (see also Laver-Benko in the 12 P-QR3 KR-N 1 13 B-Q2 was seen in Akvist-Eskols-

The continuation Appendix)

60

THE

BENKO

GAMBIT

son, Varna 1 9 6 9 : 1 3 . . . N-N5 1 4 Q-B2 N / 5-K4 1 5 N X N N X N 1 6 P-QN4? ( A premature attempt to free himself from the pres­ sure) 16

. . . P X P 17 P X P Q-B2 1 8 KR-B 1 N-B5 1 9 R-N 3 N X B

2 0 Q X N Q-B5 (Bl ack stands a l ittle better) 2 1 R l l -N 1 R-QB 1 2 2 N - K 4 Q X KP 2 3 Q-B4? B - Q 6 24 B - B 3 Q - B 7 2 5 B-Q 1 Q-B5 , White resigned. The game Baren dregt-B ukal , Zagreb 1 9 7 1 , continued:

1 3 N-Q2?!

R-R2 14 N-R2 Q-B2 1 5 P-QN4 ? ! B-N4 1 6 N-QB3 P X P 1 7 N X B R X N 1 8 R X P R X R 1 9 P X R R-R7 (The chances are about equal) 20 P-K4 N-N3 21 N-N3 N-R5 2 2 B-Q2 Q-B5 (Now B l ack has the edge) 2 3 N-B 1 R-B7 24 R-K 1 N-N 5 ! 25 B-B 1 Q-Q5 2 6 N-Q3 R-R7 2 7 Q-B 1 P-R4 2 8 B-K3 Q x KP 29 Q-B8ch K-R2 30 N-B 1 R-B7 3 1 Q-Q7 N x B 3 2 Q x N Q-B 6 ? (Black wins easily with 3 2 . . . R X P ! ) _ 3 3 N -Q3 , a n d White i s winning material and the game. The early N-Q2 by White was tried in a slightly different posi­ tion (White plays Q-B2 and R-K 1 instead o f R-N 1 and P-QR3) in a game Haigh-Vogt, Skopje Olympiad 1 97 2 : 11 Q-B2 Q-R4 12 R-K 1 KR-N 1 1 3 N-Q2? N-K 1 1 4 N-N 3 Q-N 5 1 5 N - Q 2 N - B 2 1 6

P-QR3

Q-R4

17

R-N 1

P - B 5 (Black h a s a strong initiative)

18

N/ 2-K4 (If 1 8 P-K4 N-B4-Q 6 . See the following example for more o n the dangers of P-K4) 1 8

. . .

N x P 19 B-Q2 N x N 2 0 N x N

R-R2 and Black won. An important point was demonstrated i n Lorinczi-Browne, Siegen 1 9 7 0 : after

11 R-K1 Q-R4 12 Q-B2 N-N 3 1 3 P-K4 KN -Q2 1 4

B-N 5 KR-K 1

1 5 N-Q2 , White was al ready i n trouble. The move

P-K4 m akes sense only if P-K5 can be successfully achieve d ; other­ wise the move only weakens White's Q 3 . The text move prevents Black's plan of N-B5 followed by N x NP, a possibil ity which is a strong motivating force behind B lack's earlier Q-R4. If instead 1 5 B-Q2, the answer 1 5

. . .

N-B5 is embarrassing because White's

KR prevents the Bishop from retreating to K l . Also, o n 1 5 QR-Q 1 N-B5 1 6 B-QB 1 , the unpleasant reply is 1 6 . . . N-R6 . The actu al game continued : 1 5 . . . N-K4 16 N-N 3 B-Q6 1 7 Q-B 1 Q-N5 1 8 B-R6 B-R l

1 9 N-Q2 N / 4-B 5 20 N X N N X N 2 1 P-QR4 Q x N P

(Black h a s an edge) 22 Q X Q N X Q 2 3 KR-QB l KR-N 1 2 4 P-R5 R-N6 2 5 B-Q2 N-B5 26 B-K 1 N-R6 2 7 B-B 1 B X B 28 K X B R X P 2 9 R-R2 R-R2 3 0 P-B3 N-N4 3 1

R X R N x R 3 2 K-N 2 P-B4,

Wh ite resigned. Another example of a poorly timed P-K4 was M ista-Spiridonov, Cienfuegos

1 972, which reached this variation by transposition : 61

THE

BENKO

GAMBIT

1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-KB3 B-N2 4 P-KN3 0-0 5 B-N2 P-B4 6 P-QS P-Q3 7 0-0 P-QN4!? 8 P x P P-QR3 9 P X P B X P 1 0 R-K1 QN-Q2 11 P-K4?! N-NS 1 2 N-B3 Q-R4 1 3 P-KR3

N/ 5-K4 1 4 N x N N X N 15 B-B 1 KR-N 1 (Black is definitely better already) 1 6 B X B Q x B 17 K-N2? (R-K2) 1 7 . . . N-Q6 (The pun­ ishment for P-K4) 1 8 R-K2 N x N P ! 1 9 B X N R X B 20 R X R B X N 2 1 R/ 1 -N 1 B X R 22 R X B Q-B5 23 Q-K2 Q x Q 24 R X Q R-R6. Black stands much better and won the endgame on the 48th move. E2

The advantage of 1 1 . . . Q-N 3 is the pressure Black exerts im­ mediately on the QN file. Also, the Queen is less exposed to danger on N3 than on R4. Its disadvantage lies in the fact that the Queen temporarily obstructs the maneuver N-N 3-B5 . The game Vukic-Djukic, Sarajevo 1973, continued (from the last diagram): 11 Q-B2 Q-N3 12 R-Q l (see also the game Vukic-Benko in the Appendix) 1 2 . . . KR-N 1 1 3 R-N 1 N-N 5 ! ? 14 B-Q2 N/ 5-K4 1 5 N X N N x N 1 6 P-N3 B-QB 1 1 7 P-KR3 Q-R3 1 8 P-B4 N-Q2 1 9 P-K3 Q-R6 ( 1 9 . . . B X N?! 20 Q X B Q x P? 21 R-R 1 , etc.) 20 N-R4 N-N 3 ? (Trading the bad White Knight. Better is 20 . . . N-B3 to be followed by B-Q2, or if 21 B-QB 1 Q-N5 22 P-R3 B-B4, etc.) 2 1 B-QB 1 Q-N5 22 P-R3 Q-R4 23 B-Q2 Q-R3 24 N x N R X N 25 P-QR4 (White has the better chances) 25 . . . B-B4 26 P-K4 B-Q5ch 27 K-R2 B X RP?? (The unusual rate of play in this tournament was one hour for the entire game, which partly explains this desperate try. After 27 . . . B-Q2, White still has a long way to go) 28 K X B Q-K7 2 9 B-KB 1 Q-B6 30 B-B4, and Black resigned. Instead of 12 R-Q 1 , the game Jakobsen-Westerinen, Linkoping 1 969, continued 12 R-N l?, which is a mistake because it allows Black to attack the QR pawn and QP effectively . However, the immediate 12 . . . B-B5 does not work because of 13 N-Q2 (see Szily­ Foldy below). The continuation was 12 . . . Q-N2! (Now 1 3 R-Q 1 would be met by 1 3 . . . B-B5 1 4 N-Q2 B X RP 1 5 R-R 1 B X P) 1 3 B-N5 N-N3 1 4 B X N (Here 1 4 KR-Q 1 would again b e bad because of 1 4 . . . B-B5) 1 4 . . . B X B 15 KR-Q 1 B X N ( 1 5 . . . B-B5? 1 6 P-N3 B X QP 1 7 N X B N X N 1 8 N-N5 ! P-K3 1 9 N-K4 B-K2 20 N-B3, and White is obviously better) 1 6 P X B B-B5 17 R-N 2 Q-R3 1 8 Q-K4 Q-R5 19 R/ 1 -N 1 B X QP ! 20 Q x P N-B5, with a clear advantage for Black. 62

THE

BENKO

GAMBIT

In the game between Korchnoi (White) and Quinteros at the Leningrad Interzonal 1 973, the position after Black's 1 1 th move (Q-N3) was reached by transposition. Here, too, the mistake 1 2 R-N l was played, but Black did not take the opportunity to prepare for . . . B-B5 by 1 2 . . . Q-N2 ! Instead, Quinteros played 12 . . . KR-Nl? and the game continued 1 3 P-N 3 N-K 1 (N-N5 looks more active; if then 14 B-N2 P-B5 15 P X P B X P 1 6 B-QR1 Q-R3, etc.) 14 N-Q2 Q-R4 (Black's Queen has taken two moves to reach this square. Since he could have played Q-R4 in one move, this is something to think about when considering how best to develop this important piece) 1 5 B-N2 P-B 5 ! ? 1 6 KR-B 1 (Of course not 1 6 P X P-or 1 6 N x P-because of 1 6 . . . R X B and 1 7 . . . Q X N ) 1 6 . . . N-K4 1 7 B-QR 1 R-N5 1 8 P X P N x P 1 9 R X R Q X R 2 0 N x N B X N 2 1 B-R3 ! (White now has the edge. This original maneuver is aimed at improving the King Bishop's prospects, e.g. at QB6) 2 1 . . . N-B2 2 2 P-K4 B-Q5 ?! (The Bishop i s unprotected here and therefore, as all the textbooks point out, invites a combination. Better was 22 . . . N-R3 23 B-Q7 N-B4 24 B-B6 B-Q6, etc.) 23 B-Q7 N-R3 24 B-B6 R-R2 (Certainly an awkward place for the Rook, but there is nothing better) 25 R-N 1 Q-R4 (Q-R6 was better, but White retains his advantage after 26 N-N5 B X N 27 B/ 1 X B B-Q6 28 Q-N3, etc.) 2 6 N-N 5 ! B X B (If 2 6 . . . B X N 2 7 B / 1 X B R-B2 28 B-B3 Q-R5 2 9 Q x Q B X Q 30 B-R5 , etc.) 27 Q X B, Black resigned. In a game between Dr. Szily and Dr. Foldy in Budapest 1973, the mistake 12 . . . B-BS? was played, with the continuation 13 N-Q2 ! B-R3 (Admitting his mistake and losing two tempi in the process-1 3 . . . B X RP leaves the Bishop trapped after 14 R-R 1 ) 1 4 P-N 3 KR-N 1 1 5 B-N2 N-K 1 1 6 N-R4?! Q-N 5 1 7 KR-K 1 N-B2 18 B X B K x B 19 N-QB3 N-K4 20 N-B3 (20 P-KR3 looks more promising, with the idea P-B4, even after 20 . . . P-N4?!) 20 . . . N x Nch 2 1 P x N ? ! Q-Q5 22 QR-Q 1 Q-B3 23 P-B4 N-N4! 24 N X N B x N , drawn. I t i s hard for White to make progress, despite his pawn plus. After 1 1 R-N l Q-N3, Johansson-Westerinen, Dresden 1969, con­ tinued 1 2 B-Q2?, but this is an error which returns the pawn at once and leaves Black with a positional advantage: 1 2 . . . B-B5 1 3 P-N3 B X QP 14 N X B N X N 15 N-N5 P-K3. Instead of 1 1 R-N 1 , there has been some experience with 11 R-Kl Q-N3.

63

THE

BENKO G A M B IT

12 P-K4 was played in Karlsson-Akvist, Ebso 1 970 (also see Aspler-Benko in the Appendix). The continuation was: 1 2 . . . N-N5 ! 1 3 P-KR3 N/ 5-K4 1 4 N-QR4 Q-N2 1 5 R-K3 N X Nch 1 6 R X N B-N4 1 7 N-B3 B-B5 1 8 B-B 1 B x B 1 9 Q x B KR-N 1 2 0 N-Q 1 B-Q5 2 1 P-QR4 N-K4 22 R/ 3-R3 P-B4 23 Q-K2 P X P 24 Q X P Q-Q2 2 5 K-N2 R-N 5 2 6 Q-K2 Q-N2, with Black better. As we can see, when White plays P-K4 prematurely Black's 1 2 . . . N-N 5 ! threatening the KB7 and Q 6 squares i s advantageous. Therefore, White has tried to prepare P-K4 by first preventing . . . N-N5. The game Charpentier-Hook, Panama 1 970, continued 12 P-KR3 KR-N l 13 R-N 1 N-K 1 14 Q-B2 N-B2 15 P-K4 N-N4 1 6 N x N B XN 1 7 P-R3 .

On 1 7 P-N3 Q-R3 i s very strong. The game Thorbergsson­ Gheorgh iu, Reykj avik 1 972, continued: 17 R-K3 (instead of P-R3) R X P 18 R-R3 R X R ( 1 8 . . . R-R 1 ! 19 Q-N3 is equal) 1 9 P X R Q-R3 (Now Black is clearly better) 2 0 N-Q2 B-Q5 2 1 B-B l B X B 2 2 R X Rch N X R 2 3 N x B N-Q2 24 B-N2? N-K4 ! 2 5 K-N 2 N-Q6 64

THE

BENKO

GAMBIT

26 B-B3 Q-B5 and White resigned. Continuing the Charpentier-Hook game (after 1 7 P-R3) 1 7 . . . B-R5 1 8 Q-B4? B-N6 1 9 Q-K2 R-R5 20 N-Q2 N-K4 2 1 N X B Q X N 22 Q-Q 1 Q X Q 2 3 R X Q R X RP (With a clear advantage for Black) 24 B-B 1 R-R5 25 P-B4 N-Q2 26 B-Q3 R/ 5-N5 and Black won. The game Farago-Bilek, Budapest 1 964, continued (from the diagram after 1 1 . . . Q-N 3 above): 12 Q-B2 KR-N l 13 B-R3!? (An interesting attempt to prevent . . . N-N5 while activating the Bishop) 1 3 . . . Q-N5 (P-R4!?) 14 P-K4 B-B5 15 P-K5 P X P 1 6 N X P N X N 1 7 R X N R-N2? ! (R-R2 i s better) 1 8 B-N2 N-K l ! 1 9 R-K 1 N-Q3 2 0 P-QR3 Q-N 3 ? (Q-N6 i s stronger) 2 1 N-K4 ! with an edge for White. The game Bokor-Bilek, Budapest 1 965, varied with 13 P-KR3 N-K 1 1 4 P-K4 Q-N 5 ? ! (N-B2) 1 5 N-Q 1 N-K4 1 6 B-Q2 Q-B5 ! 1 7 B-B3 N-Q6? (Black could have obtained the slightly better chances with 1 7 . . . Q-Q6 ! 1 8 Q x Q N x Q) 1 8 R-K3, and it is White who has a slight edge.

E3

The posting of the Black Queen on QB2 is the most circumspect of the three possibilities under discussion. From her safe haven she will direct the counterplay on the Queenside. This development has neither the disadvantage of being exposed to harassment nor that of being in the way of the maneuver N-N3-B5, but being in a less active position, the hard tooth-and-nail fighting on the Queenside is somewhat postponed. ( 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR 3 5 P x P B X P 6 N-QB3 P-Q3 7 N-B3 P-N 3 8 P-KN 3 B-KN2 9 B-N2 0-0 1 0 0-0 QN-Q2)11 Q-B2 Q-B2 12 R-Ql was played in Korchnoi-Gurgenidze, USSR Championship 1 957. The game continued: 12 . . . KR-N 1 1 3 P-KR3 N-N3 14 N-Q2 KN-Q2 1 5 R-N 1 N-QB 1 ? ! (The start of a plan to win a pawn which is only half successful) 16 P-N 3 Q-R4? (B X N ! ?) 17 B-N2 P-B5 1 8 B-R l ! ( 1 8 P X P? R X B) 1 8 . . . P X P 1 9 N x P (White i s definitely better) 1 9 . . . Q-B2 2 0 N-Q4 R x R 2 1 Q X R R-N 1 22 Q-B 1 B X N 2 3 R X B B X P 2 4 K-R2 (24 N x B? Q x Qch 25 N x Q R-N8) 24 . . . B-R3 25 Q-R6 N-B l 26 R-KR4 65

THE

BENKO

GAMBIT

(White has a winning attack) 26 . . . Q-N3 27 N-K4 P-B3 28 B X P P X B 2 9 N x Pch K-B2 30 R-KB4, Black resigned. Instead of 1 2 R-Q 1 , the game Quinteros-Szabo, Vinkovci 1 970, continued: 12 R-N l?! N-N 3 (Q-N2! is more logical-see J akobsen­ Westerinen in Variation E2) 13 N-Q2 QR-K l ! ? (Black intends a central break a Ia the Volga Gambit instead of Queenside play) 14 R-Q 1 P-K3 1S P X P P X P 16 N/ 2-K4 KN-Q4 1 7 P-QR3 N x N 1 8 N x N P-Q4 1 9 P-K3 R-K2 20 N-R4 N X N 2 1 Q x N B-K7 22 R-K 1 B-Q6 23 R-R 1 R-N 1 (P-BS). Black has a slight edge but he later lost. After 11 R-Kl Q-B2, the fol10wing position is reached:

12 P-KR3 N-N3 1 3 P-K4 KN-Q2 1 4 Q-B2 N-BS ! , drawn (Szabo­ Toran, Kapfenberg 1 970). 12 Q-B2 was played in Tritaliatis-Toran, Siegen 1 970, which continued: 12 . . . KR-N l 1 3 B-B 1 ? Q-B l 14 P-K4 (P-KR3, to prevent . . . N-NS, is more exact) 14 . . . N-NS 1 S B X B Q X B 1 6 B-B4 N / S -K4 1 7 N X N N X N 1 8 B X N B X B (Black's position is preferable) 19 R-K2? R X P 20 Q X R B X N 21 Q X B Q X R and Black won, illustrating the fact that a mistake by White in this gambit can be very costly! In Spassky-Szabo, Goteborg 1 97 1 , the then world champion tried 12 B-Q2?! The game continued : 12 . . . N-N 3 1 3 B-B4 (Provoca­ tive !) 1 3 . . . KR-N 1 (A routine move. 1 3 . . . N-BS is stronger) 1 4 P-N3 ! N-NS 1 S B-Q2 N-B 1 ? ! 1 6 P-KR3 N-K4 1 7 N x N B X N 1 8 R-N 1 N-R2 1 9 Q-B2 P-B S ? ! 2 0 KR-QB l R-QB 1 2 1 Q-Q l ! P X P 2 2 P X P Q-N2 2 3 N-R2 ! N-N4 2 4 N-N4 N-B6 2 S B X N B X B 2 6 N-B6. The position of this Knight finally gives White some ad­ vantage, but the game was later drawn.

66

THE

BENKO

GAMBIT

A sounder method for Black, which in this instance was ad­ mittedly favored by some cooperation on White's part, was seen in Vincenti-Andrews, Skopje Olympiad 1 972: 11 Q-B2 Q-B2 12 R-K1 N-NJ 13 N-Q2 KR-N 1 14 P-K4? (Since Black has the reply . . . N-N5, as discussed in section B, White should prepare P-K4 by first playing 1 4 P-KR3) 14 . . . N-N 5 1 5 N-B3 N-B5 ! (With a definite advantage for Black. White's P-K4 is a critical thrust and must be carefully prepared) 16 P-KR3 N/ N5-K4 17 N x N B X N ! ? (N X N is safe) 1 8 B-B 1 Q-Q2 1 9 B X N B X B 20 K-N2 P-B4 2 1 P-B3 P x P 2 2 P x P P-K3 (The opening o f the position i s to Black's advantage) 23 P-R3 Q-QN2 24 P X P B X KP 25 K-R2 Q-K2 26 Q-N2 Q-R5 27 N-Q5 B X N 28 P X B Q-Q5 (28 . . . R-N6? 29 R X B) 29 R-K2 R-N6 30 B-B4(?) B X B 3 1 P X B Q x Pch 32 K-R 1 R-KB 1 (Black has a winning attack) 3 3 QR-K 1 R-KB6 34 R-K8 R-B7 35 Q-K4 R-R7ch, and White resigned in view of mate in one. Instead of developing h is Queen, Black can first play 1 1 . . . N-N 3 , but this usually means only a transposition into the normal lines after 1 1 . . . Q-B2. After 11 R-K1 N-NJ, de Jonghe-Rosenlund, Viborg 1 970, con­ tinued: 12 N-Q2 Q-B2 13 Q-B2 B-N2?! 14 P-K4 B-QR3 (Black wanted to force White to play P-K4 by th is Bishop maneuver, but it is questionable whether that objective is worth the loss of two tempi) 1 5 P-B4? N-N5 16 N-B 1 (N-B3) B-Q5ch (Black stands better) 17 N-K3 N-B5 1 8 N-Q 1 Q-R4 19 R-K2 N-R6 ! 20 Q-Q2 N x N 21 Q x Q? (The question mark indicates a blunder, but the alternative also loses: 2 1 R X N Q X Q 22 B X Q N-B7, etc.) 2 1 . . . N-B5ch 22 B-K3 N x Q 23 B X B B X R, White resigned. A game Westerman-Lundwall, Diesen 1 970, continued : 12 P-K4 KN-Q2 1 3 Q-B2 Q-B2 ( 1 3 . . . N-B5 ! ? looks more promising, e.g. 14 P-N 3 ? Q-R4 ! as in Mrs. Honfi-E. Bilek, Budapest 1 970, and if 15 P-K5 N/ 5 X P with advantage to Black) 1 4 P-N 3 P-B5 ? 15 B-N2 KR-B 1 (P X P) 16 P-QN4 N-K4 17 N x N B X N 1 8 P-QR4 and White won. A recent game Mukhin-Georgadze, USSR 1 972, took an entirely different course: 11 R-N 1 N-K1?! 12 B-Q2 N-N 3 1 3 P-N 3 N-B2 14 N-KR4 B-N2 (P-B4) 15 P-K4 B X N?! ( 1 5 . . . B-QR3 16 R-K 1 P-B5 is u nclear, but offered better chances than the text) 1 6 B X B R X P 1 7 Q-B 1 N-N4 1 8 B-N2 N-Q2 (The chances are about equal) 1 9 Q-R6 N-K4 20 P-R3 (If 20 P-B4 R X B 21 R X R N-N5 22 Q-N 5 N-K6 threatening . . . P-B3) 20 . . . P-B3 2 1 Q-K3 N-R6 22 B X N 67

THE

BEN KO

GAMBIT

R X B. Black has a little the better of it because of White's weak QNP, but the game was eventually drawn.

E4 1 P-Q4 N-K83 2 P-Q84 P-84 3 P-QS P-QN4 4 P x P P-QR3 5 P x P 8 x P 6 N-Q83 P-Q3 7 P-KN3 P-N3 8 8-N2 8-KN2 9 N-R3?! QN-Q2 10 0-0 0-0

Of the many attempts to improve White's chances, one of the most frequently adopted is the development of the King Kn ight to R3. The idea is good insofar as the Knight does not interfere with the King B ishop's protection of the QP and, in fact, may even help in this respect by going to KB4. Nevertheless, the effectiveness of Black's Queenside play is in no way hindered by White's N-R3 and the Knight may find itself too far away for too long from the main scene of battle, as has been shown by practical experience. The immediate 1 1 N-84?! was tried in Yanofsky-Gheorghiu, Siegen 1 970. After 1 1 . . . Q-B2 12 Q-B2 KR-N 1 1 3 R-N 1 , Black took advantage of the position of White's KN with 1 3 . . . P-B5 ! With the Knight on KB4, Black does not have to worry about releasing his Q5 square since White cannot play N-Q4-B6. This is another drawback to the development of the Knight to KR3 . The game continued: 14 N-R3? (Now White decides to relocate his Knight via KN5, but his position cannot survive the wasting of so many tempi. He should have tried 14 B-Q2 N-B4 15 KR-B 1 ) 1 4 . . . N-B4 1 5 N-N5 B-QB l ! (The Bishop threatens to find new prospects on KB4. The point is that if White tries to prevent this by P-K4, 68

THE

BENKO

GAMBIT

his Q3 will be laid bare for exploitation by Black's Knight) 1 6 P-B3? N-R5 ! (Black now stands much better) 1 7 N x N Q-R2ch 1 8 K-R 1 Q X N 19 P-N 3 Q x RP 20 R-N2 P X P 21 R X Q P x Q (P X R i s flashy but unnecessary) 22 R X P P-RJ 2 3 N-R3 N X P and Black won. I n the game between Antoshi n and M ihailchisin at Sarajevo 1 970, the continuation was 11 B-Q2 Q-N 3 12 R-N 1 KR-N 1 1 3 P-N 3 N-K 1 1 4 Q-B2 N-B2 1 5 P-R4 N-B3 (N-K4 may be better) 1 6 N-B4 R-R2 (White has a little the better of it) 17 P-QN4!? P X P 1 8 B-K3 Q-N2 1 9 N-N5 B X N 2 0 B X R Q X B 2 1 P X B R x P 22 N-Q3 N/ 3 X P 23 P-K4 N-B6 24 R X P R X R 25 N x R N/ 2-N4, and White's slight material advantage was not enough to win. After 1 1 R-K1 Q-N3, an inferior l ine was chosen in Golombek­ Sefc, Trencanske Teplice 1 949( !): 12 K-R1? N-K4 13 Q-B2 KR-NJ 1 4 P-B4 N-B5 1 5 P-N3 Q-N5 ! ! (Black will now obtain an irresist­ ible initiative) 1 6 P X N N-K5 1 7 Q X N B X N 1 8 B-Q2 B X B 1 9 KR-Q 1 B-B6 20 QR-N 1 Q X P 2 1 Q X P R X R 2 2 R X R Q X RP 23 B-K4 Q-R5 24 B-B3 (If 24 B-N2 R-K 1 25 Q X P Q-R7 26 R-N 8 Q-R8ch, etc.) 24 . . . R-K 1 25 Q X P B X P 26 B-N2 Q-B7 27 R-N 1 B-Q5 (Black has a winning position) 28 Q-Q7 R-KB 1 29 P-Q6 B X R 3 0 N x B P-B5 3 1 B-B6 P-B6 3 2 P-R 3 Q-Q8 3 3 Q-B7 P-B7, White resigned. Another try was seen in Eroeybek-Schoenberg, Skopje Olympiad 1 972: 12 Q-B2 KR-N 1 13 R-N 1 N-K4! (White has improved by playing R-N 1 instead of Golombek's inferior K-R l . But his funda­ mental problems remain) 1 4 N-KN5 B-N2 !? 1 5 R-Q 1 B-QB 1 1 6 N/ 5-K4 (P-N3) B-B4 1 7 N X Nch B x N 1 8 B-K4 B-R6 1 9 P-N 3 N-N5 20 N-R4 Q-R2 (Black's freer pos ition gives him the edge) 2 1 P-K3 N-K4 22 P-B4 N-N5 (White has been given the "free" moves P-K3 and P-B4, but that price was too high-Black has a clear advantage because of White's weaknesses) 23 R-K 1 B-N2 24 B-KN2 B X B 25 K X B Q-N 2 ! 26 Q-B4 R x N (Winning) 27 Q X R Q x Pch 2 8 P-K4 Q-Q6 2 9 Q-B4 (29 B-R3 Q-Q7ch) Q x Q 3 0 P X Q R X R, and White resigned. After 11 Q-B2 Q-B2 (For 1 1 . . . Q-R4, see Denker-Benko in the Appendix), the game Sloth-Larsen, Denmark 1 97 1 , continued : 1 2 R-Q 1 KR-N 1 1 3 R-N 1 N-K4 1 4 N-KN5 B-QB 1 1 5 N/ 5-K4 N x N 1 6 B X N B-R6 17 P-N3 N-N 5 1 8 P-R4 P-B5 ! (Tactics dove­ tailing with strategy-on 1 9 P-N4 R X NP ! 20 R X R Q-B4 threaten­ ing mate and the Rook) 19 N-N5 Q-N3 20 P-K3 P X P 21 R X P 69

THE BENKO GAMBIT R-QB l (2 1 . . . R X P? 22 N-Q4) 22 Q-N l Q-B4! 23 R-N4 (23 B-R3 Q-B5 24 R-N4 Q-K7 and White cannot properly defend his KB2) 23 . . . N-B3 24 B-R3 N X B 25 Q x N ( RX N is slightly better) 25 . . . Q-B7 (With an edge for Black) 26 Q X Q R X Q 27 R/ 1 -N l P-R4 2 8 B-N2 R/ 1 -QB l 29 N-Q4 B X N 3 0 P X B (30 R X B R X B) R-K7 (Black is clearly better. The endgame is quite interesting) 3 1 B-R3 R-R7 ! 3 2 B-N2 K-N2 3 3 R-K l R-B7 3 4 R-N l R-Q7 3 5 P-R5 B-B4 36 B-B3 R X BP (And now Black i s winning) 37 R/ 1 -N2 QR X R 38 R X R R-B6 39 B-K l B-R6 40 B-B2 R-R6 41 B-K l P-B4 42 B-N4 R-R8ch 43 K-B2 P-N4 44 K-K3 K-B3 45 R-N3 R-R7 46 R-R3.

46 . . . R X P ! ! 47 R-R l (If 47 P-R6 B-B8 48 P-R7 B-B5 49 P-R8=Q R-K7ch 50 K-B3 P-N5ch and White is mated by 51 . . . R-K5 ; or if 49 R-Q3 R-R7 50 B-R3 B X P wins) 47 . . . B-N7 48 R-R2 P-R5 49 P x P P-B5ch, White resigned. Black made excellent use of his chances, even in positions that seemed drawish. An in­ structive game. An attempt to improve the N-KR3 idea lies in the development of White's Queen Bishop to QN2 in order to oppose Black's strong King Bishop and neutralize its pressure. An example of this idea is Benko-Berry in the Appendix. A similar attempt was made in Evans-DeFotis, U.S. Champion­ ship 1 972, but by a different order of moves: 1 P-Q4 N-KB3 2 P­ QB4 P-84 3 P-QS P-QN4 4 P X P P-QR3 5 P X P B X P 6 P-KN3 P-Q3 7 B-N2 QN-Q2 8 N-KR3 P-KN3 9 P-N3. In order to ef­

fectuate this move White has had to postpone the development of his Queen Knight. The game continued : 9 . . B-KN2 10 B-N2 .

0-0 1 1 0-0

70

THE

BENKO

GAMBIT

1 1 . . . Q-N1?! (The natural 1 1 . . . Q-N 3 is better) 1 2 Q-Q2 Q-N5 1 3 B-QB3 N-K5 14 B X N Q X B / 5 1 5 B X B K x B 1 6 N-B3 Q-Q5 17 KR-Q 1 Q x Q 1 8 R X Q (White has a slight edge) KR-N 1 1 9 R-N 1 R-N 3 20 R/ 2-N2 B-B 1 2 1 N-B4 N-K4 22 N-Q3 ? ! N X N 2 3 P X N B-R3. The game i s equal and was eventually drawn. White's weak pawns prevent him from getting an active game. More games are needed to examine this idea. So far, I know of only one other, G . Garcia-Peev, Cuba 1973, which, instead of 1 1 . . . Q-N 1 , continued 11 . . . Q-N3 1 2 Q-Q2 KR-N 1 1 3 B-QB3 (A measure against the possibility of . . . P-B5) R-R2? (This looks like a loss of time. Why not 1 3 . . . N-K 1 right away?) 14 R-K 1 N-K 1 15 N-R3 B X B 16 Q X B Q-N 5 17 Q-N2 N-N 3 ( 1 7 . . . P-B5?!) 18 N-B2 Q-R4 19 QR-N 1 (White has a slight edge) B-B 1 20 P-QN4 ! ? N-R5 (20 . . . Q X R P ! ?) 2 1 Q-R3 P X P 22 R X P R X R 23 N X R R-N2 24 N-Q3 . White still has the slightly better chances ; he eventually won.

ES

Some Hungarian masters have come up with a new idea for White: after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4 4 P x P P-QR3 5 P x P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 P-KN3 B-KN2, the new idea is 9 B-R3!?, which aims to prevent Black's N-N5 and provide more activity for the King Bishop. After 9 . . . 0-0 10 0-0, there are three examples:

71

THE

BENKO GAMBIT

Novarovsky-Georgadze, Tbilisi 1 97 1 , cont inued 10 . . . Q-R4?! (This does not work out too well. Black should probably postpone the development of the Queen) 1 1 R-K 1 QN-Q2 12 P-K4 N-N3 13 P-K5 ! N-K 1 (If 13 . . . P X P 14 N X P QN x P? 15 N x N N x N 1 6 N-B6 and wins) 1 4 B-B4 N-B5 1 5 Q-B 1 Q-B2 1 6 P X P P X P 1 7 B-R6 N-B3 1 8 B x B K X B 1 9 P-N 3 N-N 3 20 Q-N 2. White has a little the better of it, although the game was later drawn. In F. Portisch-Korpas, Budapest 1 972, the continuation was 10 . . . QN-Q2 1 1 R-K1 N-N 3 (Here 1 1 . . . P-R4!?, preparing . . . N-N5, has to be taken into consideration) 12 B-B4 B-B 1 ? (Better is 12 . . . N-B5 13 P-N 3 N-KR4, or 12 . . . KN-Q2, preparing . . . N-B5) 1 3 B-N 2 N-R4 (N-B5) 1 4 B-B 1 B-Q2 1 5 P-K4 N-R5 1 6 P-K5, with a clearly superior position for White, which h e later won. Another example was L. Szabo-Vasiukov, Wijk aan Zee 1 97 3 , which transposed t o the previous game b y a minor change i n move order: 9 . . . QN-Q2 (instead of 9 . . . 0-0) 10 0-0 N-N3 1 1 R-K 1 0-0, and now the continuation was 12 B-N2?! Q-Q2 1 3 N-Q2 KR-N 1 14 Q-B2 B-N2 (P-B5) 15 P-K4 N-KN5 16 P-KR3 N-K4 1 7 P-B4 N/ 4-B5 18 N x N N x N 19 P-N3 B-Q5ch 20 K-R2 N-R6 (Black has an equal game) 21 Q-Q2 B-B 1 22 B-N2 Q-N2 23 N-Q 1 Q-N5 24 B-QB3 B X B 25 N X B N-N4 26 R-K3 R-R6 (And now Black stands better) 27 R-Q3 N x N 28 Q x N B-R3 29 Q x Q R X Q (29 . . . P X Q?! 3 0 R-Q4) 3 0 R/ 3-Q 1 P-R4 3 1 P-R4 K-N2 32 B-B3 K-B3 3 3 K-N2 P-B5 (The game is equal after the release of tension and the liquidation on the Queenside) 34 KR-N 1 P-K3 35 QP X P P X KP 36 P X P R X R 37 R X R R X Pch 3 8 K-N 1 B X P 39 R-N6 R-R8ch 40 K-N2 R-R7ch, drawn. 72

THE

B EN KO

GAMBIT

E6

Other attempts to get something more out of the White position have of course been made. For instance, after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P x P P-N3 6 N-QB3 B x P 7 P-KN3 P-Q3 8 B-N2 B-KN2, the game Barlay-Benko (see the game in the Appendix) continued 9 Q-R4ch?! QN-Q2 10 Q-R4,

but without success for White. A dubious idea after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P­ QN4 4 P x P P-QR3 5 P X P B X P 6 P-KN3 P-N3 7 B-N2 P-Q3 8 N-QB3 B-KN2 was tried in Jones-Malagon, Skopje Olympiad 1 972: 9 P-K4?! 0-0 10 KN-K2 QN-Q2 1 1 0-0 Q-N 3 (White's

e arly P-K4 and the resulting weakness of his Q3 should be exploited by 1 1 . . . N-K4) 12 Q-B2 KR-N 1 1 3 R-N 1 N-K4 14 R-Q 1 R-R2 1 5 P-KR3 Q-R4 16 P-B4 N-BS 17 B-B 1 N-Q2 1 8 K-R2 R/ 2-N2 19 P-N3 N-R6 20 B X N Q X B. Black stands very well and eventu­ ally won. If Black tries to avoid the main variation for some reason, he cannot gain a satisfactory position. For instance, in a game Meleghegyi-Dr. Sallay, Budapest 1 972, Black left his QP unmoved for too long, which is rather dangerous when White fianchettoes his King Bishop : 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P x P B X P 6 N-QB3 P-N3 7 N-B3 B-KN2 8 P-KN3 0-0?!

Black's pun ishment should serve as a lesson : 9 P-Q6 ! N-B3 1 0 B-N2 P X P 1 1 Q X P R-K 1 ? ! 1 2 Q X BP N-K5 1 3 N X N R X N 1 4 0-0! B X KP 1 5 R- K 1 B X N 1 6 R X R B X R 1 7 B X B, with an obvious superiority for White. This is surely not the kind of position Black hopes to obtain when he plays the Benko Gambit ! 73

THE

BENKO

GAMBIT

F Transpositions to the Gambit There are many ways to transpose into the Benko Gambit from several other openings. Even when Black has the opportunity to play the gambit, he can postpone entering it until later, when White may make a move which does not fit any anti-gambit system or which may allow Black to transpose into an inferior variation for White. Or perhaps White may u nwittingly allow the loss of a tempo whereupon the Black pawn sacrifice may be an excellent choice either for counterattacking possibilities or to free Black's position and avoid a possibly long defense. There are times, of course, when Black has no o p portu nity to play the gambit until later in the game. See, for example, the game Calvo-Benko in the Appendix, which opened with 1 P-K4 and at first became a Pirc Defense, then transposed to a King's Indian Defense, and finally to a Benko Gambit. A similar mixed transposition occurred in a game Hort-Gaprin­ dashvili, Goteborg 1 97 1 : 1 P-K4 P-KN3 2 P-Q4 B-N2 3 P-QB4 P-QB4 4 P-Q5 P-QN4!? 5 P x P P-Q3 6 N-QB3 P-QR3 7 N-B3

(Approximating a position in the Deferred Variation) 7 . . . N-B3 8 P-QR4!? 0-0 9 B-KB4 Q-R4!? 1 0 N-Q2 P X P 1 1 B X NP B-QR3 12 0-0 Q-N5!? 13 Q-B2 B X B 14 P X B? ! (White gains a distinct advantage on 14 N x B N-R4 15 B-N5 Q x NP 1 6 Q x Q B X Q 1 7 B X P B X R 1 8 B X R R X P 1 9 B X P) 1 4 . . . QN-Q2 1 5 P-R3 KR-N 1 1 6 R-R6 R X R 1 7 P X R Q-R4 1 8 P-K5 P X P 1 9 N-B4 R-N5??? (Too bad! After 1 9 . . . Q x P or 19 . . . Q-N5 20 B X P R-R 1 the game is even) 20 N x Q, Black resigned.

F1

It is the King's Indian Defense which yields the greatest number of transposition opportunities, which is not surprising since the entire Benoni group is an offshoot of the King's Indian. After a normal King's Indian beginning: 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 P-K4 P-Q3 5 B-K2 0-0 6 B-N5, Black might be of the opinion that this Bishop move would not fit very well into the Benko Gambit. So, adding the fact that White's King 74

THE

BENKO

GAMBIT

Bishop has already moved, Black takes the first step towards the gambit: 6 . P-84 7 P-QS. .

.

Fla

At this point, the gambit may still be postponed. Two interesting examples follow: Paulson-Soltis, World Student Team Championship 1 968, con­ tinued: 7 . P-QR3 8 P-QR4 P-R3 9 B-R4? (B-B4 is better-see the next game) 9 . . . Q-R4 10 Q-Q2 P-QN4! 1 1 BP X P P X P 1 2 B X P B-Q2 !? 1 3 R-R3 (B-Q3) 1 3 . . . Q-N5 (Black stands better) 1 4 P-B3 B X B 1 5 P X B R X R? ( 1 5 . . . QN-Q2 1 6 N-K2 N-N 3 1 7 R X R R X R 1 8 0-0 QN X P leaves White i n a n unenviable position) 1 6 P X R Q X RP 17 KN-K2 QN-Q2 1 8 0-0 R-R 1 and Black won the game even though he had only a small edge here. From the diagram, the game Uhlmann-Geller, Majorca 1 970, continued: 7 . P-KR3 8 B-84 P-R3 9 Q-Q2 K-R2 10 N-B3 P-QN4!? (Better late than never!) 1 1 P X P P X P 1 2 B X N P Q-N 3 ! (The threat i s 1 3 . . . N x KP and 1 3 P-QR4 i s n o defense because of 1 3 . . . N x KP anyway, and if 14 N x N Q X B !) 1 3 B-K2 Q-N 5 1 4 P-K5 N-R4 1 5 B-N 3 B-R3 1 6 B X B R X B 1 7 P x P P X P 1 8 0-0 N-Q2 1 9 QR- K 1 N X B 2 0 RP X N N-N3 (Black has an edge) 2 1 R-K2 N-B5 22 Q-Q3 KR-QR 1 23 P-N 3 Q X N 24 Q x Q B X Q 2 5 P X N R X P 2 6 R X R R X R 2 7 R-B I B-B3 2 8 K-B l P-R4 29 N-K 1 P-N4 (Now Black's advantage has grown. White has a very diffic ult position despite the even material) 30 R-B2 R-R8 3 1 K-K2 K-N 3 32 N-B3 K-B4 33 N-Q2 P-N5 34 N-B 1 B-Q5 35 N-Q2 R-KN 8 36 N-N3 B-K4 37 N-Q2 R X P 38 K-B 1 R-R7 39 K-N l .

.

.

.

75

THE

BENKO

GAMBIT

R-R6 40 N-B l P-R5 4 1 K-N 2 P X P 42 P X P K-K5 43 R-B2 P-B 3 44 R-R2 R-R l and White resigned. To the end, Gel ler demonstrated the technique of a great master. In Novi Sad 1 972, the game Deze-Benko continued: 8 B-K3 P-QN4!? 9 P X P P-Q R3 10 P-QR4!? Q-R4 1 1 B-Q2 ! P X P 12 B X NP N-R3 1 3 N-B3 N-QN 5 1 4 0-0 B-R3 1 5 R-K 1 Q-N 3 1 6 B-QB l B X B 1 7 N X B KR-N 1 1 8 R-R3? ! N-Q2 1 9 Q-N 3 Q-N 2! 20 B-Q2 N-N3 21 B X N P X B 22 Q X P N X RP 23 N/ 3-Q4? (23 R x N R X R 24 Q X R Q x N 25 Q x Q R X Q with a slight edge for Black) 2 3 . . . B X N 24 R X N B-B4 ! 25 Q-B4 R X R 26 Q x R Q X N , White re­ signed.

Flb

After 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 P-K4 P-Q3 5 B-K2 0-0 6 B-NS P-B4 7 P-QS P-QN4!? 8 P X P P-QR3, there are two major branches, differentiated by which player makes the next capture on the Queenside . White may try to utilize his QN5 as a base for his pieces, continuing with 9 P-QR4!? Q-R4 10 B-Q2 P x P, which was played in Uhlmann-Adamski, Polanica Zd roj 1 968.

That game continued: 11 B x P B-QR3 12 KN-K2 QN-Q2 1 3 0-0 B X B 1 4 N X B Q-N 3 1 5 Q-B2 KR-B 1 1 6 B-B3 P-B5 1 7 B-Q4 Q-N 1? (With his next move, a rather strange maneuver and an expensive loss of time) 18 KR-B 1 Q-N 2 19 N-R3 (Wh ite now has the initiative and the better chances) 19 . . . N-B4 20 B x N R X B 2 1 N x P QR-B 1 2 2 P-QN 3 N x KP!? 2 3 Q X N B X R 2 4 R X B Q X N P 76

TH E

BENKO

GAMBIT

25 N-K3 Q-N 7 26 R-K 1 R/ 1 -B2 27 P-R4 P-R4 28 K-B 1 R-N2 29 P-N 3 K-B 1 30 Q-B3 R-N 5 3 1 N-KB4 Q-B3 32 K-N 2 R x R P (With the disappearance of his QRP, Wh ite's chances are now only m ini­ mally better, but Black still has a difficult de fense) 33 R-QN 1 R-B 1 34 Q-K2 K-N 2 35 R-N7 R/ 1 -QR 1 36 N-B4! R/ 5-R2 37 R-N 3 R-R7 38 Q-K4 R/ 7-R3? 39 R-KB 3 Q-R8 40 Q x P R/ 1-R2 41 N-K6ch, and Black resigned. From the diagramed position, the game Kenez-Melcghegyi, cor­ respondence 1 9 7 1-72, continued : 11 N x P Q-N3 12 Q-B2 B-Q2 13 R-R3 N-R3 14 N-R3 P-K3 (The position is unclear, with an interesting battle in view) 1 5 P X P P X P 1 6 0-0 N-QN5 17 Q-N 3 ? ! (Q-N l ) 1 7 . . . N x P 1 8 BXN PXB 1 9 QXP P-Q4 (Now Black seems to have the slightly better chances because of h is two B ishops and better center control) 20 P-R5 Q-N 2 2 1 P-B3 Q R-N 1 ?! 22 P-R6 Q-N 3ch 23 K-R 1 B X N 24 P-R7? (24 P X N is much better) 24 . . . B X B ! 25 P X R=Q Q X Q/ 1 26 Q X Q R X Q 27 KR-K 1 B X NP! 28 R-K3 N-B6 29 R/ 3 X B N X R 30 R X N B-Q5, and Black has the much better game which he conducted to a successful con­ clusion. After 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 P-K4 P-Q3 5 B-K2 0-0 6 B-NS P-B4 7 P-QS P-QN4 8 P x P P-QR3,

White may continue directly with 9 P X P.

In Filip-Calvo, Havana 1 966, the continuation was: 9 . . N x RP 10 N-B3 Q-N 3 1 1 QR-N 1 P-R3 1 2 B-R4? ! (B-B4) P-N4 1 3 B-N3 N-R4 1 4 0-0 N x B 1 5 RP XN N-B2 16 B-B4 B-N5 1 7 Q-B2 KR-N 1 1 8 N-Q2 B-Q2, and White had the slightly better chances, but Black later won. In Dr. Erdy-Meleghegyi, Szolnok 1 972, the continuation was: .

77

THE BENKO GAMBIT 9 . . . Q-R4!? 10 Q-Q2 B X P 1 1 B X B N X B 12 KN-K2 KR-N1 13 0-0?! (P-B3) 1 3 . . . Q-N5 14 P-K5 P X P 15 QR-N 1 R-R2 1 6

KR-Q 1 R-Q2 1 7 N-N3 N-K 1 1 8 Q-K2 N/ l -B2 1 9 B X P R X B 20 P-Q6 R/ 2-K 1 21 P X N N x P, and Black is slightly better; he won. The game Woodhaus-Brodie, Adelaide 1 97 1 , continued : 13 P-B3 P-B5 14 B-K3 N-B4 15 0-0 KN-Q2 16 KR-Q 1 N-K4 17 QR-N 1 N-R5 (Black has a clear advantage) 1 8 Q-B2? N-Q6 1 9 N x N Q X N 2 0 P-N3 Q X RP 2 1 Q X Q R X Q 2 2 N-Q4 B X N 2 3 B X B N-B 5 ! (And wins) 2 4 R-N2 R X R 2 5 B X R R X P 26 B-Q4 P-B6 27 K-B2 P-B7 28 R-QR 1 P-B3 29 B-K3 N-Q6ch 30 K-K2 K-B2 3 1 R-KB 1 R-N8, White resigned.

F2

If we trust in statistics we must conclude that the pawn sacrifice with . . . P-QN4 in this variation of the King's Indian offers Black very good chances. But the sacrifice can be considered in other variations as well. Here are a few examples (the King's Indian with P-KN 3 for White is examined in Variation F3). After 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 P-K4 0-0 5 B-K3 P-Q3, the game Liardansky-Vasiukov, USSR Cham­ pionship 1 964, continued: 6 P-KR3 (P-B3) P-B4 7 P-QS (On 7 P x P Q-R4, but 7 N-B3 is to be considered) 7 . . . P-QN4!? 8 P X P Q-R4 9 B-Q3 P-QR3 1 0 P X P B X QRP 1 1 N-K2 Q-N5 1 2 B X B N X B 1 3 Q-B2 Q-B5 1 4 0-0 N-QN5 1 5 Q-N 1 KR-N 1 (Black's pressure on the Queenside and in the center gives him a slight pull) 16 R-Q 1 R-R4 17 R-Q2 N-Q2 18 P-R3 Q-R3 19 Q-Q 1 N-K4 20 R-B 1 N/ 5-Q6 21 R-N 1 P-B5 . White has been defending himself stubbornly and was threatening P-B4. Black's solution, P-B5, does strengthen the position of the vital Knight at Q6, but it also gives White the use of his Q4 square, through which he is able to offer sufficient resistance. The game continued: 22 N-Q4 R-N2 23 R-K2 N-B4 24 R-B2 P-R4 25 Q-K2 N/ K-Q6 26 P-KR4 N-K4 27 P-B3 N/ B-Q6 28 R-Q 1 B-B3 29 P-KN 3 Q-R 1 30 K-N2 Q-N 1 3 1 R/ 1-Q2 R-R 3 32 B-N 1 K-R2 33 B-K3 Q-K 1 34 B-N 1 R/ 3-N3 35 N-Q 1 Q-R5 36 N-B3 Q-Q2 37 N-Q 1 Q-R5 38 N-B3, drawn . After 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-Q3 4 N-QB3 78

THE BENKO GAMBIT P-KN3 5 P-K4 B-N2 6 P-KR3, Black decides that White's last

move is a wasted tempo in Benko Gambit positions. The game Mititelu-Bilek, Zalaegerszeg 1 960, continued : 6 . . . P-QN4!? 7 P X P P-QR3 8 P X P 0-0 9 B-Q3 N X RP 10 KN-K2 N-QN5 1 1 0-0 N-Q2? (N X B) 1 2 B-QN 5 ! N-N 3 1 3 B-K3 B-Q2 1 4 B X B Q X B 1 5 P-R3 N-R3 1 6 P-QN4 N-B 5 ! 1 7 P-N5 N x B 1 8 B P X N N-B2 19 P-QR4 R-R4 ! 20 Q-B2 N-R l 2 1 R-R2, drawn. Black, with his powerful Bishop and solid blockade on the Queenside, might have continued without risk. Also see the game Barcza-Filip in Variation A. In the game Mititelu-Plachetka, Luhacovice 197 1 , another Queen­ side impasse was reached: 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 P-K4 0-0 5 P-KR3 P-B4 6 P-Q5 P-Q3 7 B-N5 P-N4!?

8 P X P Q-R4 9 B-Q2 P-QR3 10 P X P QN-Q2 1 1 N-B3 B X P 12 B X B Q X B (The chances are equal) 1 3 Q-K2 KR-N l 1 4 QR-N l R-N5 1 5 P-R3 R-N2 (R-N6) 1 6 Q x Q R X Q 1 7 0-0 R/ 3-N3 1 8 N-R4 R-R3 19 N-B3 R/ 3-N3 20 N-R4 R-R3, drawn. The game Farago-Szabo, Budapest 1 972, was an interesting fight: 1 P-Q4 N-KB3 2 P-QB4 P-KN3 3 N-QB3 B-N2 4 B-B4 P-B4 5 P-Q5 P-Q3 6 P-K4 P-QN4!? 7 P X P 0-0 8 N-B3 P-QR3 9 P X P

QN-Q2 1 0 N-Q2 N-N3 1 1 B-K2 B X P 1 2 0-0 Q-Q2 1 3 R-K 1 B X B 1 4 Q X B KR-N l 1 5 P-KR3 Q-N2? (N-R5 !) 1 6 P-QR4 ! (White has the edge) 16 . . . R-R4 17 R-R3 R/ 1 - R 1 18 N-N3 R / 4-R3? (Necessary was 18 . . . R/ 4-R2 in order to avoid the pin after White's 2 1 st move) 1 9 P-R5 N-B 1 20 P-K5 ! (This is now a decisive breakthrough) 20 . . . N x P 2 1 Q-B3 Q-N5?! (If 2 1 . . . P-K3 22 P X P and 23 N X P) 22 R-K4 N X B 23 R X N ! Q-N l 24 R X P P X P 2 5 N X P Q X P (If 2 5 . . . R X P 2 6 R X Bch) 2 6 N X R Q X R 27 Q-Q5 P-K3 28 Q-Q7 B-B 1 29 N-K4 N-K2 30 Q x P and White won.

F3

Sometimes it is possible to transpose to the Benko Gambit from the P-KN3 variations of the King's Indian Defense. For instance, after 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-KN3 4 N-KB3 B-N2 5 P-KN3 0-0 6 B-N2 P-Q3 7 0-0, a well-known position is reached (Korchnoi-Bouaziz, Sousse 1 967, for example). 79

THE B EN KO

GAMBIT

White has intentionally postponed the development of his Queen Knight in order to be able to meet 7 . . . P-K3 8 P X P B X P 9 N-N5 B X P 1 0 B X P QN-Q2 with 1 1 N-R3 ! with advantage. But Black may avoid that now dubious Exchange sacrifice line by transposing to the Benko Gambit. In the game Vukic-Velimirovic, Yugoslavian Championship 1 972, Black postponed the gambit slightly by 7 . . . P-QR3?! (P-QN4 is better) 8 N-B3 (P-QR4) P-QN4?! 9 P x P P X P 10 N x P QN-Q2 1 1 N-B3 N-N3 !? 12 R-K 1 Q-Q2 1 3 P-QR4 B-QR3 14 P-K4 (P-R3) N-N5 15 Q-B2 KR-N l 16 P-N 3 P-B5 ! (Starting a wild melee of complications) 17 P-N4 N x QRP 18 R X N B X N 19 R X B B X R 2 0 R X R B X Pch 2 1 Q X B ! R X R 2 2 Q-Q4 Q-R2 (Black has the more promising position) 23 B-N2 Q x Qch 24 N X Q R-R7 25 B-QB3 N-K6 26 B-B3 N-B7 27 N-K2 R-R6 28 K-B2 R-N6. Black has the advantage here because White's pieces are tied down to the maintaining of the blockade of Black's dangerous QBP. However, the game ended in a draw. From the diagram, the more direct transposition was seen in Portisch-Szabo, Budapest 1 972: 7 . . . P-QN4!? 8 PxP P-QR3 9 N-B3 P X P 10 N x P (Reaching the same position as in the previous example, in which White did not avail himself of the opportunity to avoid it) 10 . . . B-QR3 11 N-B3 QN-Q2 12 Q-B2 Q-R4 1 3 R-K 1 KR-N 1 (N-N3 !) 1 4 B-Q2 Q-N3 (A loss o f time. N-N3 was still the correct move) 15 P-N 3 R-R2 (P-BS) 16 QR-N 1 Q-N 2 1 7 B-B4! Q-N5 ( 1 7 . . . N X P 1 8 N X N Q x N 1 9 N-K5 and 2 0 N-B6) 18 P-KR3 R-QB 1 19 B-Q2 P-B5 20 KR-QB l R/ 2-B2 2 1 N-Q4! P X P 22 N-B6 P X Q 23 R X Q N-B4 24 R X P, and White won . The game Dzh indzhichashvil i-Grigorian, USSR Championship 1 972, varied with 12 P-KR3 Q-N 3 13 Q-B2 KR-N 1 14 R-K 1 N-K 1 80

THE BENKO GAMBIT 1 5 R-N l !? N-B2 (How about 1 5 . . . B X N !? 1 6 Q X B B-N2 to regain the pawn?) 1 6 B-N 5 ! K-B 1 1 7 Q-Q2 N-N4 1 8 B-R6 N x N 1 9 Q X N N-B3 2 0 B X Bch K X B 2 1 P-K4 Q-N5 2 2 Q X Q R X Q 23 P-K5, and White had a slight edge which h e was able t o convert to a win. In Mista-Spiridonov, Cienfuegos 1 972, White played 9 P X P (instead o f 9 N-B3). The game continued : 9 . . B X P 10 R-K1 (N-B3) QN-Q2 1 1 P-K4? ! N-N5, transposing to a position similar to Variation E 1 , where this game is given. A different approach for White was seen in Keene-Friedgood, Caorle Zonal 1 972: 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-KN3 .

4 N-KB3 B-N2 5 P-KN3 0-0 6 B-N2 P-Q3 7 0-0 P-QN4!? 8 P X P P-QR3 9 P-N6?! QN-Q2 10 N-B3 N X NP 1 1 P-K4 B-N5? ! (P-QR4)

1 2 Q-B2 B X N 1 3 B X B KN-Q2 14 R-N 1 Q-B2 15 B-K2! P-K3 1 6 P-QR4! P X P 1 7 P-R5 ! N-B 1 1 8 N X P Q X P 1 9 B-KN5 P-B3 20 B-Q2 Q-Q 1 2 1 N-B4 (White has an obviously superior position. The many holes in Black's camp are too high a price to pay for his extra pawn) 2 1 . . . R-K 1 22 B-B4ch K-R 1 23 N-K6 Q-N 3 24 P-QN4 Q-R2 25 B-Q5 QR-N 1 26 P X P N X P 27 B-K3 , and White won. At the Skopje Olympiad 1 972, the game Csom-Ghizvadu pro­ ceeded: 1 N-KB3 P-KN3 2 P-KN 3 B-N2 3 P-Q4 P-QB4 4 P-B4 N-QB3 5 P-Q5 N-R4 6 KN-Q2 P-QN4? ! 7 P X P P-Q3 8 N-QB3 N-KB3 9 B-N2 0-0 10 QR-N 1 P-QR3 1 1 0-0 P X P 12 N X P B-QR3 1 3 P-QR4 Q-Q2 1 4 P-QN4!? P X P 1 5 R X P QR-N 1 1 6 N-N3 B X N 1 7 P X B R X P 1 8 R X R Q X R, drawn.

81

THE BENKO

GAMBIT

G The Gambit Declined This section deals with possibilities when White does not accept the offered pawn on the Queenside. These lines have been corning more and more into the limelight as the gambit's reputation grows. However, when White declines the offer Black generally is able to obtain a free and active game and may even be able to seize the initiative. Some resemblance to the Polish Defense may be noted. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4!? 4 N-KB3?!

This position may come about in various ways-we can even find plentiful examples from half a century ago. For instance, Rubinstein-Spielmann, Vienna 1 922: 1 P-Q4 N-KB3 2 N-KB3 P-B4 3 P-QS P-QN4 4 P-B4 (Reaching the diagram position in a round­ about way) 4 . . . B-N2 5 P-QR4!? P X BP?! 6 N-B3 P-K3 (P-N 3) 7 P-K4 N x KP? ! (Imaginative play typical of Spielmann, but Rubin­ stein is not lacking in that department as we shall see) 8 N X N P X P 9 N-B3 P-Q5 1 0 B X P ! (The simplest. Sacrificing the piece back makes Black's idea rather dubious) 10 . . . P X N 1 1 B X Pch! K X B 1 2 Q-N3ch (White is clearly better) 1 2 . . . P-BS 13 N-K5ch ! K-N 1 1 4 Q X B N-B3 1 5 N XN Q-K 1 ch 1 6 B-K3 P X P 1 7 Q XNP PXN 18 0-0 B-Q3 1 9 Q-B3 Q-K3 20 QR-B 1 and White won in an instructive Rook and pawn endgame. While most examples continue with 4 . . . B-N2, there has been some experience with other moves. In the game Balashov-Stein, Moscow 1 97 1 , the continuation was 82

THE

BENKO GAMBIT

4 . . . P x P?! 5 N-B3 P-N3 6 P-K4 P-Q3 7 B X P B-KN2 8 P-K5 !? P X P 9 N x P 0-0 1 0 0-0 KN-Q2 1 1 N-B6!? N X N 1 2 PXN N-N3 13 B-K2 Q-B2 14 B-B3 B-QR3 15 R-K 1 QR-Q 1 16 Q-N 3 B-B5 17 Q-R 3 B-Q5 18 B-K3 P-K4 (Black's strong center position gives him the edge) 19 P-QN4! P X P 20 Q X N P B X B 2 1 P X B P-B4 22 N-N5 B XN 23 Q X B P-K5 24 B-Q 1 R-B3 25 B-N 3ch K-N2 26 QR-B 1 R-QB 1 27 Q-R6, drawn. Another approach was seen in Malich-Diez del Corral, Siegen 1 970, which continued: 4 . . . P-Q3!? 5 P-QR4? ! P X BP 6 N-B3 QN-Q2 7 P-K4 B-R3 8 Q-B2 Q-N 3 9 P-R5 Q-N2 10 N-Q2 N-K4 1 1 N X P N X N 1 2 B X N B X B 1 3 Q-R4ch N-Q2 1 4 Q X B P-N3 (Black has a small edge because of his pressure on the dark squares) 1 5 P-R6 Q-B 1 16 P-R4? ! B-N2 17 P-R5 N-K4 1 8 Q-K2 P-B5, and despite Black's slight advantage the game was eventually drawn.

Gl

After 1 P-Q4 N-KB3 2 P-QB4 P-84 3 P-QS P-QN4 4 N-KB3?! B-N2, we will examine various continuations for White on his fifth move. Fourth-move alternatives for White will be discussed under G2.

5 P-QR4!? P-QR3 was played in Kan-Keres, USSR Champion­ ship 1955: 6 RP X P P X P 7 R X R B X R 8 N-B3 Q-R4 (8 . . P-N5 9 N-QN5 Q-N3 is good for Black, as is 9 Q-R4 P X N 10 Q x B Q-N3) 9 N-Q2 P-N5 1 0 N-N3 Q-N3 1 1 N-R4 Q-B2 1 2 P-B3 P-Q3. Black stands well and eventually won. Instead of 5 · . . P-QR3, the game Balashov-Piatonov, USSR .

.

83

THE

BENKO

GAMBIT

Championship 197 1 , continued: 5 . . . P-N5 6 QN-Q2 P-Q3 7 P-K4 P-K4 8 P X P e.p. P X P 9 B-Q3 P-K4 10 Q-K2 N-B3 1 1 N-B 1 N-QS 1 2 N X N B P X N 1 3 P-B4 B-K2 1 4 N-N3 0-0 1 S 0-0 (N-BS) 1S . . . N-Q2 16 B-Q2 P-QR4 17 N-BS R x N ! (Black gets excellent attacking chances for the Exchange) 18 P X R N-B4 19 P x P P X P (The connected passed pawns in the center are worth more than the Exchange as White is soon compelled to give up material) 20 P-B6 B X BP 2 1 B-BS Q-K2 22 QR- K 1 P-K5 23 Q-N4 Q-Q3 24 R-B4 P-K6 2S B X P P X B 26 R X P R-Q 1 (Black has an obvious superior­ ity) 27 R-K 1 Q-Q7 (Q-B3) 28 R/ 1 -KB 1 Q-K6ch 29 R/ 1-B2 B-QS?! (Black starts to go astray. Stronger was 29 . . . R-Q7) 30 K-R 1 Q-K 1 ? (Here 30 . . . Q X R was correct. The text relinquishes Black's advantage and the game was finally drawn.) In the USSR Championship of 19S7, the game Klaman-Keres continued: 5 P-KN3 P-N3 6 B-N2 B-N2 7 0-0 P-Q3 8 R-K 1 P X P 9 P-K4 0-0 1 0 KN-Q2 QN-Q2 1 1 N X P R-N 1 1 2 B-Q2 N-N3 1 3 N-RS B-R 1 1 4 N-R3 KN-Q2 1 S R-N 1 P-K3, with slightly better chances for Black. The game was later drawn. In the game Johannessen-Fischer, Havana Olympiad 1 966, the continuation was: 6 . . . P x P 7 N-B3 B-N2 8 0-0 0-0 9 N-K5 P-Q3 10 N x QBP QN-Q2 11 R-Kl B-QR3 12 Q-R4 Q-B 1 1 3 N-RS N-N3 (Black is slightly better) 1 4 Q-R4 R-K 1 15 B-N S Q-B2 1 6 N-B6? ! (The Knight only seems strongly posted here, but actually it has little to do. Furthermore, it will soon be undermined with disastrous consequences) 16 . . . B-N2 17 P-K4 QN-Q2 1 8 P-B4? ! K-R l ! (Preparing to destroy the Knight's suppor t by avoiding a possible check) 1 9 P-KS P X P 20 P X P (N X P/ S) 20 . . . N X QP ! 2 1 N x N Q x N (Black has a clear advantage) 22 P-K6 N-K4 2 3 R X N B X R 24 P X P R-KB 1 (And wins) 2 S P-KR3 ( 2 5 N-B4 B-QSch and 26 . . . R X P!) 2S . . . R x P 26 N-B4 R X N, and White resigned since 27 B X Q is answered by 27 . . . B-QSch . Instead of 1 1 R-K 1 , the game Castagna-Lundwall, Biel 1 970, continued: 1 1 P-K4?! B-QR3 12 Q-R4 Q-B 1 13 R-K 1 N-N 3 1 4 N X N P x N (Black has a slight edge) 1 S Q-N 3 N-Q2 (P-QN4!?) 1 6 B-NS R-K 1 1 7 B-R3 Q-B2 (Q-N2) 1 8 N-NS B X N 1 9 Q x N N-K4. The game is even and was later drawn. In an exciting game between Guimard and Keres, Goteborg 19SS, the continuation from the last diagram was: 5 N-B3?! P-NS 6 Q-N 3 Q-R4? ! (Q-B2 !?) 7 N-Q 1 P-K3 (P-N 3) 8 P-K4!? N x KP 9 B-Q3 N-KB3 10 0-0 P-N 3 (B-K2) 1 1 B-N S B-N2 12 N-K3 0-0 1 3 84

THE

BEN KO GAMBIT

P-KR4 Q-B2 (More urgent was P-Q3 and QN-Q2) 1 4 QR-K 1 P-Q3 15 B X N B X B 16 N-N4 B-N2 17 P-R5 ! KP x P? (P-K4) 1 8 RP x P RP X P 1 9 N-N5 ! Q-Q2 20 R-K6 ! This position deserves a diagram:

White has a strong attack, but unfortunately he spoils it at a critical moment. 20 . . . P X R 2 1 B X P R-B4 22 Q-K3? (Much stronger was 22 B X R ! P X B 23 Q-KR3 ! with a powerful attack, e.g. 23 . . . P X N 24 Q-R7ch K-B 1 25 R-K l !) 22 . . . R X N ! 23 Q X R N-B3 24 P X P Q-K2! (The advantage is now Black's) 25 Q-R5 N-K4 26 R-K 1 B X P 27 Q-R7ch K-B 1 28 N-R6 Q-B3 29 R X N (Time-pressure desperation) 29 . . . P X R 30 P-B3 B XN 3 1 Q x Bch K-K2 3 2 Q-R7ch K-Q3 , White resigned. A dubious plan for White is 5 P X P?! B X P, for he has exchanged a strong center pawn for a flank pawn, giving Black the better of the deal. The game Tarusevich-Zaitsev, Moscow 1 969, continued : 6 N-QB3 B-N2 7 Q-R4 P-Q4 8 N-K5 P-K3 9 N-B6 Q-N 3 1 0 N-R5 (White has spent three tempi to try to capture Black's Queen Bishop, which could now simply move away. Black, however, prefers to continue his development) 10 . . . QN-Q2 1 1 P-K3 B-Q3 12 N X B Q X N 1 3 Q-R6 Q-N 1 1 4 B-K2 0-0, and Black has a n advantage in development and a strong center, which he successfully exploited. Another game with Keres as Black was that against Hartoch at Amsterdam 1 97 1 , which continued: 5 P-K3 P-N3?! (5 . . . P-K3 !? 6 QP X P BP X P 7 P x P P-Q4, similar to the Blumenfeld Gambit, had to be taken into consideration) 6 N-B3 P-N5 7 N-K2 B-N2 8 N-N 3 P-K3 9 P-K4 P X P 1 0 KP X P 0-0 1 1 B-K2 P-Q3 1 2 0-0 QN-Q2 (Black's strong King Bishop and good piece coordination give him the preferable position) 1 3 B-B4 N-N3?! ( 1 3 . . . Q-N3, to be followed by . . . QR-K 1 , was a good plan) 1 4 Q-Q2 R-K 1 1 5 85

THE BENKO GAMBIT B-Q3 B-QR3 (P-QR4) 1 6 QR-B 1 R-QB 1 1 7 P-N3 R-B2 1 8 P-KR3 Q-K2 (If 1 8 . . . R/ 2-K2 1 9 B-N5 followed by N-R2-N4) 1 9 KR-K 1 Q-B 1 20 R X R N X R. The chances are roughly equal, but White eventually won. The game Kolarov-Peev, Varna 197 1 , continued 5 QN-Q2 P-Q3 6 P-K4 P-N5 (P X P !?) 7 B-Q3 P-N 3 8 0-0 (P-QN3) B-N2 9 R-K 1 QN-Q2 1 0 N-B 1 0-0 1 1 R-N 1 P-QR4 1 2 P-QR4 P X P e.p. 1 3 P X P Q-B2 1 4 N-K3 KR-N 1 1 5 Q-B2 P-K3 1 6 B-N2 P X P 1 7 N X P B X N 1 8 KP X B R-K 1 , drawn.

G2 After 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-QS P-QN4, White has several ways of continuing on his fourth move.

The game Lombard-Diez del Corral, Montigny 1973, continued: 4 N-Q2?! (see also Lyman-Benko in the Appendix) 4 . . . P-Q3 (On 4 . . . Q-R4 there is a possible countergambit with 5 P-QN4!?) 5 P-K4 P X P 6 N-K2!? (Original ! The plan is N-QB3 and N x P!) 6 . . . P-N3 7 N-QB3 B-KN2 8 N x P! 0-0 9 B-K2 B-QR3 1 0 0-0 QN-Q2 1 1 B-N5 R-N 1 1 2 Q-B2 B X N 1 3 B X B N-K 1 ? (Better is 1 3 . . . P-KR3 14 B-R4 P-N4 and 1 5 . . . N-R4, or 1 3 . . . Q-R4) 14 QR-K 1 N-B2 1 5 P-B4 (White has the edge) R-N5 1 6 P-QN3 N-N3 1 7 P-QR3 B-Q5ch 1 8 K-R 1 N X B 1 9 P X R N-K6 20 R X N B X R 2 1 Q-Q3 P X P (B-Q5) 2 2 N-K2 ! B-B4 23 B-R6 R-K 1 2 4 P-B5 (White has clearly better chances) P-B3? (Q-Q2) 25 P X P P X P 26 P-K5 ! K-R2 27 N-B4 R-N 1 28 R-B3 K X B 29 R-R3ch K-N2 (29 K-N4 30 R-R7) 30 Q X Pch K-B 1 3 1 Q-R6ch K-K1 32 P-K6? 86

THE BENKO

GAMBIT

(R-N 3 !) N X KP 3 3 P x N Q-R4! 34 Q-N6ch K-Q l ! 35 Q x Rch K-B2 36 P-N3 Q-R8ch 37 K-N2 Q-N8ch 38 K-B3 Q-Q8ch 39 K-N2 Q-B7ch 40 K-B3 P-B4! 4 1 N-Q5ch K-N2 42 N-K3 Q-K5ch, White resigned. 4 N-QR3?! was played in Velimirovic-Niegovan, Yugoslavia 1 966: 4 . . P-N5 (P-QR3 or PXP is playable) 5 N-B2 P-K4 6 P-KN3 P-Q3 7 B-N2 B-K2 (P-N3) 8 P-K4 0-0 9 N-K3 N-K 1 1 0 N-B3 P-N3 1 1 P-N4!? N-N2 12 P-KR4 P-B4 13 KP X P P X P 14 N x BP N X N 1 5 P X N B X BP, and White's game is slightly preferable. A violent attempt at refutation was a counter-sacrifice by White in the game Peev-Pedersen, Skopje Olympiad 1 972, which con­ tinued: 4 P-QR4?! P X BP (P-N5) 5 N-QB3 P-Q3 6 P-K4 B-R3 (P-N3) 7 P-B4 QN-Q2 8 N-B3 Q-N 1 9 Q-K2 Q-N2 10 P-KN3 P-K4 1 1 P X P e.p. P X P 1 2 P-K5 N-Q4 1 3 N x N Q x N 14 P X P 0-0-0 ( A rather unusual position with unclear ramifications) 1 5 B-N2 B X P 1 6 N-N5 Q-B4 1 7 0-0 N-N 1 1 8 B-R3 Q-Q6 1 9 Q x Pch K-B2 20 Q-B7ch R-Q2 2 1 B X R N X B 22 N-K6ch K-B 1 23 R-R3 P-B6 24 R-K 1 B-N2 25 N-N5 (25 R X P Q-Q4) 25 . . . P-B5 26 R X P (26 B-K3 Q-B7) B-B4ch 27 B-K3 Q-Q7 , and White resigned. An interesting fight. 4 P-QN3?! P x P 5 P x P P-Q3 was played in Masera-Benko, Reggio Emilia 1 97 1 , which continued: 6 N-QB3 (N-Q2) P-N3 7 B-N2 (R-N 1 ) B-KN2 8 P-B3 QN-Q2 9 P-K4 R-QN 1 10 Q-B2 (If 10 R-N 1 R X B ! 1 1 R X R N X QP, etc.) 10 . . . Q-R4! (White's routine developing moves have led to difficulties for him. His King is unsafe and if e.g. 1 1 N-K2 B-QR3) 1 1 K-82 N x Pch 12 P X N R X B 1 3 Q x R B XN 1 4 Q-B 1 N-B 3 ! 1 5 N-B3 N X Pch 1 6 K-K3 B X R 17 Q X B N-B3 1 8 B-Q3 0-0 19 P-KR3 B-R3 20 R-QN 1 P-K4 2 1 P X P e.p. P X P 22 R-N3 Q-B2 23 R-R3 B-B 1 24 Q-N2 P-Q4 25 Q-K5 N-N5ch!, White resigned, for if 26 P X N R X Nch and 27 . . . Q x Q. The game Torner-Berglund, Linkoping 1 970, continued in the same vein but with a different order of moves: 5 . . P-N3 6 B-N2 P-Q3 7 P-K4 B-KN2 8 B-Q3 0-0 9 N-K2 QN-Q2 (The chances are equal) 10 P-B4? ! R-N 1 (But now Black has the edge, e.g. 1 1 Q-B2 R X B). 4 B-N5?! was tried in Pardo-Benko, Olot 1 97 1 : 4 . . . P X P 5 N-QB3 P-Q3 6 P-K4 P-N3 7 B X P B-KN2 8 KN-K2 0-0 9 0-0 QN-Q2 10 N-N3 Q-R4 1 1 Q-Q2 (Q-K2) 1 1 . . . R-N 1 1 2 QR-B 1 B-QR3 1 3 B X B Q X B 1 4 R-B2 R-N5 ! 1 5 Q-K2? ! Q x Q 1 6 KN X Q? .

.

87

THE BENKO GAMBIT (A mistake, but even the better 1 6 R X Q KR-N 1 1 7 P-N3 P-B5 leaves Black with a clear advantage) 16 . . . N x KP ! (Wins) 1 7 B x P? ( 1 7 P-QR3 R-B5) 1 7 . . . R-K 1 1 8 B-R4 N X N 1 9 N x N R X B 20 P-B4 B-Q5ch 2 1 K-R 1 R X BP, White resigned. The play on the QN file is instructive. Instead of 4 . . . P x P, the move 4 . . . N-KS was tried in Shashin-Domes, USSR 1 97 1 , leading to a sharp fight: 5 B-B4 !? (If 5 B-R4 P-N4) 5 . . . Q-R4ch 6 N-Q2 P-Q3 (P X P might h ave been tried, to avoid the unclear complications which now ensue) 7 P-QN4! Q X NP 8 R-N 1 Q-B6 9 R X P (If 9 R-N 3 Q-B3) 9 . . . N X N 1 0 B X N Q X P 1 1 P-K3? ! (Better was 1 1 Q-N 1 N-Q2 1 2 P-K4) 1 1 . . . Q X RP 1 2 B-B3 P-K3 1 3 B-Q3 B-Q2 (Not 1 3 . . . Q x P 1 4 Q-N 1 threatening both R X N and B-K4) 1 4 P X P Q x P 1 5 N-K2 ! (White has a strong initiative) 1 5 . . . P-Q4 1 6 N-B4 Q-Q3 1 7 0-0 P-Q5 1 8 P X P ! Q x N 1 9 R-K 1ch B-K2 20 R X P K-B 1 2 1 B-K4 N-B3 22 B X N B X R 23 B X R (White has a winning position) 23 . . . B-K2 24 Q-K2 Q-Q3 25 R-N 1 P-B3 26 Q-B4 P-N 3 27 B-N4 Q-K3 28 B-Q5 Q-B4 29 B X Bch K X B 30 Q-B5ch K-Q 1 3 1 R-N 8ch B-B 1 3 2 Q-Q6ch Q-Q2 3 3 R X Bch, Black resigned. 4 Q-B2 P x P was played in Enevoldsen-Beyer, Munich 1 9 5 8 : 5 P-K4 N-R3? (Black is too rash in trying to take advantage of the position of White's Queen. It was better to develop by . . . P-Q3 and . . . P-N3) 6 B X P N-QN5 7 Q-K2 Q-R4 (The Knight was already threatened by P-QR3) 8 N-QB3 P-Q3 9 N-B3 P-N 3 10 0-0 B-KN2 1 1 P-QR3 N-R3 (If 1 1 . . . 0-0 1 2 P X N !) 1 2 P-K5 ! (White already has a won game) 1 2 . . . N-Q2 (A better, but still losing, choice was 12 . . . P X P 13 N X P and 14 N-B6) 13 P X P 0-0 1 4 P X P R-K 1 1 5 P-Q6 B-N2 1 6 B X Pch, Black resigned because o f 1 6 . . . K X B 1 7 Q-B4ch. A game in 197 1 between Lim and Browne continued : 4 P-B3?! P X P 5 P-K4 P-Q3 6 B X P P-N 3 7 N-B3 B-KN2 8 KN-K2 0-0 9 0-0 QN-Q2 10 B-K3 (Better was B-B4) 10 . . . N-K4 1 1 B-N 3 B-QR3, with Black a little better. White's P-B3 is much stronger if preceded by 4 P x P P-QR3. The game Visier-Gheorghiu, Adelaide 197 1 , continued: 5 P-B3 P-N 3 (If 5 . . . P X P 6 P-K4 threatening both P-K5 and B X P) 6 P-K4 P-Q3 7 N-B3 B-KN2 8 B-N5 (Simpler is accepting the pawn by 8 P X P B X P 9 B X B N x B 10 KN-K2) 8 . . . 0-0 9 Q-Q2 R-K 1 1 0 P-QR4 QN-Q2 1 1 N-R3 N-N3 1 2 N-B2 P-K3 !? 1 3 P-R5 !?

88

TH E B E N K O G A M B IT RP X P 1 4 B X P B-Q2 1 S N-N4? (Better was 1 S P X P and if 1 S P X P 1 6 Q X P B X B 1 7 Q X Q KR X Q 1 8 N X B N-BS 1 9 0-0, o r if 1S . . . R X KP 1 6 B-K2) 1S . . . B X B 1 6 N X B N-BS (Now Black has the better game) 17 Q-B3? (A better choice was 1 7 Q-B 1 , allow­ ing 17 . . R X P) 1 7 . . . N XN 1 8 Q x Bch K x Q 19 B X Q KR X B 2 0 P X N QR-N 1 2 1 N-B3 R X P and Black won. After 4 P X P P-QR3, possible is 5 N-QB3 (instead of S P-B3) S P X P 6 P-K4? Here White must take the pawn. After 6 N x P B-R3 7 N-B 3 , we have a regular gambit variation with each side having lost a tempo. The game Vannrud-Akvist, Eksjo 1 970, con­ tinued: 6 . . . P-NS 7 P-KS P X N 8 P X N P X N P 9 B X P Q-R4ch 10 Q-Q2 Q x Qch 1 1 K X Q N P X P and White is already lost. .

•

.

.

G3

Another way for White to decline the gambit is to accept the pawn at first and then return it with P-N6 more or less immediately. The resulting positions are somewhat similar to those after Black plays NP X BP, seen earlier in this section (see also Reshevsky­ B rowne in section D). Due to the nature of the pawn structure, Black must pursue active operations on the open QN file at once. Black is favored in these lines somewhat more than in the usual closed positions. 1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P-N6?!

89

THE BENKO GAMBIT There is not much experience with this variation, and in fact there are only two principal lines for Black, 5 . . . Q x P and 5 . . . P-Q3 followed by . . . QN-Q2 and the later capture of the NP. The immediate capture by the Queen is not of great urgency-de­ velopment of the Knight first offers Black a choice of capturing with the Queen or Knight according to circumstances. The game Mendoza-Benko, Malaga 1 969, continued: 5 . . . Q x P 6 N-QB3 P-N3 7 P-K4 P-Q3 8 R-N 1 B-KN2 9 Q-R4ch? QN-Q2 10 B-Q2 0-0 1 1 N-B3 R-N 1 12 B-Q3 P-B5? 1 3 B-K3 ! ( 1 3 Q x BP? N-N5 threatening both Q X Pch and N-K4, or 13 B X P? N-B4 1 4 Q-B2 KN x KP! with Black o n top) 1 3 . . . N-B4 1 4 Q x BP KN-Q2 1 5 0-0 P-QR4 1 6 N-Q4, and in this unclear position, the game was agreed drawn. White varied in Agdamus-Gheorghiu, Buenos Aires 1 970: 7 N-B3 (instead of P-K4) 7 . . P-Q3 8 N-Q2 B-KN2 9 P-K4 0-0 10 N-B4 Q-B2 1 1 B-K2 QN-Q2 1 2 B-B4 N-N 3 1 3 N-K3 P-QR4 14 Q-B2 P-R5 (Black has the edge) 1 5 0-0 B-QR3 (B-Q2) 16 B X B R X B 1 7 P-KR3 KN-Q2 1 8 B-N5 R-K 1 1 9 K-R 1 P-B5 (White is unable to generate any initiative compared with Black's purposeful play on the Queenside) 20 QR-K 1 P-R6 2 1 P X P N-B4, and Black successfully exploited his superior position. The more flexible method for Black is to postpone the capture of the QNP. The game Bagnoli-Bilek, Reggio Emilia 1964, continued: 5 . . . P-Q3 6 N-QB3 QN-Q2 7 P-K4 P-N3 8 B-Q3 B-KN2 9 KN-K2 0-0 1 0 P-QR4? Q X P 1 1 P-R5 Q-B2 1 2 0-0 R-N 1 1 3 N-N 3 P-B5! 1 4 B-B2 (B-K2) 1 4 . . . N-B4 1 5 Q-K2? N-N2! and Black won. In a game Wastfeld-Green, Stockholm 1 97 1 , the continuation was: 7 P-KN3 P-N3 8 B-N2 B-KN2 9 N-B3 0-0 10 0-0 N x NP 1 1 P-K4 B-N5 (KN-Q2) 12 P-KR3 B X N 13 Q X B Q-Q2 (P-B5) 14 Q-K2 N-R5 15 N XN Q XN, with equality. A more recent try, Pfleger-Benko, Skopje Olympiad 1 972, con­ tinued: 7 N-B3 P-N3 8 P-K4 B-KN2 9 B-K2 0-0 10 0-0 N X NP 1 1 B-KB4 B-N5 1 2 N-Q2 (If 12 P-KR3 B X N 13 B X B KN-Q2 14 B-K2 P-B5 with equal chances) 1 2 . . . B X B 13 Q X B Q-Q2! (Now Black has the better chances because White's Bishop is misplaced and he must either consent to the following maneuver or allow Black to run free on the Queenside after . . . N-R5) 14 P-QR4? ! (P-KR3 was the better choice) 1 4 . . . Q-N5 ! (Black looks for the endgame which favors him because of White's new weaknesses on .

90

TH E BEN K O G A M B IT the Queenside) 1 5 Q-K3 N-R4 1 6 B-R6 B X B 1 7 Q X B Q-B5 1 8 Q x Q N x Q 1 9 P-QN 3 QR-N 1 20 P-N 3 N-Q6 2 1 P-B4 P-B4! (White is in trouble) 22 P-R5 N-Q2 23 QR-Q 1 P X P 24 N/ 3 X P (If 2 4 N/ 2 X P N-N7 followed by . . . R X NP) 2 4 . . . N-N5 25 N-KB3 (If 25 N-B4 N-B7 and . . . N-Q5) 25 . . . P-R3 ?

Black could have won a pawn with 2 5 . . . R-N4! , a s 26 N-B3 R X RP 27 N-Q2 does not trap the Rook because of 27 . . . N-N 3 . The game continued: 26 N-B3 N-B7, and here Black offered a draw which was accepted by White, of course. However, 27 R-B2 N-Q5 28 N x N P X N 29 R X P (29 N-K2 R-N4) R X P is still difficult for White. Also interesting is 27 R-B2 N-R6 !? By transposition, the same opening position was reached in a game Dzhindzhichashvili-Alburt, USSR Championship 1 972: 1 N-KB3 N-KB3 2 P-Q4 P-B4 3 P-Q5 P-QN4!? 4 P-B4 P-N 3 !? 5 P X P P-QR3 6 P-N6 P-Q3 7 N-B3 QN-Q2 8 P-N7?! (An interesting idea to try to force the Black QB to block the QN file) 8 . . . B X P 9 P-K4 B-N2 1 0 B-K2 0-0 1 1 0-0 Q-B2?! 1 2 B-KB4! (Preventing . . . P-K3) QR-N 1 (KR-N 1 ) 1 3 Q-Q2 B-B 1 (Was White's 8 P-N7?! justified?) 1 4 KR-K1 Q-R4 15 QR-N l N-N 3 16 P-K5 N-R4 17 B-R6 B-N5 18 B X B N X B 19 P X P P X P 20 P-N4! P X P 21 N-K4 P-B4? (N-K 1 ) 22 Q-R6! (White obviously has the better game) 22 . . . N-K 1 (P X N 23 N-N5) 23 N/ 3-N5 R-N2 24 N-K6 R/ 1-B2 25 B X B P X B 26 N-Q8 N X P 27 N X QR R X N 28 N-B5 R-K2 29 N-K6 N-N2 30 N xN RXN 3 1 R-K8ch K-B2 32 R/ 1-Kl Q X P 3 3 Q-N5, Black resigned.

91

TH E

BENKO GAMBIT H

The Gambit Reversed Finally, the question arises: If the gambit has been able to prove its worth in practice, would it not be still better to adopt it with the White pieces a tempo ahead? There have been only a few examples of this. Probably, since White already has some initiative by virtue of his having the first move, it is considered unnecessary to offer a pawn sacrifice which is de­ signed to gain the initiative. However, without prejudice, we can discuss it as a possible varia­ tion of the future. An early treatment of the gambit reversed was seen in Geller­ Mikenas, USSR Championship 1 95 5 : 1 N-KB3 P-Q4 2 P-84 P-Q5 3 P-KN3 P-QB4 4 8-N2 N-Q83 5 0-0 P-K4 6 P-Q3 8-K2 7 P-QN4!? P X P 8 P-QR3 P X P 9 Q-R4! B-Q2 1 0 B X P N-B3 1 1 Q-N 5

(White has an edge) 1 1 . . . 0-0? ! (Better was 1 1 . . . B X B 1 2 N X B Q-K2 1 3 Q xNP 0-0) 1 2 N X KP N x N 1 3 Q X N B X B 1 4 R X B ! B-B3 1 5 B X B P X B 1 6 KR-K 1 Q-N 3 1 7 N-Q2 (Now White has a definite advantage) 17 . . . Q-N5 (KR-K l ) 18 Q-QR5 Q-Q3 19 R-N 3 KR-K 1 20 R-N7 R-K4 (P-R4) 2 1 Q-B7 (And wins) 2 1 . . . Q-K3 22 N-B3 R-KR4 2 3 N x P Q-K 1 24 KR-N 1 , Black re­ signed. In a game Lein-Kytov, USSR 1 96 1 , the continuation was: 6 . . . N-83 (instead of B-K2) 7 P-QN4!? P X P 8 P-QR3 P X P 9 N x RP? ! (QN-Q2) B-K2 1 0 P-K3?! 0-0 1 1 P X P P X P 1 2 R-K 1 B-QN5 1 3 B-Q2 B-QB4 1 4 N-K5 (N-B2) 1 4 . . . N X N 1 5 R X N B-Q3 1 6 R-K 1 P-QR3 1 7 N-B2 B-QB4 1 8 B-B4 R-R2 1 9 Q-N 1 B-KB4, and Black had a slight advantage and later won. A more modern treatment of the gambit reversed was seen in Larsen-Portisch, Palma Interzonal 1 970: 1 N-KB3 P-Q4 2 P-84 P-Q5 3 P-KN3 P-Q84 4 P-QN4!? P x P 5 P-QR3 P x P 6 8 x P N-Q83

92

THE

BENKO GAMBIT

A picture of the "classical" Benko Gambit with cotors reversed! The game continued: 7 P-Q3 P-KN 3 8 B-KN2 B-N2 9 0-0 N-R3 10 QN-Q2 0-0 1 1 N-K 1 ? ! (Q-R4) 1 1 . . . B-Q2 12 N-B2 Q-B 1 ! 1 3 N-N4 N x N 1 4 B X N R-K 1 (Black has a slight edge thanks to White's rather slow play) 1 5 R-R2 B-R6 1 6 Q-R4 B X B 1 7 K x B N-N5 1 8 N-B3 P-K4 1 9 P-R3 N-B3 2 0 B-Q6 P-K5 !? 2 1 N X P P X P 22 P X P R-Q l 23 B-K5 N-R4 2 4 B X B N X B 2 5 N-K2 R X P 26 N-B4 R-Q5 27 N-Q5 ! Q-QB4 (Not 27 . . . Q x P? 28 Q x Q R X Q 2 9 N-N6) 28 Q-Q7 Q X P 2 9 R X P! R-KB 1 30 N-B6ch K-R 1 3 1 Q X NP. White has a slight pull, but the game was drawn in 85 moves.

93

THE

BENKO GAMBIT

TWENTY SELECTED GAMES played by PAL BENKO

95

THE

BENKO GAMBIT LIST OF GAMES

Game No.

Opponent

Page

1

H. Avram

2

F. Visier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

3

R. Gross

4

J . H. Donner

5

M. Taimanov

6

C. Chellstorp

7

T. Ghitescu

8

Z. Vranesic

9

L. Kaufman

10

L. Portisch

11

F. A. Kuypers

12

D . Blumin

13

R . Laver

14

M. Vukic

15

J. Aspler

16

A . Denker

17

J. Berry

18

L . Barlay

19

R. Calvo

20

H. Lyman

97

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 02

.

.

.

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . 1 05

. . . . . . . . . . . . . . . . . . . . . . . . . . . . 108

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 12 .

.

.

.

.

.

.

..

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

1 15

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 18

.

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . 1 19

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 22

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 24

.

.

.

.

.

.

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 04 .

.

.

.

.

.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 27 .

.

.. .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. .

.

.

.

.

.

.

.

.

.

. . .

.

.

.

.

.

.

.

. .

.

.

.

.

.

.

.

.

.

.

.

.. .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. .

.

...

.

.

.

.

.

.

.

.

.

.

.

1 29 131 133

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 .

.

.

.

.

.

.

.

. .

.

.

.

. . .

96

.

.

.

.

.

.

.

.

.

.

.

.

. .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.. . .

.

.

.

.

.

.

.

.

. 1 37

. . 1 40 .

THE BENKO

GAMBIT

GAME 1

U. S. Open 1968

aggressive instincts, as in the follow­ ing course of the game he will be forced to go on the defensive.

White: H. Avram Black: P. Benko 1 2 3 4

P-Q4 P-QB4 P-QS PXP

N-KBJ P-B4 P-QN4

It is interesting to note that dur­ ing an earlier game between the same opponents (American Open 1 967), Avram declined the gambit by playing 4 P-B3. 4 5 PxP 6 N-QBJ 7 P-K4

9 10 K-N2 1 1 P-KRJ

P-QRJ BxP P-QJ

I rather expected the more con­ sistent 1 1 P-KR4 (which would h ave been answered by 1 1 . P-R4) instead of this cautious move, which is played to permit the de­ velopment of the King Knight.

The text is aimed at getting rid of Black's strong Queen Bishop, even if this means that White can­ not castle. The forfeiture of the castling privilege, however, can be rectified by castling "by hand." 7 8 KXB 9 P-KN4??!

B-N2 0-0

.

11

BXB P-NJ

. .. . . .

.

P-KJ

Black's plan is to open files in the center and then launch an at­ tack against the loose position of White's King. Should White allow Black to play P X P, White would end up with an isolated pawn in the center. But my opponent has his own plans, as we shall see.

A bold move characteristic of the aggressive style of my opponent. It is doubtful, however, if White is in the position to give in to such 97

THE BENKO GAMBIT 12 P X P 1 3 P-K5!

have been 21 . . . QR-B l , threaten­ ing . . . Q-B7.

PxP

B y returning the extra pawn in this way White breaks up Black's promising center. If Black now accepts the sacrifice the position would still be even following the further exchange of Queens, and Black would have uncertain com­ pensation for his demolished pawn structure. Therefore the pawn offer (as well as the exchange of Queens) is declined with hopes of an attack against the White King. 13 14 P X P 1 5 Q-Kl 16 P-B4

N-K1 NXP Q-Q2

NXN N-K4

R-tn

.

NxN

Correct was t o exchange Queens and go into an endgame a pawn down-after 22 Q X N Q X Q 23 R X Q P-Q6 ! ? (also to be considered is 23 . . . KR-QB l or 23 . . . QR­ QB l ) 24 B-Q2 B X P 25 R-Q l R X P 2 6 K-B 3 , Black's victory would be far from easy because of the active White King.

N-B3 N-Q5

Also 17 . . . QR-K 1 threatening . . . P-K4 came into consideration. Perhaps that would have been stronger than the text. 18 19 20 21

.. ...

22 R X N

This move, not as bad as it may seem, alleviates the pressure on the KB file. 16 17 N-B3

21

22 23 24 25 26 27

PxN Q-B3 KR-Q1

B-Q2 R-Q1 R-N4 K-N3 Q-K4?

P-Q6 BXP RXP Q-Q3 B-B3

White is obviously losing, but he overlooks a direct threat. In order to put up stiffer resistance, 27 R­ N7 or R-N3 should have been tried.

Q-B3?

Better was 21 Q-Q3, blocking Black's QP. The continuation would 98

THE 27 28 R X R 2 9 R/4-N2 30 R X B 31 R X P

BENKO GAMBIT

RXB B-B6 B X R/N7 P-Q7

skin with 32 Q x KPch and petual check. 31

. . . . .

.

a

per­

Q-R6ch

White resigned. After 32 K-R4 Q-K2ch and 33 . . . R X R White has no compensation for the Rook.

White's last trick. In case of 3 1 . . . Q X R White would save his ,

99

THE

BEN KO

GAMBIT

GAME 2

Malaga 1969

13 N x P 14 N/ 2-B3

White: F. Visier Black: P. Benko 1 2 3 4 5 6 7 8 9 10

P-Q4 P-QB4 P-Q5 PXP PXP N-QB3 P-K4 KxB P-KN4??! P-B3

The pin 1 4 B-N5 is not danger­ ous because of the answer 1 4 . . . N X N.

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 BXB P-N3 B-N2

14 15 PxN

. . . . . .

NXN

The reply 1 5 N x N has its draw­ backs too, as the White Queenside would be difficult to develop due to Black's pressure there. In addition, Black would gain a strong outpost on Q5 . I probably would have answered 1 5 N x N by 1 5 . . . P-B4.

Varying from the previous game (against Avram) where P-KR3 was played (one move later). White is apparently contemplating a King­ side attack with P-KR4. 10

N-B3

15

. . . . . .

N-N5!

0-0

Black can afford to castle "into it" because White's own King is not particularly safe. If White con­ tinues with P-KR4, Black, too, can play P-R4. 11

KN-K2

P-K3

Black follows the known pattern of opening the center to reduce the possibilities of an enemy Kingside attack. 12 K-N2

PxP

This is better than the more natural-looking 1 5 . . . N-Q5 . The text move builds a strong post on Q6 and keeps the King Bishop' s diagonal open.

1 00

THE 16 P-KR4 17 P-N5?

B E N KO GAMBIT

P-B4

White gives up all his Kingside chances. He should have tried 1 7 B-N5. 17 18 P-R3

P-QB5

This is j ust a wasted tempo as the Knight is now forced to occupy its best spot. However, it is hard to suggest a constructive plan for White. 18 19 QR-N 1 20 Q-Bl

N-Q6 Q-N3 KR-K1

White resigned. This decision may seem a little premature, but an ex­ amination of the position reveals that White's game is hopeless. The threat is 2 1 . . . N-K8ch, and if White tries 2 1 Q-Q2, then 2 1 . . . B x N. The only reasonable try seems to be 2 1 K-N 3 , but then a simple line is 2 1 . . . B-K4ch 22 K-N2 B XN 23 P X B Q-B4, and wins.

101

THE

BENKO GAMBIT GAME

U. S. Open 1968

the White Queenside and instead openly reveals his plan to attack the center with the aim of breaking it up.

White: R. Gross Black: P. Benko 1 2 3 4 5

6 7 8 9 10 11

P-Q4 P-QB4 P-Q5 PxP PxP N-QB3 P-K4 KXB P-KN3 K-N2 KN-K2

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 BXB P-N3 B-N2 0-0

13 P-N3 14 P-B3 15 P x P

15 1 6 B-K3 17 Q-Q2?

PxP Q-B3

White underestimates the danger from Black's advancing mobile cen­ ter. Here was his last opportunity to neutralize the position with 1 7 P-QN4 P x P 1 8 R X P although in that case 1 8 . . . P-Q4 would offer Black good chances, e.g. 1 9 R-N6 Q-B5 20 Q-N3 (20 Q-Q4? N-N 5 ! ) 20 . . . Q-Q6, etc.

Q-N3 N-R3

,

17 18 19 20 21 22

It should be noted that Black here forsakes the original idea of bringing pressure to bear against

N-B2 P-K3

I t i s hardly a sound idea to give up the center, especially because Black will now have an open file for his Rook.

The position appears safe enough for White although he renounces the more aggresive plan with P-B4. 11 1 2 QR-N1

3

PxP B-B4 N-R4 KR-KB1 R-B2

P-Q4 PXP P-Q5 N/ 3-Q4 N-K3

The position is hopeless for White, despite his extra pawn, be­ cause of the ideal deployment of 102

THE

BENKO GAMBIT

Black's pieces and pawns in the center. Black can wind up the mat­ ter in many ways, for instance : 22 . . P-B5. I even toyed with the

Finally I decided in favor of the shortest and simplest solution. 22 23 N-Nl

.

P-Q6

The pawn is taboo, for after 23 QXP R X B, Black wins a piece. 23 24 P x N 2 5 K-Rl 26 R/ 2-Bl

idea of 22 . . . R X N, as after 23 P X R P-B5 the two connected passed pawns would be irresistible.

N/3 x Bch N x Pch B-QS N-K7!

White resigned. After 27 Q x P N x N 28 R X N R X P 29 Q-B4ch K­ R l 30 R-KN2 R-B7 3 1 R-N l R X P, there is no defense to the mating threats.

103

THE BENKO GAMBIT GAME 4

Black is obviously aiming for P -BS and N-B4.

Palma de Majorca 1971 White: J. H. Donner Black: P. Benko 1 2 3 4 5 6 7 8 9 10 11 12 13

P-Q4 P-QB4 P-QS PXP PXP N-QB3 P-K4 KxB P-KN3 K-N2 N-B3 Q-B2 QR-N1

N-KB3 P-B4 P-QN4 P-QR3 BxP P-Q3 BxB P-N3 B-N2 0-0 QN-Q2 Q-N3 KR-B l!?

14 P-N3

N-NS!

Instead of this, 14 . . . P-BS fails either to 1S P x P or 1S P-QN4. 15 P-KR3 16 N x N

N/ S-K4 BXN

Also 1 6 . . . N X N was t o be seriously weighed. Though White could repel the Knight with 17 P­ B4, this would involve a further loosening of White's King position . P-BS!

17 B-Q2

After this Black stands slightly better. 1 8 P-QN4 is a bit dubious because of 1 8 . . . R-R6. 18 KR-QB1 19 P x P 20 N-Q1

PxP Q-NS RxQ

Drawn

An interesting new idea instead of the common 1 3 . . . KR-N l .

The players agreed on a draw here . However, after 21 B X Q R X R 2 2 R X R R-N 1 , Black regains his pawn with a slightly better position.

1 04

THE BENKO GAMBIT GAME 5

became readily accepted in prac­ tice. Also, Black's N-R3 is not forced, but in this game he was not able to find anything better.

Wijk aan Zee 1970 White: M. Taimanov Black: P. Benko 1 2 3 4 5

6 7 8

P-Q4 P-QB4 P-Q5 PxP PXP N-QB3 N-B3 N-Q2

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 P-N3

10 N X B 1 1 N-K3 12 0-0

This game saw the birth of this new idea, whose aim is to recapture with the Knight after Black plays B X B, thus maintaining the right to castle. 8 9 P-K4

The break 1 2 . . . P-K3 is pre­ mature because of the possible posi­ tional sacrifice of a pawn by 1 3 P x P P x P 1 4 P-K5 ! , destroying the Black center. H aving the opin­ ion that White's Queenside is se­ cure, Black elects to prepare the center break. 13 14 15 16

B-KN2 B X B(?!)

This natural move is far from best. Now after 1 0 N x B 0-0 1 1 N-Q2 N-R3 1 2 0-0, the same position is reached as if Black had refrained from playing B X B and had instead played 9 . . . 0-0 1 0 B X B N X B 1 1 0-0. But there i s an important difference: after the text the White Knight is not compelled to return to Q2 but may go to the more active square K 3 . Due to these considerations, Black played 9 . . . 0-0 against Toran some weeks later, which is a definite im­ provement of the line and which

0-0 N-R3 N-Q2

Q-K2 B-Q2 QR-N 1 P-QN3

Q-B2 Q-N2 N-B2 P-K3!

The purposes of this move are to open the King file and to weaken White's hold on his Q5 square. If 1 6 KR-N 1 then 1 7 P-QR4 would 1 05

THE

BENKO GAMBIT

prevent Black's counterplay by N-N4. 17 18 19 20 21 22

P-QR4 N-B4 Q-B3 NxN KR-Q1 B-B4!

attack. These lines are really beau­ tiful but Bl ack is not forced to enter them. The text continuation should lead to a drawn endgame. But Black is rather short of time and does not play accurately.

KR-K1 Q-R3 N-K4 BXN QR-N1

A correct plan - the trade of Black's strong centralized Bishop is to White's advantage. 22 23 P X P 24 BXB 2 5 P-R3

PxP R-NS RxB Q-N2 27 28 29 30 31

Black eyes the weak points on Q4 and N6. The position is about even. 26 K-R2!?

RXQ N X QP RXR R-R7

RXQ R-R6 R x QP NXR P-B4(?)

R X NP?!

With his last move White pre­ pared a clever combination, but against the correct defense it does not offer much. Black might never­ theless have prepared the recapture of the pawn by 26 . . . K-N2. 27 N-K4!?

The point ! Now on 27 . . . R X R 2 8 N x QP R-B4 2 9 N X R R X R (on 29 . . . P x N 30 Q-N3ch and 3 1 R X R followed by 3 2 Q X N with White much better) 30 N-R6ch ! K-N2 3 1 Q X Pch with a winning

White has only a slightly better endgame which would not be diffi­ cult for Black to draw, given enough time. Even the text move, a mistake, should not lead to the loss of the game ; indeed, several other questionable moves are needed. The simple 3 1 . . . R-R7 or 3 1 . . . P-R4 would have been suffi­ cient. White is not threatening to capture the KBP because the Black QBP may advance rapidly. But now White's Rook on the seventh rank becomes very strong because of Black's error, which was incorrect

106

THE

BENKO GAMBIT

in principle as well as in fact. 32 P-R5 33 N-BS!

chances in a 3-against-2 ending. 39 R-Q7ch 40 R-Q4!

R-R5(?)

Preventing the Black QBP from advancing, as on 33 . . . P-QB5 34 N-N6 N x N 35 P x N wins. Better than 32 . . . R-R5 was 32 . . . R-R7. 33 34 N-N6 35 N-Q7

N-N5 R-R7 P-QBS(?)

Almost all of Black's moves are mistakes. 35 . . . R X BP was better, and neither 3 5 . . . N-Q6 nor 35 N-B3 would have lost immediately. 0

0

0

36 N-B6ch 37 N x Pch 38 N-B6ch

K-Bl K-Kl K-Ql?

The last blunder. 3 8 . . . K-B l 39 R-B7 R X RP 40 R X P would still have offered fair drawing

1 07

K-B l R-B7

The time pressure is over and so is the game, for all practical pur­ poses. The remaining technical task is solved by White without any problems. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

N-K8 N-Q6 N x QBP N-KJ NxN R-Q5 RXP R-NJ RXP R-R2 P-R6 R-RJ K-NJ P-R4 K-RJ R-R4 K-N4 K-B5

K-N l RXP P-N4 N-B7 RXN P-BS P-B6 PxP R-B4 K-R2 R-B5 R-KR5 R-Rl R-Nlch R-KRl R-R2 R-N2ch Resigns

THE

BENKO GAMBIT GAME 6

U. S. Open 1971 White: P. Benko Black: C. Chellstorp 1 2 3 4 5 6 7

8 9 10

P-Q4 P-QB4 P-Q5 PxP PxP N-QB3 N-B3 N-Q2 P-K4 B-K2

N-KB3 P-B4 P-QN4 P-QR3 BXP P-N3 P-Q3 B-KN2 0-0

Usually White plays B X B, but that loses a tempo and helps Black to develop. The text move was in­ troduced by Spanish International Master Roman Toran against me. My present opponent, however, plays better here than I played against Toran. 10 11 0-0

KN-Q2 Q-R4!?

This move was originally my in­ vention in a slightly different posi­ tion (see Kaufman-Benko, game 9). Even so, it really surprised me . After calculating all the variations I was off-balance enough to make an inferior move. The obvious 1 2 N-B4 QB x N 1 3 B X B B X N 1 4 P X B Q X BP 1 5 Q-N3 Q X R 1 6 B-N2 Q x Rch (even 1 6 . . . Q X P ! ? 1 7 Q-QB3 Q X B/ 7 1 8 Q x Q R-R5 is possible) 1 7 K X Q leads to a n unclear position in which Black has two Rooks for the Queen but White has two Bishops and attack­ ing chances. If these lines are not to Black's taste, he can go into an ending with 1 5 . . . Q X Q instead of 1 5 . . . Q x R, in which case the position probably favors Black be­ cause of his protected passed QBP, despite White's two Bishops . The question of White's QRP would be critical: Is it strong or weak? Another possibility is 12 N-N3 B X B 1 3 Q X B Q-N3, when it seems to me that Black has good chances because the White Knight is not well placed on QN3 . This was all very annoying. Look­ ing at the situation from the "wrong" side of the board, it seemed that all the lines favored Black ! 12 B x B?

Certainly not the best (perhaps 108

THE

BENKO

the best is 1 2 N/ 2-N l ! ?) since White admits the loss of a tempo (B-K2 X B) and goes into the older variation, but Black's Queen turns out to be not at all badly placed. 12 13 N-B4 14 Q-Kl

17 B-NS 18 N-K3

Q x N/ 3

Better w a s 1 6 B-N5 to force Black either to lose time to protect his KP or to sacrifice it for compli­ cations. 16

. . . . . .

Black's Knight begins a maneuver to reach Q5, a new worry for White. Now I could feel the same pressure I used to inflict upon my opponents when playing this gambit for Black !

NXB Q-NS N-N3!

Of course not 14 . . . B X N 1 5 P X B Q X P 1 6 B-N2, and White has fine attacking chances, as he usually does in similar situations. If Black is to retain reasonable fighting chances he is obliged to eliminate White's Knight on QB4. Of importance, moreover, is the fact that despite his pawn minus Black does not have to fear trading pieces. With fewer pieces on the board, White will find it too diffi­ cult to organize a successful King­ side attack and may be unable to cope with the ever-increasing pres­ sure on the Queenside. 15 N x N 1 6 N-Ql?

N-Bl

Now that the White Knight has withdrawn its protection of QN5,

GAMBIT

P-K3

Losing patience, I decided to give back the pawn for cou nterplay­ otherwise Black will find it easy to increase the pressure by means of natural and obvious moves. 18 19 P X P 20 Q x Q?!

PxP QxP Drawn

I was originally intending 20 Q-B3, but I now rej ected it because of 20 . . . N-N4 and the Knight will come strongly to Q5 . Neverthe­ less, 20 Q-B3 should have been played as it offers more tactical chances for counterplay than the text move. After I made my move, my op­ ponent offered a draw which I was unwillingly forced to accept, since White has the worse position, After 20 . . . B X Q 2 1 QR-Q 1 (2 1 QR-N 1 R X P) 2 1 . . . P-B3 22 N-B4 B-K4 23 N x B B P X N and White has two weaknesses, the QP and QRP, while Black has only one, the QP. The winning chances, if any, would have been Black's.

109

THE

BENKO

GAMBIT

GAME 7

R X KP 1 7 Q X R B X R 1 8 B-N 5 ! B-B3 1 9 B x B P x B 20 P-QN4 , White would be, in effect, a pawn up because of Black's doubled pawns.

Siegen Olympiad 1970 White: T. Ghitescu Black: P. Benko 1 2 3 4 5 6 7

8 9 10 11 12 13 14 15 16

P-Q4 P-QB4 P-Q5 PXP PxP N-QB3 N-B3 N-Q2 P-K4 BXB 0-0 Q-K2 N-B3?! R-Q1 NXN P-QN3!?

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 P-N3 B-KN2 0-0 NXB N-Q2 N-N3 N-B2 N-R5 RXN

16 17 18 19 20 21 22 23

With this unexpected move White tries to get rid of the pressure on the Queenside and offers to return the pawn. However, Black rejects the offer because after 1 6

PxR Q-B2 B-Q2 Q-B4 B-N5 Q-B2 Q-B4

B xR B-N2 Q-Q2 R-N 1 Q-B1 R-N5 R-N7 R-N5

Drawn by repetition (or overcau­ tion). Must Black accede to the draw or can he attempt to win? I disliked the economical 23 . . . K-B 1 because of P-K5. That left the "ugly" 23 . . . P-B3 24 B-Q2 ! Q-R3 25 Q X Q N x Q 26 P-QR3 , leading to a difficult position: Black has a protected passed pawn but White is a pawn to the good (de­ spite the fact that his pawns are doubled). Who knows which side

1 10

THE BENKO GAMBIT has the better chances? Black could attack the center with . . . P-B4, but then White could obtain coun-

111

terchances against the weak Black KBP and KP.

THE

BENKO

GAMBIT

GAME 8

14 15 R-R3

Toronto 1971 White: Z. Vranesic Black: P. Benko 1 2 3 4 5 6 7 8

P-Q4 P-QB4 P-Q5 PxP PXP N-QB3 N-B3 N-Q2

Disclosing another part of White's plan: to place the Rook on a pro­ tected square and to make possible a future R-KR3 (for i nstance) to aid in a Kingside attack. More exact, however, was the immediate 15 P-R5 N-N5 and now 16 R-R3 .

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 P-N3

15 16 Q-K2

P-K4 B XB 0-0 N-B4 N-K3! P-QR4

B-Q5

In order to prevent the threat­ ened 16 . . . B X N(K3) 17 B x B N­ B5, etc.

At the time this game was played this characteristic Knight move was virtually the main variation for White in this opening. 8 9 10 11 12 13 14

KR-N1

B-KN2 0-0 NXB N-Q2 N-N3 Q-Q2

A committing move which is ob­ viously intended to secure the strong QB4 square for the White Knight on K3 by first driving away Black's Knight on QN6. On the other hand, the move seriously weakens the squares on the QN file, as well as the two Queenside pawns. What will predominate now, strategy or tactics?

16

Q-N2!

Of course, 16 . . . B X N(B3) would win back the pawn, but after 1 7 P x B N X RP 1 8 B-Q2 N-B2, the position would be barely equal and with only meager winning ch ances for Black. Further, because of the absence of Black's King Bishop,

1 12

THE

BENKO GAMBIT

White could try for an attack against the Black King. Black is willing, naturally enough, to win back his gambit pawn, but not if this would mean giving up his positional advantage. The spirit

22 23 24 25

NXQ PXN PXR NXB

NxN RXR RXP PxN

of this gambit requires that Black retain w inning chances after re­ gaining the pawn!

With the text move, Black begins a regrouping of his pieces in antici­ pation of White's following ma­ neuver. 17 18 19 20

K-R1 P-RS N-B4 P-B4?

N-B2 N-Q2 Q-R3

White believes that his prepara­ tions h ave been sufficient for him to launch a Kingside attack, but it is still too early. He has overlooked Black's strong reply. In any case, the position is already difficult for White as his Queenside weaknesses are ripe for exploitation. 20

0

0

0

0

0

0

26 K-N l

QxQ

P-B4!

After this typical breakthrough in the center, White cannot avoid the loss of a pawn.

R-NS!

A very strong move which wins at least a pawn. It is impossible for White to maintain his Knight on QB4 by 2 1 P-QN3 because of 2 1 . . . B X N. Black i s now ready to win back his pawn because this time he does not have to trade his positional advantage for it. 21 N-N6

This i s the endgame Black wanted; all the winning chances are his. It is difficult, if at all possible, for White to hold his position. If 26 R-Q 1 R-N8 and the pin is very unpleasant. On 26 B-Q2 R-N7 27 B-N4 N-R3 , or 27 R-Q 1 N-N4, White is in extremis.

27 P x P 2 8 K-B2

PXP NxP

Now Black will have two con­ nected passed pawns in the center and White will be unable to resist their advance. Instead of White's last move, had he played 28 R-Q l ,

1 13

THE

BENKO GAMBIT

then after 28 . . . R-N8 29 K-B2 N-N4 30 R-K I N-B6, the outcome would be the same. 29 30 31 32 33 34 35 36 37 38

R-Q1 R-Q3 P-R3 P-QR4 8-Q2 P-RS 8-N4 R-QN3 K-K2 PxP

39 8-K1 40 K-83

R-R7ch P-RS

Obviously threatening mate. Of course 41 B X P is answered by 4 1 . . . N-Q7ch.

P-K4 K-82 P-R4 R-N8 R-QR8 N-83 R-RS N-K5ch K-K3 PxP

41 42 43 44

R-N6ch 8XP K-K2 K-K3

K-Q2 R-R6ch P-Q6ch P-Q7ch

White resigned in view of 45

K-K2 R-R8, etc.

1 14

THE

B E N KO

GAMBIT

GAME 9

of his King Bishop in anticipation of Black's . . . B x P.

U. S. Open 1968 White: L. Kaufman Black: P. Benko 1 2 3 4 5

P-Q4 P-QB4 P-QS PxP P-K3?!

9

0

0

0

0

0

Q-R4!?

N-KB3 P-B4 P-QN4 P-QR3

In his books on the openings, Ludek Pachman strongly advocates this move, even giving it an ex­ clamation mark. However, he con­ fines his analysis to continuations after 5 . . . P-K3, which is more akin to the Volga Gambit than to the Benko Gambit. As a matter of fact, I had played 5 . . . P-K3 in a game against R. Gross at the Amer­ ican Open in 1 966. But this time I decided to find out what White would play against the fianchetto, which is a perfectly logical contin­ uation. Furthermore, White's P-K3 may prove to be a loss of tempo if he should later find it desirable to play P-K4. 5 6 N-QB3 7 PxP 8 N-B3 9 P-K4

0

P-N3! B-KN2 0-0 P-Q3

So White has played P-K4 after all. He postpones the development

Black is in no hurry to take the pawn. Instead, he paves the way for the Queen to get to the QR3KB8 diagonal after the exchange of Bishops. Now 10 B-Q2 would not be very good because it would interfere with White's necessary efforts to get a Knight to QB4 via Q2. And 10 B-Q3 would be a serious blunder because of 1 0 . . . N X QP ( 1 1 P X N B X Nch). 10 11 12 13

N-Q2 BxB Q-K2 N-84

BxP QxB KN-Q2 P-B4!?

An original move which is neces­ sary if Black is to maintain his in­ itiative, which he must do since he is a pawn down. The move has two aims: first, it opens lines in the cen-

1 15

THE

BEN KO

GAMBIT

White has castled artificially, but it costs him a pawn. Had he played 23 K-K 1 instead. Black could have continued his offensive by 23 . . . Q Q5, followed by . . . Q-N8ch and . . . Q X RP.

ter and weakens White's QP after . . P X P ; second, the possibility of . . . P-KB5 secures a safe outpost on K4 for a Black Knight. A further advantage of the text move is that White cannot castle here because of 1 4 . . . B X N 15 P x N N-K4, winning a piece. .

-

23 24 B-K3

Q-Q5ch

14 P-B3?

Caught by surprise, White comes up with a passive reply. He should have met the challenge by 14 P x P. 14 15 PXP

PXP

Even here 1 5 N x KP was better, for after the text Black not only gets his K4 square, White is also prevented from castling indefinitely. 15 16 17 18 19 20

NXN R-B1 R X Rch B-K3 N-Q1

N-N3 QxN N-Q2 RXR R-N1 N-K4

24

Black could have recaptured his pawn here with 20 . . . B X P, but there was no hurry. Instead, he in­ creases the pressure in expectation of a better opportunity. 21 K-B1 22 B-Q2 23 K-N1

Forced, since 24 K-R 1 fails against 24 . . . Q x B !, and 24 Q­ K3 N-N 5 ! would lead to an even quicker debacle. . . . . . .

Q X KP

Having regained his pawn Black has succeeded not only in restoring the material balance, he has also preserved - as is often the case in this opening - his positional superiority. It is this fact that de­ cides the issue.

Q-N5 R-B1ch

25 N-B3 26 R-KB1 27 B-B1

1 16

Q-QB5 R-N1

THE

BENKO GAMBIT

A sad necessity. If 27 Q-Q2, then 27 . N-N5 28 R-B4 B X N 29 R X Q B X Q 30 B x B R x P, etc. .

39 40 41 42 43 44 45

.

27 28 N x Q 29 P-QN3

QxQ N-Q6 N-N5

P-N3 K-N2 K-B3 R-Q1 NXB R-K1

B-B7! R-R7 N-K5 P-Q4 B�Q5 BP x N

Black's positional superiority has led to the win of a pawn and with it the game. White, however, does not yet give up the fight and re­ sorts to tactical twists to try to save the game. 30 31 32 33 34 35 36 37 38 39

B-N5 R-Q1 N-B4 K-R1 R-KB1 R-K1 N-K6 B-R4 R-KB 1 B-Q8

N x QP P-K3 B-Q5ch N-B6 R-KB1 P-K4 R-B7 R X RP R-R1

Black was threatening to weave a mating net with 45 . . . P-R4, and if 46 R-KB l N-Q7ch, etc. Black's center is not something you see every day.

White's efforts have apparently not been in vain as he seems to have considerable counterchances . Black's next move, however, shat­ ters all his hopes.

45 46 K-N4 White resigned.

1 17

R-B7ch RxP

THE

BENKO GAMBIT GAME 10

1 2 Q-K2

Palma de Majorca 1971

This move has proved to be good enough, as demonstrated in my game against Kaufman (see the preceding game). But I have also played 1 2 . . . QN-Q2 successfully, i.e. against Farago at Vrnjacka Banj a 1 97 1 .

White: L. Portisch Black: P. Benko 1 2 3 4 5

P-Q4 P-QB4 P-Q5 PXP P-K3

N-KB3 P-B4 P-QN4 P-QR3

1 3 P-QR4

A cautious move that has been recommended by Pachman, but it is actually intended for use against the system with P-K3 for Black, as Pachman notes in his theoretical works. It is true that at the time Pachman made his recommenda­ tion, the lines undergoing analysis were those within the framework of the Volga Gambit, which is char­ acterized by an early Black P-K3. 5 6 7 8 9

N-QB3 N-B3 PXP P-K4

KN-Q2

Q x Qch

This exchange is already forced as White was threatening N-N5. But forced or not, the move is a good one, now that White has weakened his Queenside. 14 K X Q 15 QR-N1

N-R3

Necessary. On the natural 1 5 N­ B4 KR-N l followed by . . . R-N5 is very strong.

P-N3! P-Q3 B-KN2 0-0

White has been induced to play P-K4 with a loss of tempo. Part of the reason was to open a path for his Queen Bishop. 9

. . . . . .

Q-R4!?

An interesting adventure but it is not certain that it is necessary. 10 N-Q2 11 BXB

BXP QXB

15

. . . . . .

P-B4!?

Draw agreed. The position is rather difficult, but Black's chances are no worse than White's.

1 18

THE

BENKO

GAMBIT

GAME 1 1

White is already forced to make artificial moves, since 1 5 P-QN3 would be met by 1 5 . . . N-R6 ! The power of the two Black Bishops is striking.

Wijk aan Zee 1970 White: F. A. Kuypers Black: Pal Benko 1 2 3 4 5 6 7 8

P-Q4 P-QB4 P-QS PxP P-K3 N-QB3 PxP B-NSch

N-KB3 P-B4 P-QN4 P-QR3 P-N3 P-Q3 B-KN2

15 16 P x P

The position is ripe for a central explosion. Also good here was 1 6 . . . N X P in connection with a later . . . N-Q5.

This is a questionable place for the Bishop. Black is able to parry the check with a useful developing move and he will later exploit the White Bishop's position to gain tempi. 8 9 10 11 12

KN-K2 0-0 P-K4 B-Q3

17 R-N1 18 P X P 1 9 B-Q3

P-Q4 PXP

19

P-QS

KN-Q2 0-0 NxP N-B2 N-K4

Of course this is not a permanent square for the Knight, but Black is of the opinion that if White wishes to drive the Knight away by means of P-KB4, White's own position will be weakened. 13 B-B2 14 P-B4 15 R-B3

P-K3 PXP

B-QR3 N-BS

. . . . . .

This advance was not urgent and in fact turns several vital squares over to the opponent. Black would have had an excellent game after 19 . . . B-N2 20 B x N (20 P-BS is inferior because of 20 . . . N-K4) 20 . . . P X B 2 1 Q x Q QR X Q 22

1 19

THE

BEN KO

K5 Q-Q4, but in time pressure he feared 29 P-B5 with an attempt at complicating the game.

R-B 1 R-Q6, despite the exchange of Queens and Black's doubled pawns. In order to force the rapid advance of the central pawns, how­ ever, best was 19 . . . N-Q3 , and only after the point K5 is under Black's control is he ready to push the pawns. 20 21 22 23

N-R4 P-QN3 BXB B-R3

28 29 B-R3

B-QS N-QB4?

White would have been satisfied with a draw, but Black had no desire to reconcile himself to this because of the excellent position he obtained in the opening, although the situation is now double-edged. Another factor was that he had not lost a single game with this opening, until this one.

N-K3 N-Q3 RXB R x N?!

Black was intending this Ex­ change sacrifice, although it is double-edged. It was possible to play 23 . . . Q-R4 to defend the QBP, and then to try for . . . P-B5 . 24 P X R

GAMBIT

30 N-K2?

P-BS

The two passed pawns really look dangerous, but one of their sup­ porters - the Knight on Q3 - is pinned. This is not Black's only problem, however - he is also fighting a time shortage. 25 26 27 28

QR-Bl N-N3 K-Rl B-N2

P-Q6 B-QSch B-B4

30

To be considered was the return of the Exchange by 28 R x QP P x R 29 B X B N X P, with a compl icated position. After the text, Black has an opportunity to secure control of the center by 28 . . . Q-R1 29 B-

. . . . . .

P x N?

Now th� game is lost, and just at the moment when Black had good winning prospects with 3 0 . . . N / 3 K5 ! , e.g. 3 1 N X B Q x N 32 B X N Q X B and Black obviously stands better, and on White's attempt to eliminate the Black passed pawns by 3 1 R X BP P X N 32 Q x P, the

1 20

THE

BENKO

reply 32 . . R-K l ! leaves White's position critical because of the situ­ ation of his King and Queen. .

34 R-K1

Q-R4

The last trick: on 35 R X P?? Q­ K8ch would lead to mate.

N-Q6

31 Q x B

Even here 3 1 better.

GAMBIT

.

. . R K l was -

35 36 37 38

Q-N3ch Q-N4 P-RJ B-N2

N-B2 Q-Q4 P-R4 N-Q3?

32 R x N!

38 . . R-K6 would have prolonged the game. .

Of course, the rest easy. 32 33 Q x P

IS

relatively

39 Q-B3 40 Q-R8ch 41 Q-B6ch

PXR R-K1

121

N-B4 K-B2 Resigns

THE

BENKO

GAMBIT

GAME 12

this will become the source of his future troubles. Black's Queen Bish­ op, on the other hand, will exert strong control of its diagonal. Black had nothing to fear in 1 2 B X B, for after 12 . . . R X B followed by . . . Q-R 1 and . . . R-N 1 , he would have a dominating position both in the center and on the Queenside, with good compensation for the sacri­ ficed pawn.

Atlantic Open, N.Y. 1969 White: D. Blumin Black: Pal Benko 1 2 3 4 5 6 7 8

9 10

P-Q4 P-QB4 P-QS PxP P-K3 N-QB3 B-B4 KN-K2 0-0 PXP

N-KB3 P-B4 P-QN4 P-QR3 P-N3 P-Q3 B-KN2 0-0 QN-Q2 N-N3

12 13 R-N1

Another inaccuracy. 1 3 R-K 1 was more urgent.

Black tries to exploit the position of White's B ishop on QB4 to gain tempi for the development of his pieces. 11 B-NS

R-N1

13 14 P-B4

N-NS!

A further weakening of White's position, although it is true he had to parry the threat of 14 . . . N-K4, after which 1 5 B-N5 would have been answered by 1 5 . . . B X B 1 6 N X B N x QP !

BxP

14

1 2 B-B6?

The Bishop has found an appar­ ently cozy spot on QB6, but in fact

KB X N

It might seem surpnsmg for Black to willingly exchange his val­ uable King Bishop, but the move is actually a logical exploitation of White's weakened position. The White Queen Bishop is doomed to an inferior role, without any op­ portunity to control the long diag­ onal.

1 22

THE 1S P X B 16 RXR 17 Q·R4

BENKO GAMBIT Q-N5. But he has no defense at all against the text move.

N-BS QxR

A s can b e seen, White's position is rather precarious. After 1 7 R-B3 Black's invading Queen ( 1 7 . . . Q­ _N8) would be decisive. It is to be noted that in all variations White's Bishop on QB6 is completely out of action. 17 1 8 P-K4

Q-N3 N-Q7

The same move would have been played after 1 8 R-B 3 , after which the Knight would have been taboo, e.g. 19 B X N Q-N8ch, etc. If now 19 B X N B X N, Black threatens 20 . . . P-B5ch. 19 R-Q1 20 R X N 2 1 R-Q4

BxN P-BSch Q-N8

In case of the seemingly strong . . . P-K4, White would save his skin with 23 P X P e.p. P X P 24

22 Q-R3 23 P-RJ 24 B-R4

Trapping the Rook. But Black had other winning continuations, for instance 24 R-R l . 2 S K-B2 26 B X N 2 7 K-N3

N-B7 Q x Bch

The fact that there are Bishops of opposite colors is of no signifi­ cance whatever, the more so be­ cause White's Queen Bishop could have been won by 27 . . . R-N8. 27 28 29 30 31 32 33

P-BS R-Q2 K-R2 K-N1 R-Q1 R-Q2

22

123

R-N1 N-K6 B-Q6

White resigned.

Q-K7 BxP Q-K6ch Q-BSch R-N8 Q-N6 Q-K8ch

THE

B E N KO

GAMBIT

GAME 13

American Open, California 1967 White: R. Laver Black: Pal Benko 1 2 3 4

P-Q4 P-QB4 P-Q5 PxP

N-KB3 P-B4 P-QN4

On several occasions the pawn sacrifice was not accepted, my op­ ponents trying 4 P-B3 or the timid 4 P-QN3 . Naturally, White cannot hope to obtain any advantage in the opening by such measures. For some analysis on the gambit de­ clined, see the appropriate chapter in the theoretical section of this book. 4 5 6 7 8 9 10 11 12

PXP P-KN3 B-N2 N-QB3 N-B3 0-0 R-N1 P-QR3

P-QR3 BXP P-Q3 P-N3 B-KN2 0-0 QN-Q2 Q-R4

12 13 R-K1 14 B-Q2

KR-N1 N-K1 B-B5!

Summing up the opening, Black has considerable pressure on the light squares and against White's Queenside pawns, while White has no effective counter. With his next move White tries to find counter­ play on the Kingside. 15 P-R3

Q-R3

If 1 5 . . . R-N6 1 6 B-N5 with complications, the threat being 1 7 N-Q2. It would be a mistake for White to try now to trap the Bishop with 1 6 P-N3 because of 1 6 . . . Q x P, etc.

(See diagram next column.)

White is trying to free his posi­ tion with P-QN4, but Black never allows it. In fact, he is even able to take advantage of the weakened light squares on White's Queenside.

16 N-KR2

B-N6(?)

More direct was 1 6 . . . R-N6 and doubling or even tripling on

1 24

THE

BENKO GAMBIT

the QN file. The Bishop only in­ terferes with Black's open file. 17 Q-B1

20 21 B-B3

Forcing White's next move, as on 22 Q-B4, B-B7 wins the QN pawn.

N/ 2-B3

Black is confusing several plans. The idea of the text move is to transfer the Knight from K 1 to QN4. For this reason I did not want to move that Knight to KB3 . How­ ever, one of the points of 1 6 . . . B-N6 was to free the QB5 square for a Knight, but obviously both plans are not simultaneously play­ able. Furthermore, I was concerned about White's possibilities after get­ ting his Knight to KN4 and I thought it best to return one of my Knights to KB3 to meet that pos­ sibility. Actually, White's action turns out to be less dangerous than I feared. 18 N-N4 19 P x N

R-R2 Q-B1!

22 B X N 2 3 K-N2

BXB

White tries to work up an attack along the KR file, but this is not to be feared by Black since his unop­ posed King Bishop simultaneously defends the Kingside and exerts pressure on the enemy Queenside. 23 24 25 26 27

R-KR1 R-R4 Q-Q2 B-K4

B-N2 R/ 2-N2 Q-B1 P-R3 B-BS

It is not too late to admit the mistake on the 1 6th move. Now White's pieces are tied to the de­ fense of the Queenside.

NxN N-B3

Switching back to the plan N-Q2N3-B5. I was also hoping to pro­ voke 20 P-N5, blocking White's own Bishop. 20 B-NS

A strategical error leading to the exchange of White's valuable Queen Bishop. Better, even though the Bishop would be blocked, was 20 P-N5.

28 R-R2?

It was better to think at once of defense with 28 R/ 4-R 1 , but White was probably still thinking of dou­ bling Rooks on the KR file. It was now possible for Black to win back his pawn by 28 . . . B-R7, but I felt that with opposite-color Bishops remaining, the win would be problematical. I preferred to maintain my positional superiority.

1 25

THE 28 29 R/ 2-R1

BENKO GAMBIT

R-N6

Now that White's Kingside "at­ tack" has come to naught, Black is ready to roll him up in the center. 29 30 P-B3

Q-B1 P-K3!

This breakthrough in the center quickly decides the issue. Suddenly everything in the White camp is hanging. The attempt to win the QN pawn by 30 . . . Q-N2 would misfire after 3 1 B-B2 R X NP? 32 R X R Q x R 3 3 R-QN 1 , etc. 31 P x P 3 2 B-Q3?

This is not satisfactory, but the threats of 32 . . . P-Q4 and 3 2 . . . B X N are not easy to meet. If 3 2 KR-Q 1 P-Q4 3 3 B X QP B X N and Black wins a piece. White's best move seems to be 32 N-Q 1 which, although passive, seems to hold for a while, at least. 32 33 Q X B 34 RXR 3 5 N-K4

BXB R X NP RXR P-BS!

This was probably overlooked by White. The point is 36 Q x QP R X Pch 37 K-R3 R X N 3 8 Q-N8ch Q-K 1 and wins.

QXP 36 37 38 39 40 41 42 43

Q-Q1 N-B2 R-B1 QXP QxP RXR K-R3 P-NS

P-Q4 RXP Q-K6 B-QS R X Nch Q x Rch Q-N8 B-K4

White resigned in view of un­

avoidable mate.

1 26

THE BENKO

GAMBIT

GAME 14

16 B-Q2 17 P-NJ 18 P-K4

Sarajevo 1967 White: M. Vukic Black: Pal Benko 1 2 3 4 5 6 7 8 9 10 11

P-Q4 P-QB4 P-Q5 PxP PxP N-QBJ N-B3 P-KNJ B-N2 0-0 Q-B2

White has succeeded in getting rid of Black's strong Queen Bishop, but this means a weakening of his King's position.

N-KBJ P-B4 P-QN4 P-QR3 BXP P-QJ P-N3 B-KN2 0-0 QN-Q2 Q-NJ

19 K X B 20 R-K1 21 K-N2

It is difficult to determine the best place for the Queen. I have tried 1 1 . . . Q-R4 as well as 1 1 . . . Q-B2 and the text move with­ out reaching a firm conclusion. All three moves are examined in detail in the theoretical section of this volume. 12 R-Q1 13 R-N1 14 B-N5

KR-N1 N-K1 Q-Q1

The Queen returns, not unwill­ ingly, to its original square, having permitted the King Rook to get to QN l . But now the KP required protection and White thus gains a tempo. 15 B-KB1

N-B2 N-NJ BxB

P-RJ

1 27

Q-Q2 K-R2 P-KJ

White has managed to secure his Queenside; therefore Black endea­ vors to create new problems for him in the center. In addition to the text, many other plans were available to Black, for instance pawn breakthroughs by P-B5 or P-B4. 22 23 24 25 26 27

PXP N-K2 N-B4 RXP NxN P-QR4

NxP P-Q4 PxP N-Q5 PXN

THE

BENKO

A blunder allowing Black to re­ gain the gambit pawn and achieve an overwhelming position. But White was in difficulties in any case, the threat being 27 . . . R­ QB I . If then 28 Q-N2 P-Q6 . If White tried to blockade the pawn by 27 N-Q3, Black would have played 27 . . . R-QB l followed by 28 . . . Q-N4 or 28 . . . Q-B3 29 P-B3 Q-B7 . 27

. . . . .

.

Q-N2!

A double threat against the Rook ( . . . P-B4) and the QR pawn which White thought he was saving. 28 P-83 2 9 QR-K1?

NXP

Desperation. His expected coun-

128

GAMBIT

terplay fails t o materialize. 29 30 31 32 33 34 35 36

QxQ R-K7 R/ 1-K2 BxN RXP K-81 R x 8ch

QxP RXQ R-N7 N-86 PXB R/ 1-R7 P-N4

Not even this can help. On 36 R/ 2-K7, the answer is 3 6 . . . R-N8ch. 36 37 N-K6ch 38 N-Q4 39 N X R White resigned.

KxR K-83 RXR R-R8ch

THE BENKO GAMBIT GAME 15

but this is premature here. White should prepare with P-KR3 .

Vancouver 1971 White: J. Aspler Black: Pal Benko 1 2 3 4 5 6 7 8 9

P-Q4 P-QB4 P-QS PXP PxP N-QB3 N-B3 P-KN3 B-N2

12

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 P-N3 B-KN2

.

N-NS!

Not only preventing P-K5 but also threatening to come strongly into K4 aiming at Q6. 13 Q-B2 14 P-KR3

Fianchettoing White's King Bish­ op may be considered one of the main lines in this gambit. 9 10 0-0 1 1 R-K1 12 P-K4

. . . . .

0-0 QN-Q2 Q-N3

KR-N 1

White is obviously worried about his KB2. After, for instance, 1 4 R-N 1 o r 1 4 P-N3, the move 1 4 . . . P-BS ! threatens simultaneously 1 5 . . . B X N and 1 5 . . . Q x Pch. But the text move allows Black to gain another tempo. 14 15 N x N 1 6 P-N3?

N/ 5-K4 NxN

I t was absolutely necessary to protect Q3 by 1 6 R-Q 1 (but not 1 6 B-B 1 ? N-B6ch). After the error of the text, Black's whole position quickly collapses. 16 17 R-Q1

Preparing the thematic break­ through in the center with P-K5,

N-Q6 P-BS!

The threat is 18 . . . P X P. If 1 8 P X P Q X Pch 1 9 Q X Q N X Q 20 K X N B X N, and White's Queen Rook is trapped.

1 29

THE

BENKO

If now 23 KR-B 1 N-Q6, winning the Exchange anyway, but the K8 pawn is also hanging. If 23 QR-8 1 , then 23 . . . 8 X R 24 8 X 8 N-R7 wins two Exchanges (25 N X N B X Q). White i s hardly to be blamed for his next desperate move.

As we can seem the Black pieces work together like a well-oiled machine. Black's position is beauti­ ful, is it not? 18 B-K3

GAMBIT

Q-NS

There is no hurry to play 1 8 P X P because after 1 9 P x P Q x P 2 0 Q x Q R X Q 2 1 R X N, White escapes. But with the text Black conveniently improves his position. 19 B-Q2 20 R-KB1

Q-B4

Losing the Exchange, but so does 20 B-K3 P X P 2 1 P x P Q x N 22 Q X Q B X Q 23 R X B R X R 24 R X N. 23 Q-R3 20 21 P X P 2 2 Q-N2

BxN

PXP O f course, 23 . . . B X R 2 4 Q x R R X Q 2 5 R X Rch 8-B 1 2 6 B-R6 would be too naive, even for me.

N-NS R-QBl!

Again, there is no hurry to win material. After 22 . . . B X R 23 B X B R X R 24 Q X R R-QB 1 25 B­ QB4, the game is lost for White, of course, but longer resistance would be possible.

1 30

24 B X B 2 5 KR-B1 26 B-B1 White resigned.

QxB Q-QS R-B7

THE

BENKO

GAMBIT

GAME 16

U. S. Open 1969

1 2 B-Q2 seems more natural, but White obviously wished to hinder Black's maneuver N-K 1 -B2.

White: A. Denker Black: Pal Benko 1 2 3 4 5 6 7

P-Q4 P-QB4 P-QS PxP PxP N-QB3 P-KN3

12 13 KR-K1 14 QR-N 1 15 B-Q2

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3

KR-Nl R-R2 R/ 2-N2 P-BS

White temporarily abstains from playing P-K4 and first develops his King Bishop. I must confess that I would be at a loss to decide which continuation to choose as White if someone played this variation against me. 7 8 B-N2 9 N-R3?!

P-N3 B-KN2

The text, instead of the more natural 9 N-B3, was a small sur­ prise to me. In similar positions in other openings, with the fianchetto of the White King Bishop, I too have played this move on occasion. But here I do not see its advantage over N-B3. 9 10 0-0 1 1 Q-B2 12 B-NS?!

QN-Q2 0-0 Q-R4

131

With this well-considered deci­ sion, Black gives the game new character. White is now obliged to try to get rid of the annoying pres­ sure on the QN file. 16 P-QN4 17 R x P

P x P e.p.

If 17 P x P, then 17 . . . R-QB l puts White's Queen in an unpleasant pin on the QB file. After the text, however, Black's Bishop gets an ideal post on QB5, which is suffi­ cient compensation for the sacrificed pawn.

THE 17 18 R X R 19 N-KNS

BENKO

22 23 24 25 26

B-BS RXR

1 9 N-B4 seems more prudent. 19 20 R-QB1 21 B-R3

GAMBIT NXP BXN PxB BxP B-B3

NxN BXN Q x BP QxP

White chooses the simplest solution.

Q-R3 R-N1 N-B4?

27 Q-K3

Instead of this, Black could have regained his pawn with advantage by 2 1 . . . N-N3 ! , since the QP is loose. If 22 P-K4 P-R3 23 N-B3 B-Q6 and wins the KP. In case of 22 R-N 1 , Black would reply 22 . . . R-R l . After the recapture of the pawn, White's weaknesses - QR2 and K2 - would remain while Black's pieces would be excellently posted. 22 B-K3

White grasps the initiative in order to liquidate into a drawn position following the return of the extra pawn. 1 32

White has chosen to avoid all the possible complications and makes no attempt to win the game by playing 27 Q-B7. However, because of the three-against-four pawn con­ figuration, winning prospects are few. A possible line is 27 . . . R­ K l ! ? (27 . . . Q-N3 is safe) 28 R-B6 Q x P 29 R X B Q-Q8ch (on the quiet 29 . . . B-B5, the unbelieva­ ble 30 P-N4 ! would save the game) 30 K-N2 R-K8 3 1 N-B3 B-Q4, and now on 32 B-N4 P-R4, or on 32 B-K6? ! R-N8ch 33 K-R3 B X Bch. 27

. . . . .

BXN

.

Draw

THE

BENKO

GAMBIT

GAME 17

Queen Knight i s deprived o f its natural developing square, QB3, and the King Knight must be developed via KR3 to avoid the loss of the QP. Nevertheless, I thought the idea was worth trying.

Canadian Open 1971 White: Pal Benko Black: J. Berry 1 2 3 4 5 6 7 8 9

P-Q4 P-QB4 P-QS PXP PxP P-KN3 B-Nl B·Ql B-QB3!?

N-KB3 P-B4 P-QN4 P-QR3 BXP P-Q3 P-N3 B-KN2

9 10 11 12 13

N-KR3 0-0 N-B4 P-N3

0-0 QN-Ql N-N3 Q-Ql

Necessary, to prevent 1 3 R5. 13

N-

R-R2

Black has so far built up his posi­ tion logically, but the text is prob­ ably too slow. Sharper seems 1 3 . . . P-N4, and i f then 1 4 N-Q3 of course the QP is lost, but after 1 4 N-KR3 interesting positions occur. 14 P-KR4

This was a new idea at the time of this game. The system was tried and recommended by Dr. Elod Macskasy, who told me of his suc­ cess with it after discouraging re­ sults initially with the gambit. The plan is to neutralize the long black diagonal, which is normally powerfully exploited by Black's King Bishop. The plan is reason­ able and well motivated but has two serious drawbacks: Black's 133

Denying Black the above-men­ tioned possibility, but even now Black can try 1 4 . . . B-R3 to make the future of White's Knight on B4 insecure. 14 15 R-K1

R-N1 P-BS

This looks like the only way to open lines on the Queenside, but

THE

BENKO

White is developed enough to meet it. 16 17 18 19 20 21

B-Q4 N-B3 PXP Q-Q3 P-QN4 P-NS!

KR-N1 Q-Q1 P-N6 R-R7

26 R X R 2 7 R-N4

RXR N-K4

Admitting his failure to win the pawn. After 27 . . . N X NP, the strengthening of the pin on the QN file by 28 Q-N3 wins.

R-B2 PXP B-QB1 R/ 2-N2 Q-K1

This pawn, weak or strong, will decide the issue. White always man­ ages to find a way to defend it, but Black should probably try 2 1 B-Q2 anyway. 21 22 23 24 25

GAMBIT

QN-Q2 N-K4 B-Q2 N-BS Q-Q1

Not 25 . . . N X NP? 26 R X R and 2 7 R X N.

1 34

28 Q-N3 29 B x N! 30 N-Q3

Q-QB1 PxB

After the previous exchange, Black's position collapses because of the threats of P-Q6 and N x P. In desperation he plays . . . 30 31 32 33 34 35 36

NxP BxN BXB R-QB4 R-B6 Q-B4

P-KS NxN B-B4 PxB Q-Q1 B-QS Resigns

THE BENKO GAMBIT GAME 18

ding. But Black shows that the idea is Jess than perfect.

New Hampshire 1968 White: I. Barlay Black: Pal Benko 1

2 3 4 5 6 7 8

P-Q4 P-QB4 P-Q5 PXP PXP N-QB3 P-KN3 B-N2

10

P-R3!

N-KB3 P-B4 P-QN4!? P-QR3 P-N3 BxP P-Q3 B-KN2

The course of this game, begin­ ning with the following maneuver, is quite different from those pre­ viously seen with this opening, il­ lustrating once again the wide range of possibilities within the frame­ work of the gambit. 9 Q-R4ch!!?

The start of an interesting Queen maneuver, which, however, fails to prove its worth. Although the Queen check came as a surprise, it was a pleasant one. 9 10 Q-R4

. . . . . .

QN-Q2

The aim of the Queen maneuver is now obvious. White makes plans for a Kingside assault in the event Black castles, or at least tries to frighten Black into postponing cas135

The situation changes at one stroke ! The hunter has become the hunted. White's Queen is constant­ ly exposed to the threat of . . . P-N4, preventing his King Knight from developing naturally to KB3 or KR3 . 1 1 P-B4 12 P-K3 13 N-K2

QR-N1 Q-R4 B-Q6!

Exploiting the absence of the enemy Queen, Black deploys his Queenside forces offensively. White is already in sorry straits. The na­ tural 14 0-0 fails against 14 . . . P-N4 1 5 Q-R3 ( 1 5 P x P P X P 1 6 Q x P B-KR3 1 7 Q-R4 B X Pch, win­ ning the Queen) 1 5 . . . P-N5 1 6 Q-R4 N-B 1 1 7 P-B5 B X P 1 8 R X B N-N3, trapping the Queen.

THE 14 K-B2 15 P X P 1 6 Q-R4

BENKO

2 3 B-K4

P-N4 PXP

A sad retreat, but White had no other choice since 16 Q x P B-KR3 wins the Queen as before. 16 17 N X Q 1 8 K-K1

QxQ N-NSch N X RP

Black has recaptured his gambit pawn with an overwhelmingly su­ perior position. White's forces are undeveloped and his pawns scat­ tered. Black's pieces, on the other hand, are ideally posted . It follows, therefore, that White cannot hold out for long. 19 N/ 4-B3 20 K-Q1 21 P-R4

N-K4 N-BS

A desperate try, since the QN pawn must fall anyway (2 1 P-N3 B x Nch, etc.). 21 22 BXN

GAMBIT

N X NPch RXB

Because of the threat of . . . R X N or . . . B X Nch. 23 24 N X B

The apparently more effective 24 . . . R-N5 does not win a piece: 25 R-QN 1 R X B 26 R-N8ch K-Q2 27 R X R and 2 8 R X N. 25 R x Rch 2 6 R-N 1

BXR

Owing to the threat of . . . R X N or . . . N-B7ch, there was no way for White to avoid the loss of a piece. 26 27 28 29 30 31 32

K-Bl R-N8ch B-BSch P X Pch B X Pch RXB

White resigned.

136

B x Nch N-NS

N-B7ch RxN K-Q2 P-K3 PXP KXB N-Q6ch

THE

BENKO

GAMBIT

GAME 19

Palma de Majorca 1968 White: R. Calvo Black: Pal Benko 1 2 3 4 5 6 7

P-K4 P-Q4 P-QB4 N-QB3 B-K2 B-N5 P-Q5

P-KN3 P-Q3 B-N2 N-KB3 0-0 P-B4 15 B X B

The opening has begun as a Pirc Defense, but by transposition we have reached a variation of the King's Indian, one that became popular after the 1 966 Piatigorsky Cup Tournament in Santa Monica. To avoid all the well-analyzed lines, I decided to employ my favorite pawn sacrifice, although the posi­ tion is slightly different from those usually arising in the Benko Gam­ bit.

M y opponent had thus far seemed unimpressed by my pawn sacrifice, playing without hesitation. I was quite surprised when Larsen asked me after the game if I knew that Calvo had played the same opening with Black against Grandmaster Filip in the Havana Olympiad in 1 966. If, instead of the text, 15 B-N3, Black could capture the Bishop after 1 5 . . . N-R4, while if 1 5 B­ Q2? B X B, followed by 16 . . R X P. Therefore, White was com­ pelled to exchange Bishops, even though the position of Black's Queen is improved. .

7 8 9 10 11 12 13 14

PXP PXP N-B3 QR-N1 0-0 Q-B2 B-KB4

P-N4!? P-QR3 BXP Q-N3 QN-Q2 KR-N1 P-R3 P-N4!

15 1 6 B-K3

QxB

More prudent was 1 6 B-Q2, since it is only Black who can profit from the maneuver White seems to be provoking. 1 37

THE 16 17 B-Q2 18 N x N 1 9 P-B4?

BENKO

N-NS N/ 5-K4 NXN

2 3 R-N3 24 K-B1

Obviously, White was now ex­ pecting 1 9 . . . N-N5, after which he was willing to sacrifice the Ex­ change: 20 P X P B-Q5ch 21 K-R 1 N-B7ch 22 R X N B X R, and then either 23 P-K5 or 23 P x P would lead to unclear complications. After 19 . . . N-N5, even the immediate 20 P-K5 is possible. However, Black has a simple way of avoiding all White attacking lines. 19 20 Q x Q

GAMBIT

Q-Q6!

Exchanging Queens means, of course, no attack for White, but things would be much worse after 20 Q-B 1 N-B5 21 R-Q1 N X B 22 R X N Q-K6ch 23 K-R 1 B x N, win­ ning a piece.

Of course, 24 R X P is met by 24 . . . B-R3 , and 24 B X P is impos­ sible because of 24 . . . B X N. In this interesting position, Black is a pawn down and his KN pawn is hanging. Nevertheless, he has the better prospects because of the ex­ cellent placement of his pieces and the great pressure he exerts on the Queenside. There is no hurry to take the QN pawn since it cannot be saved in the long run. 24

20 21 P X P

NXQ P x P?!

Better was 21 . . . B-Q5ch 22 K-R 1 , after which 22 . . . N-B7ch, winning the Exchange, would be unclear, but 22 . . . P x P would lead to the same position as in the game but with White's King in the corner. 22 R-B3

P-BS

138

K-B1!

. . . .. .

B X N!

The KN pawn could not be pro­ tected by 24 . . . P-B3 because the vital King Bishop would be suffo­ cated. 25 B x B 26 P-KR4

P-B3

White tries to obtain some coun­ terplay. The other possibility was 26 R-K3 (26 P-QR3? R X RP) 26

THE

BENKO

. . . R X RP 27 R-K2 with an at­ tempt at passive defense. But this would be very difficult after 27 . . . R-N6, threatening, as in the game, N-B4-R5, etc. 16 17 18 19 30

R-R3 RXP R-R8ch R-B8

PXP R X RP R-N6 K-B1 N-B4

GAMBIT 31 R-Kl 31 R X P 3 3 R-B7

O f course 3 3 B X N R/ 6 X B would be hopeless with two Black Rooks on the seventh rank. But now the Knight is still working. 33 34 R-Ql

Taking the pawn at once leads to only a drawish position after the exchange of a few pieces. After the text, threatening . . . N-R5, the sit­ uation is entirely different as the Bishop will also be attacked. White's best chance for a longer resistance is offered by the sacrifice of the Exchange by 3 1 R x N P x R 32 P-K5, etc., to get rid o f Black's lively Knight.

139

N-RS NxP

N-Q6

If 3 4 R-R l R X B, etc., or 34 R-K3 R-N8ch 3 5 B-K l R-B7ch 36 K-N l R-Q7 ! , etc. 34 35 36 37 38 39

B-Q4 R-Kl K-Nl K-Rl K-R3

White resigned.

N-B4 NxP N-N6ch N-K7ch R-Nl R-R6ch

THE

BENKO GAME

White: H. Lyman Black: Pal Benko P-Q4 P-QB4 P-QS N-Q2?!

12 13 14 15 16 17 18

N-KB3 P-B4 P-QN4

This way of declining the gambit had not previously been seen by me. It appears no worse than other ways. The move prepares P-K4 while protecting QB4 and leaving the long diagonal open for the Queen Bishop. 4 s

6 7 8 9 10 11

P-K4 BXP P-QN3 B-N2 KN-B3 0-0 R-K1

20

White has to make further preparations, e.g. Q-B2 and QR-Q l .

New York 1970

1 2 3 4

GAMBIT

BxN N-B4 PXP BXB Q-Q2 N-NS

KN X P NXB B-N2 PXP KXB Q-B3

This is another symptom of im­ patience, but White was in a diffi­ cult situation anyway with a linger­ ing and ultimately thankless de­ fense. Black was threatening 1 8 QR-Q 1 , freeing his Knight. 18 19 P-N3

PXP P-Q3 P-N3 B-KN2 0-0 QN-Q2 N-N3 R-N1

N-BS!

There was no better move, since 1 9 N-K4 is answered by 19 . . . B X N 20 R X B Q X Rch.

Black tries to get play on the QN file, as usual in this type of position. Since Black has not sac­ rificed a pawn, the game may be considered fully equalized. 12 P-KS?

This advance here is a serious blunder, since it fails to take into consideration the fact that Black's last move has actually prevented it. The plan itself may be good, but

19

. .. . . .

N-R6ch!

A petit combinaison which aims at forcing the White King from his castle.

1 40

THE BENKO GAMBIT 20 21 22 23 24

NxN Q-N2ch K-Bl K-K2 Q-B6

Not only winning a pawn but also activating the Queen. Of course, there is no danger for Black on the open KR file because Black's attack will succeed first.

Q-B6 K-Nl Q-N7ch QxN B-R3

Opening a file on the Queenside to prevent the White King from hiding there. 25 K-Q2 26 P x B 2 7 QR-Bl

28 R-KRl 29 QR-KN l

Q-N7 Q-N2

White resigned. There is no de­ fense against the penetration of Black's heavy pieces.

BXN R-N3 Q x RP

141

THE

BENKO GAMBIT

BIBLIOGRAPHY

A list of the material used in the preparation of this volume.

BOOKS:

Saltsjobaden Interzonal 1 948 (V. Pirc) Trencanske Teplice 1 949 (L. Pachman)

Benoni Counter Gambit (Zechiel and Crane) Volga-Gambit (Berglund and Akvist) Chess Informants 1 - 1 4 (Matanovic, ed.) Kapfenberg 1 970 Title Chess: U. S. Championship 1 972 (B. Hochberg)

International Grandmaster Tourna­ ment: Zurich 1 953 (D. Bronstein) Indische Spezialitaten (M. Euwe) Geschlossene Spiele (L. Pachman)

PERIODICALS AND BULLETINS:

Chess Life & Review The Chess Player Zagreb 1 955 Goteborg 1 955 USSR Championships 1 95 5-57 Havana 1 960 Budapest Championship 1 968 Malaga 1 969 Ljubljana 1 969 Buenos Aires 1 970 Vincovci 1 970

Palma de Majorca 1 970 Olot 1 9 7 1 Vrnjacka Banja 1 97 1 Bebrits Memorial 1 972 Yugoslavian Championship 1 972 Graz 1 972 Chess Olympiad, Skopje 1 972 USSR Championship 1 972 Novi Sad 1 972 Vrnjacka Banja 1 973 Leningrad Interzonal 1 973

1 42

OUTLINE OF VARIATIONS

A: The Gambit Deferred

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-Q3 4 N-QB3 P-KN3 5 P-K4 P-QN4 6 P X P . . . . . . . . . . . . . . . . . . . . . . 23 B: White Gives Up Castling

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q3 P-QN4 4 P X P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 P-K4 B X B 8 K X B P-N 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . ' . . . . . . . . . . . . . . . . 29 C: White Retains Castling

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 N-B3 P-N3 8 N-Q2 B-KN2 9 P-K4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 D: Quiet Variation

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P X P P-QR3 5 P-K3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 E: Fianchetto Variation

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 4 P x P P-QR3 5 P X P B X P 6 N-QB3 P-Q3 7 N-B3 P-N 3 8 P-KN3 B-KN2 9 B-N2 0-0 10 0-0 QN-Q2 . . . . . . . . . . . 59 F: Transpositions to the Gambit

See text

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

74

G: The Gambit Declined

1 P-Q4 N-KB3 2 P-QB4 P-B4 3 P-Q5 P-QN4 PXP G 1 : 4 . . . B-N2 5 P-QR4 . . . . . . . . . . . . . . . . 4 . . . B-N2 5 P-KN3 . . . . . . . . . . . . . . . 4 . . . B-N2 5 N-B3 . . . . . . . . . . . . . . . . . 4 . . . B-N2 5 P X P . . . . . . . . . . . . . . . . . G2: 4 N-Q2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 N-QR3 . . . . . . . . . . . . . . . . . . . . . . . . . 4 P-QR4 . . . . . . . . . . . . . . . . . . . . . . . . . . 4 P-QN3 . . . . . . . . . . . . . . . . . . . . . . . . . . 4 B-N5 . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Q-B2 . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 P-B3 . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 P X P P-QR3 5 P-B3 . . . . . . . . . . . . . . . G 3 : 4 P X P P-QR3 5 P-N6 . . . . . . . . . . . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

4 N-KB3 .

.

.

.

.

.

.

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

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

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

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

.

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

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

82 83 84 84 85 86 87 87 87 87 88 88 88 89

H : The Gambit Reversed

1 N-KB3 P-Q4 2 P-B4 P-Q5 3 P-KN 3 P-QB4 4 B-N2 N-QB3 5 0-0 P-K4 6 P-Q3 B-K2 7 P-QN4 . . . . . . . . . . . 92 6 . . . N-B3 7 P-QN4 . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 1 N-KB3 P-Q4 2 P-B4 P-Q5 3 P-KN 3 P-QB4 4 P-QN4 . . 93 143

AN IMPORTANT NEW BOOK It is an obvious fact that whether you are serious about chess or merely an "armchair grandmaster," the com­ plete record of an important inter­ national tournament is indispensable. Magazines and newspapers must nec­ essarily restrict their coverage of such events, due to the limitations of space. But the whole story, in permanent book form, should be a part of every chess library. This is why R.H.M. Chess Publish­ ing has just issued "San Antonio '72," the complete record of one of the strongest tournaments ever held in the United States. With 31 games anno­ tated by the inimitable Grandmaster Bent

Larsen,

and

many

others

by

International Master David Levy, the well-known chess writer and authority on the openings, "San Antonio '72," under the editorship of Burt Hoch­ berg, makes its bid for a place among the classics. Considering also the fact that each player has annotated his own most in­ teresting games, it is hard to imagine any chess lover's library without it, particularly at the low price of $3.95. Including Petrosian, Karpov, P®rtisch, Hort,

Additional Books Scheduled for Early Publication by R·H·M Chess Publishing

Gligoric,

Mecking,

Evans,

Browne,

Lopez,

Saidy,

Keres,

Kaplan,

and

Ken

Suttles,

D.

Larsen,

Byrne,

Campos­

Smith,

• How to Open a Chess Game, by Petrosian, Keres, larsen, Portisch, Hort, Gligoric, Evans

• The Byrne System in the King's Indian Defense, by Robert Byrne

• The Gruenfeld Defense, by Botvinnik and Estrin

• The French Defense, by Uhlmann and Gligoric

• Strategic and Tactical Significance of the Rook Pawns, by larsen

the

tournament was a vital test for both

R·H·M CHESS PUBLISHING

the experienced grandmasters and the

Sidney Fried, Publisher Burt Hochberg, Editor-in-Chief Svetozar Gligoric, Chairman, Editorial Board lubomir Kavalek, Consulting Editor

ambitious young stars of the future. You will certainly agree with us that San

Antonio

tournament

'72

was

and that

a

significant

"San

Antonio

'72" is a significant tournament book. Paperback,

271

pages,

including

photos, biographies, cross-tables, in­ dexes, and special articles, "San An­ tonio '72" is a great value at $3.95.

send

To order San Antonio '72 with your name and address (be sure to include zip) to:

$3.95

R·H·M CHESS PUBLISHING 840 WilliS AVENUE ALBERTSON, NEW YORK 11507