Miscellaneous Opening Analysis

In this chapter, we present analysis and example games on various opening variations. There are sections on:

a) The Semi-Tarrasch Defence;

b) The Tarrasch Defence;

c) The Queen's Gambit Accepted;

d) The Modern Benoni;

e) The Benko Gambit.

The Semi-Tarrasch Defence

The Case Against the Semi-Tarrasch

In Chapter 4 we demonstrated strong advantages against all defences to the Exchange Variation of the QGD. That is, after 1 d4 d5 2 c4 e6 3 £>c3 £>f6 4 cxd5 exd5 5 ±g5, Black seems to have interminable difficulties that we have found no way of overcoming. However, it is possible to play 4...^xd5, and go into the Semi-Tarrasch Defence. This defence leaves White with a lead in development and complete control of the centre. It has always seemed to me to be rather lacking in concept for Black. However, Black does exchange two minor pieces, which reduces White's attacking chances, and Black will get a queenside pawn majority for the ending, if he survives to get there.

An Unusual Departure

After Black's 4...£>xd5, White's best play is 5 e4 £>xc3 6 bxc3 c5 7 £rf3 (D).

Clearly White has been playing all System moves. On his last move, he is allowed to block his f2-pawn since White already controls the whole centre, and the pawn is not needed to support it. The only deficit of 7 £tf3 is that it allows Black to exchange another pair of pieces, thus reducing the advantage of controlling the centre. In the above position, Black usually continues 7...cxd4 8 cxd4 i.b4+ 9 i.d2 i.xd2+ 10 «xd2 0-0 (D).

Then White (to move) has a development count of -4, and Black -5. So, White has a 1.5 tempi lead in development, and controls the whole centre. However, he must show what he can do with this. We consider this position below.

First, we dispose of an interesting sub-line played by Korchnoi, as it again shows how one should reason about applying The System. In the game Kupka-Korchnoi, Luhacovice 1969, Korchnoi played the unusual l..Mcl (D), and it is this move that gives us an interesting chance to study the application of The System.

Black's last move can hardly be described as a developing move. Its sole purpose is to prevent the natural development of White's fl-bishop, since 8 J.c4 cxd4 now forces 9 «xd4. Also, 8 i.d3 is not playable since 8...cxd4 9 cxd4 «c3+ wins a piece. Now, clearly the most logical moves are 8 J.d3 and 8 J.c4, but after rechecking the above variations, we come to the conclusion that neither is playable.

What about 8 i.e2? That is what was played in the above-quoted game, but it is a poor move since the bishop will have to move later to get a better post. This is like giving Black a free move. Nor are purely preventive moves such as 8 i.d2 or 8 flfb3, that are intended to keep the black queen out of c3, justified. Also 8 i.b5+ i.d7 9 i.xd7+ £>xd7 10 0-0 only aids Black's development. Remember the System principle that one does not make defensive moves during the opening, unless material has already been won, or they are part of normal development.

So what is the answer? Can the System survive such artificial attacks as l..Moll Most certainly so; what about 8 Sbl? This is, in a very real sense, a move that improves the position of one of White's pieces, and very importantly it side-steps the threat of ...lfc3+ with an attack on the rook. Now, if Black continues 8...J.e7, White can calmly play 9 i.d3 as 9...cxd4 10 cxd4 ®c3+ 11 &e2 will turn out very well for White. His king is in no real danger, while Black must lose time retreating his queen to safety.

A very instructive example, the moral of which is: "When you know you are in a System position, don't let your opponent bluff you into making an inferior move such as 8 i.e2".

The Main Lines

However, the main line, to which we now return, is 7 cxd4 8 cxd4 i.b4+ 9 i.d2 i.xd2+ 10 ®xd2 0-0 (diagram repeated here).

I have always wondered at the audacity of Black to play this position. His development count is -5, while White's (to move) is -4. Besides being 1.5 moves ahead in development, White has complete control of the centre. All Black has to show for this is the queenside pawn majority, which at this stage of things does not amount to much. However, we are


m a*®j


s ü a


¡¡¡Alt Ü



1 ffl^ü

A v

m b A

a ;

- i

now faced with one of our regular System decisions which has a strong bearing on the value of this position: where is the fl-bishop to go? One of the popular moves is 11 J.c4, which exerts strong influence on the centre. However, one should ponder the future of this bishop at c4. It can support an eventual d5, but that would lead after ...exd5 to either:

a) J.xd5 with a strong bishop at d5 in a position of severely diminished material; or b) exd5 with a passed pawn of doubtful value with so much material off the board.

The other alternative, 11 J.d3, seems rather passive until one considers the pawn structure. With the indicated future e4-e5 advance, White will increase the scope of the bishop significantly and make it harmonious with the pawn structure, while signalling a possible attack on the black king. These facts make it clear that White should play 11 J.d3

as the bishop is needed on the bl-h7 diagonal, where it cooperates beautifully with the white pawns on dark squares to control the centre.

Now Black has two usual lines of play: 1 l...£>c6 or 1 l...b6. In the first, Black plans to swap queens and go into an ending with a spatial inferiority, but hopes to capitalize on the backward d-pawn. In the second he plays the middlegame and hopes to keep White from working up a significant attack against the black king. I originally believed that White could keep a meaningful advantage against either line. Here was my analysis:

The ll...£ic6 Variation

1 l...^c6 brings us to the following position (D):


i i

it Mimi


^St i 11 m fSt IP'




m aaa


HI iii ¿I fi

which White has a minimal space advantage, which hardly seems enough to have real hopes of winning. I have won some correspondence games from this position, but in each case the defence was somewhat suspect. However, the real issue has always been the 11 ...b6 line.

After 11 jtd3 b6 (D) we have a critical position.

White replies 12 e5, the move that he planned when playing 11 Ad3. Now play continues 12...®a5 and the queens will be exchanged, after

12 e5! ±bl 13 £ig5! (attack!) 13...h6 14 h4 and now:

a) 14...hxg5 15 hxg5 g6 (there is no other hope; 15...f5 16 gxf6 is hopeless, and 15...Se8 16 ®f4 £ic6

17 i.h7+ <&f8 18 ±g6 wins immediately) 16 ®f4 &g7 17 ®f6+ ®xf6

18 gxf6+ <£g8 19 Sh2 ±xg2 (on 19...£>c6, 20 <&d2 threatens the unstoppable mate starting with Sahl) 20 Sxg2 ¿hc6 21 Sh2 ^xd4 22 Sh3 and the mate is stoppable only by sacrificing the knight with 22...£rf3+.

Black can, of course, delay capturing the g5-knight, and we now look at this.

b) On 14...i.xg2! 15 Sgl Ab7 (Black must not block the d-file with 15...i.d5, as after 16 ^xe6! fxe6 17 ®xh6 Sf7 18 ®h7+ *f8 19 Ag6 Wc7 20 Sg3 £>c6 21 Scl Black is essentially helpless) 16 £tf3!? &h8 (not, of course, 16...i.xf3 17 ®xh6 with mate to come) 17 $Lc2 and now: bl) 17...i.xf3 18®d3g6 19®xf3 with a very strong attack for the pawn, for example 19...£M7 20lfe3! <&h7 (20...*g7 21 h5 g5 22 f4; or 20...®xh4 21 Ae4!) 21 h5 Sg8 22 Shi Iff 8 23 hxg6+ fxg6 24 Ab3 Se8 25 Scl with an overwhelming attack.

b2) However, 17...f5!! 18 exf6 ®xf6! 19 ^e5 £ic6! holds the defence and establishes a win for Black.

The above lines are very tactical, and I have spent hundreds of hours with my computer analysing them. At move 16 in line 'b\ there are other possibilities than 16 &f3!?, such as 16 £ixe6 fxe6 17 ®xh6 but 17...Sf7 holds. And I have also examined 15 Sh2, which will keep the g5-knight at its post for a long time.

To the best of my understanding, the attack beginning with 13 £>g5 just barely fails; however, a more powerful computer could very well justify the move, and there are many traps as well. In view of the above, and the fact that the ll...^c6 line yields White only a tiny advantage, it may very well be that the correct System move is not 7 £\f3, but instead 7 Sbl, which prevents the exchange of bishops. This line has recently found favour in GM play.

The Tarrasch Defence

We now take up another example of White's opening play in following the principles of the System. I have always regarded the Tarrasch Defence as being a particularly weak opening for Black, but after the success that Spassky had with it (one win, four draws) in his 1966 world title match with Petrosian, maybe this view should be re-examined1. The opening starts with the moves 1 d4 d5 2 c4 e6 3 £ic3 c5 (D).

1 This section was written in 1967.

The reader will easily recognize White's first three moves as being System moves.

With his last move Black counter-attacks in the centre. White moves such as 4 GM3 are not adequate because of 4...cxd4, and White's centre is destroyed while Black's is still intact. Moves such as 4 e3 violate the Option principle by blocking in a bishop unnecessarily. So 4 cxd5 must be correct.

After Black replies with 4...exd5, his d5-pawn is forever in danger of becoming isolated and weak. Therefore, White adjusts his strategy; he can no longer dominate the centre, but instead prepares to take advantage of the weak black set-up there. Therefore, White plays 5 supporting the d-pawn so that ...cxd4 would not have to be met with ®xd4, exposing the queen. Here, blocking the f-pawn is no longer important; the position is open and piece-play is required for centre control.

After Black replies 5...^c6, White must decide how he is going to mobilize the remainder of his pieces. It seems premature to decide whether the cl-bishop is better placed at f4 or g5, and e3 is unthinkable. The only alternative plaA is to fianchetto the fl-bishop along the long diagonal by first playing g3. Up to now in our discussions, we have never mentioned this idea, which stems from hypermodern practice. Indeed, since

System moves are dedicated to relentless pursuit of set objectives, this noncommittal form of development seldom fits in. Here, however the move helps to intensify the attack against the already weak d-pawn, and is therefore the correct plan. It is interesting to note that of the thousands of lines of System play that I have developed over the years, this is the only one in which this manoeuvre is used (it was first employed by the great Akiba Rubinstein in the early 1900s).

Play now continues 6 g3 £}f6 7 Ag2 Ae7 8 0-0 0-0 9 Ag5 (D).

All these moves are natural and easy to understand. Now, we have arrived at a well-known 'book' position; one of the few times that 'book' and the System agree! This position has been generally regarded as favourable to White and so it is. But in the 1966 World Championship match, four games continued 9...cxd4 10

¿hxd4 h6, and now each time Petro-sian played 11 Ae3. This move seems quite illogical, and is certainly not in accord with the principles of the System, as the earlier moves appear to be. It is clear that the bishop must move, and 11 Axf6 ¿xf6 leads to nothing for White as Black now has the initiative in the centre. But what about 11 ±f4 (D)l

It is true that this is a retreating move; however, the retreat only comes after the bishop went to g5 to resolve the tension in the centre, and it succeeded at that. Also, the retreat is no time-loss, as Black's ...h6 is hardly a time-gain. So now, retreating is clearly better than exchanging. At f4, the bishop is very well placed, even helping to control the centre. Any attempt to dislodge it by ...4àh5 or ...g5 only weakens the black position. Now, in fact, White can continue with ®a4 followed by Sfdl obtaining an overwhelming position.

I can find no real defence against this procedure. It would have been interesting to see what Spassky would have done against it.

The Queen's Gambit Accepted

I began playing 1 d4 d5 2 c4 dxc4 3 e4! (D) in the 1960s.

I had known for a long time that this was obviously the System move, but was intimidated by a genera) lack of confidence in the move among top players, and a lack of any analysis of my own to justify the move. However, when it became clear that in many variations White could allow Black to capture the d4-pawn with a pawn and not recapture, the basis for solid System play for White became apparent.

After 3 e4, Black can proceed in many ways. Among the standard moves are 3...e5 and 3...^c6. Later, we take up an unlikely alternative: 3...b5?!. After 30+ years of advocating 3 e4, the world has now caught up with how to play the white side correctly, and we leave it to excellent opening books to document this.

Here, we want to show a few things that may not be generally known. The game below was played between David Bronstein and my computer Hitech. I had, of course, spent a great deal of time evolving private opening analysis for Hitech, and included among those lines was a lot about how to play the Queen's Gambit Accepted, its favourite line against 1 d4. Here we show one of the critical lines.

The 3...e5 line; David Bronstein Plays The System - up to a point

David Bronstein - Hitech

AEGON Man-Machine Tournament 1990

1 d4 d5

2 c4 dxc4

4 exd4

When it finally became clear (I attribute this mainly to Korchnoi) that White should not play 5 ®xd4 or 5 £\xd4 here, but should instead leave the d4-pawn alone to be a defensive burden for Black, then it was possible to find the correct way to play against the QGA.

The strategy is akin to some things that Nimzowitsch tried against the French Defence. Leave a centre pawn to be taken by an opponent's pawn, and then let him worry about how to keep it (see Nimzowitsch's My System). Here the themes are a little different from those in the French, but the main point is that one does not put a piece on d4 to be attacked by Black, but instead just develops, ignoring the pawn deficit for a while.

After 5...Ab4+ 6 ^bd2 £ic6 7 0-0 Black is not as well off as in the present line.

Play since the 5th move has been best for both sides, and is certainly in accordance with System principles. At first, when studying this line, I was reluctant to believe that 9 ®xb7 could be a System move. However, there are several reasons for that:

a) It recaptures the pawn;

b) It creates weakness for Black on the c-file and elsewhere on the queenside;

c) The queen is not going to get chased around very much and is actually well placed.

What is White's best move?

Black last move, 1 l...Ab4, is the invention of Hitech. It threatens 12...jtxd2, which leaves the white centre in bad shape. Hitech was not able to find the correct white reply, but together we found it. David Bronstein at the table was also not up to the task.

12 Wd3?

This is an error, as the sequel shows. Correct is the imaginative 12 £ic4 0-0 (on 12...£ixe4, 13 &fe5 £ixe5 14 ^xe5 Sb6 {14...®d6 15 £ic6 Sb6 16 ®c8+ *f7 17 ®xh8 ®xc6 18 ®xh7 wins} 15 ®xa7 is too strong) 13 a3 i.e7 14 flel 2b3 15 £g5! with advantage. As can be seen, the light squares on the queen-side are a headache for Black.

Now with the above error, the headaches switch to White, who has trouble guarding his e-pawn.

Now Black's position is much freer, he is ahead in development, and his passed d-pawn is assuming threatening proportions. White's next move solves none of these problems.

Apparently White expected to gain some time with this move, but he overlooked Black's reply, which is decisive.

This is the end for White. If now 16 ®xe5, 16...2b5 wins the queen. So White must retreat and the knights start to bore into the position.

On 20 Sbl, 20...^c3 wins material, but now the back rank is vulnerable.

21 Wxf2 Sxf2

24 hxg3 fixg2

In the older way of meeting the QGA, namely 3 ^f3, Black's reply 3...b5 was quickly refuted. It had to do with the fact that White would play a4, e3 and get play on the hl-a8 diagonal and on the light squares in general. It also had to do with the fact that with the e-pawn on e3 it was not a target for counter-attack as it would be after 3 e4, and did not hinder activity on the hl-a8 diagonal.

With 3 e4, it is appropriate to ask if playing 3...b5 (D) might not work now.

We would be surprised if it did, but it is worth investigating. After 3...b5, 4 a4 must be correct as it attacks Black's major weakness before he has time to consolidate, and also creates pressure to regain the sacrificed pawn. Black must reply 4...c6, whereupon it is System correct to play 5 axb5 cxb5 as this stabilizes the pawn structure prior to making any further development decisions.

Now, however, there are many possible paths for White, but again, it should be kept in mind that when there are several good developing moves, those that attack are to be preferred. Thus, it really comes down to whether 6 £\a3 or 6 £\c3 is best. One should look with a gimlet eye at 6 £\a3 as this is not a normal square for the knight, and in this case only may have some merit because it cannot be met by 6...b4. However, in this set-up, White must be very careful about the weakness of his d4-pawn if and when Black gets to play &b4+. There are many lines after 6 £\a3 where this can happen, For instance: 6 £>a3 Ad7 7 b3 e5! and White is in trouble. Eventually, it becomes clear that White's best move must be 6 £\c3 (D).





■ // /






- ' AAA


®xa7 i.xe4 14 f3 £ic6 Black stands considerably better.

I spent a long time here trying to understand what White has for his pawn. At the moment, it is only an advantage in space and development. Once Black plays ...e6, he will be fully in the battle. So maybe it would be good to increase the space advantage and prevent ...e6 by 7 d5! (Dwhich also prevents the development of the black queenside.

Now Black must find the best way to proceed. 6...^.a6 does not work because after 7 $Lf4 (threatening both ^.xb8 and £M5) there is no defence for Black. That the end should come so quickly bodes no good for Black. Also, 6...b4 7 e6 8 i.xc4 exd5 9 $Lxd5 wins the rook in the corner.

So Black must pretty well play 6...Ad7. Now, the immediate 7 b3 b4 does not appear to work as there is no good way for White to continue. After 8 i.xc4?! bxc3 9 i.xf7+ (one should eschew such moves on general principles as this must be a premature attack) 9...4>xf7 10 ®h5+ g6 11 ®d5+ <&g7 12 ®xa8 ±c6 13

Now, if 7...e6 or 7...e5, White plays 8 dxe6 and Black has no good recapture. If 8...fxe6, 9 ^xb5 i.xb5 10 Wh5+ gives White a great superiority, while 8...i.xe6 9 !fxd8+ &xd8 10 £>xb5 leaves Black sorry he ever undertook this defence. So Black must find another development plan, and this is not easy. 7...^f6 8 e5 is unappetizing for Black. If all these normal moves are bad, then Black must be in deep trouble. Indeed, the best I have been able to find is 7...a5 8 and Black really has not much hope here, with £\e5 in the wind and White having such a large space advantage.

The only other reasonable line besides 3...e5 is 3...^c6. Now, the System move is clearly 4 which develops and defends the centre, whereas the only other reasonable move, 4 d5, would prematurely commit the white centre. According to the books, Black is now best advised to play 4...Ag4, which is answered simply by 5 $Lxc4. Now, Black dare not try to capture the d-pawn as f7 is too weak, so he must play 5...e6. Now, the threat against the d-pawn is real again, and White must decide how to proceed. It is easy to find 6 d5! (D), which attacks the weak black queenside and forces some decisions on him.

He cannot play 6...^e5 because of 7 ®a4+, nor can he play 6...Axf3 7 ®xf3 ^e5 8 Ab5+ c6 9 ®c3 cxb5 10 ®xe5 when White is, in effect, a pawn ahead.

So, Black must meet 6 d5 with 6...exd5 7 Jixd5 and now White has many threats. He is much better developed, and is ready to attack while Black still has his kingside pieces at home. Best appears to be 7...£>f6 8 .&xc6+ bxc6 9 ®a4 with a strong game for White.

This is only a very sketchy treatment of this opening. The main lines follow 3...e5 and there are many defences. However, it is clear that White has a space advantage and easy development of his pieces, and this is definitely the right way to play against the QGA.

The Modern Benoni

The Benoni Defence was looked upon as something close to crazy in the 1920s and 1930s when it was first employed. It was felt that leaving White so much of the centre could not possibly be good. Black mostly was only interested in obtaining a draw, and would usually assume passive positions and eventually lose. The idea of Black playing ...e6 and ...exd5 was tried, but the mechanics of how to make it work were beyond the then practitioners of this defence.

Then in the late 1950s, Mikhail Tal showed what one could do with the black pieces. He would open the e-file for Black and aim his pieces at the white e4-pawn, and begin his counterattacks with any of ...c4, ...b5, or even ...^g4. Black's play hinged mainly on the weakness of the e4-pawn, and the weaknesses that White made in defending it. Anyone who has followed The System up to here, knows that the premature &f3 leaves the e4-square in need of support, and the way the Modern Benoni was introduced was usually after White had already played this move, or White (not knowing or caring) would play £}f3 at some time anyway. This, of course, meant that the e4-pawn would have to be defended by pieces that could be overloaded with other functions they should also fulfil. Below is an example of how White should play against the Modern Benoni.

Eastern Open 1969

1 d4

2 c4 c5

3 d5 e6

4 Qsc3 exd5

5 cxd5 d6

The System approach to a defence such as the Modern Benoni is just to develop according to System

principles. So the above white moves are easy to understand. Now comes the first important System decision: Where should the fl-bishop go? It is possible to postpone this decision by playing £\f3, and then deciding among e2, d3, and c4 for the bishop's location. However, this entails a certain neglect of the centre. In particular, the e4-pawn will be attacked by Black, and it would be easiest just to defend it and the light squares in that vicinity by f3. However, the commitment to f3 means that the gl-knight must go to e2, and this means that the fl-bishop must be developed first. The question is "To what square?".

The alert reader will notice that the above set-up resembles the recommended set-up against the King's Indian and also the Queen's Gambit Declined. This set-up is very solid since the knights mutually defend each other, which in this case is even more important as the g7-bishop will exert strong pressure on the long diagonal. So, if White wishes to play this set-up, he must continue with 7 JLd3 right now. The only alternative, 7 JLb5+ to exchange off his less good bishop, is premature in so far as the situation in the centre is still too fluid to make an exact determination of the worth of the minor pieces.

Black has the choice of this deployment of the knight or to play ...£>bd7 with an eye toward an eventual ...£>e5 and the exchange of the d3-bishop. However, 9...£>bd7 10 £>g3 £ie5 11 ±e2 followed by f4, e5 and on ...dxe5 then f5 gives White a very strong attack that has basically put this line out of business. Details of this can be found in any good book on this opening.

White's 10th move was far from standard in 1969 when this game was played. The reasons have already been given: solidity from which to build. White can now play Ae3 when required and not have to worry about the reply ...£}g4. Now Black had to make another decision. 10...£ib4 does not work because of 11 ^.bl and then a3 will drive the knight away. However, Black may just envisage the general advance of his queenside pawns. This could be better enforced with 10...Ad7 or 10...Hb8, supporting an eventual ...b5, and leaving open for the moment the question of where the a6-knight is bound.

12 Sbl!

This is the star move of this game. During the 1960s I had been analysing many of the System formations arising in various defences to 1 d4.

This frequently came up in connection with the World Correspondence Championship in which I was participating in at the time. One of the things that my understanding of The System did not include at the time was exactly what to do with rooks. The rules for the minor pieces were securely in place, but how and when does one utilize the rooks?

The conclusion that I came to is that in many of these openings when Black has created no direct object for attack, White should make a space-grab on the queenside with the move b4. This frequently has to be prepared, almost always by 2b 1, and never by a3 unless Black prevents b4 with ...a5. So this is a rook move with this purpose. Openings such as the Dutch Defence, Old Indian and King's Indian fall into this category.

In this case, there is actually much more to it. When one looks at the current position, one can see that White's game is beautiful, except he has been denied the use of the d4-square. If this could be occupied by a knight, then White's position would be overwhelming. So if White can play b4, a challenge to c5 would ensue. This could be reinforced by Ae3 putting more pressure on the pawn. Black could attempt to meet this with an eventual ...b6 defending the c5-pawn sufficiently. However, in doing that, he would create a static formation on the queenside which White could attack in a number of ways. He could play for bxc5 and then depending upon how Black captures, play on the b-file or advance d6. Or he could play b5! and place a blockading piece on c4 from where it will stymie Black's position while White quietly prepares the advance e5. Contrary to these dynamic possibilities, a move such as 12 a4, which is frequently advocated by chess writers, represents a passive prevention of ...b5, and this leads to a passive position for White on the queenside.

Black could cautiously await events on the queenside, in which case White has clearly won the opening, or Black could just attempt to continue his queenside pawn advance. This is the way the game now continues.

Now Black is faced with a threat to his b5-pawn after White plays bxc5. He has several choices:

a) 13...c4, which leaves his queen-side pawns blocked and also gives up control over d4;

b) Defending passively by 13...a6 or 13 &d7, when White will continue the attack on the queenside pawns and the squares they guard with a4; and c) The game continuation. Black opts for freedom, but White's space advantage is too great.

14 Sxb4 a5

15 Sbl Ml

The opening can now be appraised: White's pieces enjoy great freedom and are harmoniously placed, while Black is struggling to find meaningful scope for his.

Black makes a bid for a place in the sun; however, it is ill-advised as he is behind in development, and the exchange of his good bishop on d7 will further weaken his position.

Now White is assured of control of the c-file and play against the weak d6-pawn and queenside pawns. Black is now positionally lost because of his many weaknesses.


1 SÜ






¿h A

Ü i

/ JL

A ^

A -

m m ;|p

22 Hb7

23 M4 Hd7?

This loses a pawn. However, even after the better 23...Sa8, 24 Sfcl leaves White very much in command, as Black can undertake nothing.

Black cannot now play 25...Hxa5 since 26 Sxb4 ®xa2 27 Sb8+ Af8 28 ®h6 leads to mate. So, he tries to save the situation by tactical trickery, which unfortunately for him does not work.

Threatening Jkc3 with mates upcoming, and so winning more material.


















The idea of Sbl followed by b4 is present in all alternative defences. White will stabilize the centre when needed with f3, but basically continue his attack on c5 with the above plan. I can find no effective way to counter this except the idea already mentioned above, namely ...£\bd7. However, this line is never played any more because of the Penrose attack which yields a very strong white attack against the black kingside.

A typical line of play is 9...a6 10 a4 He8 11 &g3 &bd7 12 f4 Hb8 and now the break with e5, meeting ...dxe5 with f5, will leave Black helpless. The d5-pawn is very strong and there is the ever-present threat of fxg6 while White develops his pieces in the direction of Black's king. A possible continuation would be 13 e5 dxe5 14 f5 c4 15 ±c2 b5 (15...£ic5 16 ±g5) 16 axb5 axb5 17 £\ce4 with a very strong attack well worth the pawn.

The Rather game, to me, epitomizes the way White should play. Black really never had a chance. I also wish to call the reader's attention to the Response Pairs present in this opening:

a) Black's ...^bd7 should be met by f4 to keep the knight out of e5.

b) If the b8-knight decides to move to a6, then play will be against the queenside after f3, since f4 is no longer needed to keep the knight out of e5.

The Benko Gambit

Some Philosophy

As has been mentioned many times previously in this book, the management of pawns is of the utmost importance. No other piece contributes as much to the success of a position, because of the tyranny of the weak. The pawn is worth little and can therefore intimidate all pieces of greater value.

One issue that we have said little about is the capture of pawns by pawns. We have indicated that central pawns have a greater value than less central ones, and this leads to the age-old maxim of Capture toward the Centre. However, there are situations where this maxim does not apply and pawn-capture decisions must be made. It is those situations that are investigated in this section, and we aim to give firm rules for all situations.

Let us begin by looking at a very simple opening: 1 d4 £}f6 2 c4 c5. Black offers a pawn sacrifice to get a pawn away from the centre. It is well known that after 3 dxc5, Black will regain his pawn and thus have achieved his goal of reducing White's majority in the centre. Thus, this offers White no hope of advantage. So, if White hopes to gain an advantage, he must play 3 d5. Now, it is possible to continue in many ways, most of which are taken up in this book. However, here we wish to see what we can learn from how to proceed against 3...b5?! (D).

There are a number of things that should be immediately clear. Black again wishes to lure a more central pawn away by a capture of a less central one. How should White proceed? If he plays 4 b3 and stabilizes the pawn situation, he makes a totally uncalled-for defensive move and compromises his position at this early stage. If he lets Black capture on c4, White will be able to recapture the pawn with a piece, but he will have allowed Black to achieve the goal of swapping the b-pawn for the c-pawn. There may be some merit to this for White, as he will gain somewhat in time.

One possibility is 4 f3 bxc4 5 e4 e6 6 ±xc4 exd5 and Black has no real difficulties ahead. Also possible is 4 e3, which is clearly not a System move as it gives up on an early e4. Many other moves, such as 4 ®c2,4 &f3, 4 &d2, and 4 a4, have been played here, but the essence of all these attempts is that Black will capture on c4 and thereafter be able to operate effectively against the white centre and on the b-file. So how is White to play?

One thing that affects all these decisions is the discovery in the 1960s that if White plays 4 cxb5, Black can satisfactorily offer the permanent pawn sacrifice 4...a6!.

If this is met by 5 bxa6, Black gets a fine game on the queenside with play against the white pawns there, and on the half-open a- & b-files. Is there anything else that should be considered here?

The System Approach to the Benko Gambit

It took me a long time to understand that 4 cxb5 could be a System move. It captures away from the centre, but there is a critical difference: after 2...c5 White can avoid the capture by advancing; here he cannot. No matter how White plays, his c-pawn will be exchanged for the black b-pawn unless he plays the horrible 4 b3?. So, if this exchange is inevitable, why not make Black pay in some way? The most obvious and correct way is to capture: 4 cxb5!. Now, however, after 4...a6!, we should not be anxious to play 5 bxa6, but instead consider whether there is some positional advantage that can be gained while Black is recovering his pawn.

The principal advantage that is available is control of the centre. White cannot do this by 5 £>c3 as after 5...axb5 he still cannot play 6 e4 b4, and the alternative of 6 £>xb5 is not appealing at all. However, there is a way of taking the centre that we have seen again and again in System play. It is 5 f3! (D).

Black's decision as to whether he wants to recover the pawn now, or challenge the centre by ...e6. 5...e6 6 e4 exd5 7 exd5 We7+?! 8 £f2! c4 9 £>c3 axb5 (9...*fb4 10 We2+ &d8 11 Ifxc4 J.C5+ 12 <&g3! i.d6+ 13 &h3 leaves White two pawns ahead; his king is not really in much danger) is a line that has been played in top-level competition during 1997. Now White can play the excellent 10 d6! (D).

The 5 f3 line

Now it is Black who has a problem. 6 e4 cannot be prevented and it is

Then 10...®xd6 11 ®xd6 ±xd6 12 £ixb5 ±c5+ 13 ±e3 ±xe3+ 14 is winning for White. A much better line for Black is 10...*fe5! 11 ®e2! ®xe2+ (1 l..Jbcd6 12 £xb5) 12 £igxe2 ±xd6 13 £ixb5 ±c5+ 14 ±e3 Axe3+ 15 <&xe3 <&e7! 16 £>ed4! d5 17 ±e2 ±dl 18 b3 ±xb5 (18...cxb3 19 axb3 £ic6 20 £ixc6+ ±xc6 21 «¿>d4 with a strong endgame advantage because of the outside passed pawn and the better bishop) 19 £>xb5 cxb3 20 axb3 £>c6

21 £M4! ^xd4 22 <&xd4 with a clear advantage. It should be noted that White's play in this line begins by disrupting Black's attempts to attack. 10 d6 splits the black forces, and after the tactics that follow, Black has exchanged his good bishop, and is left with weak central pawns that are on the same colour as his remaining bishop. White then proceeds to exchange the c-pawn, leaving him with an outside passed pawn in a very strong position.

The 5...axb5 line

If the aggressive 5...e6 fails, as it apparently does, then Black has nothing better than 5...axb5 6 e4 ®a5+ (Black dare not play 6...b4, 6...iLa6 or 6...®b6 as the reply 7 e5 is too strong) with the following position (D).

This sets before White an interesting problem. It appears as if there must be some easy way to refute this premature check. There are many possibilities: 7 JLd2, 7 &c3, and even 7 b4?!. After having invested quite a bit of time studying these alternatives, it became clear that 7 ±d2 b4 8 £ia3 d6 9 £ic4 ®d8 (D) is very good for White.

Black appears ready to put serious pressure on the d5-pawn, and White must be able to meet this without making any important concessions, such as giving up more of the centre. From a System point of view, he could continue with either 10 a4 or 10 a3 and have the advantage on the queenside. However, his centre has become weak as his d5-pawn is not supported by the queen and cannot be easily supported in the future by pawns or pieces. So Black can soon play ...e6, and thereafter will be able to wage a reasonable fight for the centre and the future of the game.

White must strike now on the queenside, before Black completes his development. He has the choice of 10 a3 or 10 a4. 10 a4 is a good positional move, but immediate action on the queenside is required. After 10 a4 e6 11 £ie3! exd5 12 exd5 ±e7 13 ±b5+ £ibd7! 14 £ie2 0-0 15 0-0 Ab7 Black has sufficient play against the d5-pawn to assure equality.

After the correct 10 a3, it is clear that for Black to play 10...bxa3 is a mistake because of 11 Hxa3 Hxa3 12 bxa3 and now White has a passed a-pawn and a great deal of room to support its advance as both bishops can now roam freely on the queen-side. Examination of the passed-pawn situation shows that White's a-pawn is stronger than Black's protected passed pawn, since the a-pawn's advance is easily supported by pieces while the c-pawn's is not.

So, 10...e6, with the idea of challenging the centre, must be correct. Now, White must play a move that I am not at all sure is a System move. It is 11 £\e3! (D).

It moves a well-placed piece a third time; however, in the process it makes the deployment of White's kingside forces much easier. Also, the transgression is not that bad, as Black has moved his queen twice and arrived back at its original square. The development situation is actually quite well balanced: White's tempo count is -4 since his al-rook is essentially developed, and Black's

is -5 as his c8-bishop is essentially developed. Computer analysis likes 11 £\e3, and it could well be the key move to this whole variation.'

The important point is that, if White can maintain the d5-pawn, he will have split the black forces into two camps and they will not be able to coordinate on anything offensive, while White can operate on both wings. Black must now apparently acquiesce to 1 l...bxa3 12 Sxa3 Sxa3 13 bxa3 exd5 14 exd5, whereafter White can easily develop his king-side while Black struggles to find meaningful locations for his pieces. A possible continuation is 14...^.e7 15 ±b5+ £ibd7 16 £ie2, with an excellent position for White.

I do not believe there is any merit in the sacrificial line ll...JLe7?!, which is met simply by 12 axb4 Sxal 13 Wxal exd5 14 exd5, when Black has very little to show for his pawn minus. The e3-knight in this very unusual post holds everything together nicely, and White now has a wonderful position besides his extra pawn.

Clearly, this whole variation needs some testing, as I believe it is completely new, having originated with this book. The theme of 11 £ie3! is to maintain the d5-pawn until reinforcements arrive in the best style of Western movies. It turns out that Black just cannot attack the d5-pawn more times than it can be defended, and when the pawn survives, it destroys Black's ability to operate. The fact that variations in this section are very different from any encountered anywhere else in this book does not mean that they are not System Correct. 11 £>e3! is a very surprising move, but it has excellent effects, and it only gives up a unit of time that had already been gained chasing the black queen.

Once having come to the realization that White's 4 cxb5 is a correct System move, the first few moves in these lines against the Benko Gambit are not difficult to understand.

However, the play beyond that is very complicated, involving ideas that have not been invoked in any other parts of this book. However, as everywhere in this book, the centre is paramount, and the impeding of the opponent's development by controlling the centre is the main theme. Many of these lines are completely new, and have come into being as the result of examining current (1998) opening theory and dabbling with my friend Fritz 5.0 to see where inadequacies in White's play can be overcome. Clearly, in an enterprise such as this, there will be mistakes. However, I have spent a great deal of time and effort to find the right way, and the analysis above, while not fully complete, appears to be it. The nature of the play is such that, while there may still be room for improvement for Black, it is not a very inviting position to play, and there is probably at least one improvement for White for every one for Black. There will be few players who will want to test these variations as Black.

0 0

Post a comment