Caro-Kann Defence E-Book

Anatoly Karpov

• Herausgeber: Batsford
• Kategorie: Lebensstil
• Sprache: Englisch

Seitenzahl: 630

Veröffentlichungsjahr: 2014

Beschreibung

The Caro-Kann Defence has a justified reputation for reliability, making it an attractive choice for players with a preference for positional play. However in recent years many new dangerous ideas have been discovered for White which makes a study of the theoretical main lines imperative for competitive players. The Panov Attack against the Caro-Kann, 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4, covered in this second volume in the series, leads to a rapid opening of lines and very lively piece play with prospects of a quick attack on the kingside, queenside or the centre. Former world champion Anatoly Karpov and his collaborator Mikhail Podgaets select and analyse the essential main lines that provide Black not only with security, but also with active play. About the Author Anatoly Karpov is a living legend of the chess; he has dominated the world of the chess for over a decade as the world chess champion between 1975 and 1985. Garry Kasparov, the current champion, dethroned him and the two ideologically opposed Russians have had a bitter rivalry ever since. Anatoly Karpov is also the winner of the largest number of tournaments in the history of the chess, with more than 160 due tournaments. He is considered the best positional player of history of chess and is the author of several other books on chess theory. 

Leseprobe

More Great Titles from Batsford

sign up for our newsletter here tap to read more

@Batsford_Books

www.anovabooks.com

Contents

Foreword

Chapter One

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 c6 6 f3

Index to Chapter One

Chapter Two

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 c6 6 g5

Index to Chapter Two

Chapter Three

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 g6!?

Index to Chapter Three

Chapter Four

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 e6 6 f3 c6

Index to Chapter Four

Chapter Five

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 e6 6 f3 b4

Index to Chapter Five

Chapter Six

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 e6 6 f3 e7

Index to Chapter Six

Appendix: Steiner System

1 e4 c6 2 c4

Index to Steiner System

Illustrative Games

Index to Illustrative Games

Foreword

The opening system, characterised by the moves 1 e4 c6 2 d4 d5 3 ed cd 4 c4!?

is so called in honour of the Soviet master and theoretician Vasily Panov, who published his analysis in 1930. For the sake of accuracy we mention that as far back as 1925 the idea of exchanging on d5 followed by the dash of the c-pawn was tried by A.Alekhine (in a game against Tartakower) but without success. In 1931 was played the famous game Nimzowitsch - Alekhine, in which the 4th world champion was successful in his fight against the Panov Attack, but, starting the following year, he included it in his own opening repertoire with stunning successes. Eight wins in ten games – such a score is the envy of every opening variation!

In 1933 the variation was tested by M.Botvinnik in a match against S.Flohr – with variable success (one out of two), but in the following years the Panov Attack served Mikhail Moiseevich faithfully. Spectacular and convincing victories over H.Kmoch (Leningrad 1934), R.Spielmann (Moscow 1935), A.Budo (Leningrad 1938), A.Konstantinopolsky (Sverdlovsk 1943), H.Golombek (Moscow 1956) – are proof of this.

In our day the Panov Attack has rather receded into the background, but in no way has it become a second class opening system. It is enough to say that it is employed by elite grandmasters – M.Adams, J.Polgar, A.Morozevich and V.Ivanchuk.

The fortune of the Panov Attack in matches is likewise remarkable. We recall the world championship matches Chiburdanidze – Ioseliani (Telavi 1988) and Karpov – Kamsky (Elista 1996); in both contests there was a dispute over the Panov Attack, and only with great difficulty did Black contain the opponent’s attack.

The material presented in this book is laid out in the following way.

In the first half the authors deal with those defences in which Black refrains from the move e7-e6. Thus, Chapter One is devoted to the variation 1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 c6 6 f3;Chapter Two – the variation 1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 c6 6 g5. In Chapter Three is given an analysis of the continuation 1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 g6.

In the second half we deal with the main line: 1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 e6 The positions arising after the exchange on d5 (or c4), with the isolation of the d4 pawn, are analogous to several schemes in the Queen’s Gambit Accepted, Nimzo-Indian Defence, or completely transpose into them. Consequently readers who study the given formations have at their disposal a universal scheme, suitable for immediate application in a number of openings which at first sight look completely different.

The material in the second half is divided into three: after 6 f3 Black can choose between 6…c6 (Chapter Four), 6…b4 (Chapter Five) and 6…e7 (Chapter Six).

In the Appendix we analyse the individual Steiner System: 1 e4 c6 2 c4!?

In certain cases this system inevitably transposes into the Panov Attack, but in others its branches resemble Indian or Slav opening schemes. It is important to mention that by examining the Panov Attack together with the Steiner Attack, the reader obtains exhaustive information both on the attacking potential of the c2-c4 idea in the Caro Kann Defence and also on Black’s possibilities of neutralising the attack.

The Illustrative Games section includes fresh practical material to supplement the theory of the Panov Attack.

Chapter One

1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3 c6 6 f3

And so, we return to the plan where Black rejects the immediate advance e7-e6. Obviously, he intends to resolve the problem of the centre in another way.

There are two main paths: 5…c6 and 5…g6, but first we deal with 5…e6!? (after 1 e4 c6 2 d4 d5 3 ed cd 4 c4 f6 5 c3)

The idea looks artificial, but with improvements, above all by the English grandmaster Anthony Miles, the thrust of the bishop gained a reputation of being an interesting and in any case useful move for the continuation of the struggle.

6 ge2 Considered the most dangerous – the knight heads for f4, from where it will ‘exert itself’ over the d5 and e6 squares. No advantage comes out of 6 cd xd5 7 xd5 xd5 8 e2 c6 9 f3 c4 10 e2 in view of 10…e5! 11 b3 a6 12 xc6+ xc6 13 0-0 d8 (Kosten – Miles, Edinburgh 1985), while 6 c5 will be examined later under a different order of moves – 5…c6 6 f3 e6 7 c5.

6…dc 7 f4 g4 A necessary intermediate move. Weaker is 7…c8 8 xc4 e6 because of 9 d5! e5 10 0-0! He cannot accept the piece sacrifice – 10…ef? 11 e1+ e7 12 d6, while 10…d6 led to an undoubted advantage for White in the game Hebden – Martin (Edinburgh 1985): 11 h5 xh5 12 xh5 0-0 13 e4 e7 14 d2 d7 15 c3 f5 16 xd6 xd6 17 e2 etc.

In reply to 7…g4 White usually goes into the variation 8 f3 d7 9 xc4 e6 10 d5 e5 (but not 10…ed?! in view of 11 e2+ e7 12 fxd5 xd5 13 xd5 e6 14 f4! xd5 15 0-0-0 0-0 16 xd5 with an enormous advantage, Yurtaev – Fette, Lungby 1990) 11 d3, but after 11…d6 things do not turn out badly for Black, for example: 12 g5 f5 13 f2 bd7 14 d3 xd3 15 xd3 (Dzhandzhava – Komarov, Novosibirsk 1989), and here, in the opinion of L.Dzhandzhava, 15…b6! secures full equality.

A new (and very successful) fight for the advantage was undertaken by the English grandmaster A.Kosten: 8 a4+!? (instead of 8 f3) 8…d7 9 xc4 e6 10 e2 c6 11 0-0 c8 12 d1 d6 13 d5!

To exploit the opponent’s backward development, White has every right to deliver a blow in the centre. Taking the pawn is too dangerous; the knight must retreat. If 13…b4?! 14 b3 ed, then 15 a3 c6 16 fxd5 with advantage. He could win a pawn: 13…e7 14 b3 xf4 15 xf4 fxd5, but after 16 xd5 xd5 17 d6 c6 18 g3 the weakness of the dark squares is deeply felt.

In the game Kosten – Komarov (France 1994) Black preferred 13…a5 14 d3 e5, not noticing 15 e6! A beautiful tactical blow secures White slight, but stoic pressure in the endgame (he has the two bishops to his credit): 15…fe 16 de xe6 17 xd6 xd6 18 xd6 e7 19 d1 hd8 20 e3 etc.

We will return to the idea c8-e6, when going into the position after 5…c6 6 f3.

5…c6 The idea of the move in comparison with 5…e6 is clear: Black immediately attacks the d4 pawn, leaving the light-squared bishop with more room for action on the c8-h3 diagonal.

White has two possibilities of fighting for the initiative. The first is linked to Botvinnik’s idea 6 g5 (the whole of Chapter Two is devoted to this), but here we deal with 6 f3. Because of the deployment of the knights against one another this system is still called the Four Knights.

Of course, 6…g4 looks the most natural, but we will also deal with other bishop moves.

6…f5?! An almost completely forgotten continuation. According to an analysis by Nenarokov White has a slight positional advantage after 7 c5! e6 8 b5! d7 9 f4 e7 10 h3 0-0 11 0-0.

6…e6!? is already known to us, although with the inclusion of the moves 5…c6 6 f3 things are changed somewhat (White has neither a check on a4, nor the manoeuvre g1-e2-f4, but play appears on the pin of the knight c6). After 7 c5 (on 7 b3 simplest is 7…dc! 8 xc4 xc4 9 xc4 e6 – the recommendation of E.Vladimirov) Black is faced with a not very easy choice.

7…g6 is insufficient for equality in view of 8 b5! g7 9 e5! In the old game Dake – Alekhine (Pasadena 1932) Black got a bad position without any hint of counterplay: 9…c8 10 a4 d7 11 0-0 0-0 12 f4 a6 13 xc6 bc? 14 fe1 h5 15 d2 a7 16 e2 e8 17 ae1.

Of course, 13…bc? is a serious positional mistake; in general Botvinnik considered that after the correct 13…xc6 Black should not experience difficulties: 14 xc6 bc 15 fe1 h5 16 g5 g4!, and dangerous is 17 xe7 because of 17…f4! with an attack.

But a desire to test Botvinnik’s analysis in practice is something we do not see. For example, the game Anand – Miles (Wijk aan Zee 1989) developed not ‘à la Botvinnik’ but just ‘à la Alekhine’: 9…d7 10 xc6 bc?! (why not still 10…xc6!?), and White again obtains the sought for advantage: 11 0-0 0-0 12 e1 e8 13 h3 h8 14 f4 g8 15 b4 f6 16 f3 d7 17 a4 a6 18 h2 g5 19 e2 h5 20 e6!

Not leading to equality is 7…g4!? (instead of 7…g6) 8 b5 xf3 9 xf3 e6. Black has secured himself against the threat f3-e5, but at the cost of the loss of the important light-squared bishop. The further continuation of the game Brunner – Miles (Bad Worishofen 1989) was 10 0-0 e7 11 f4 0-0 12 xc6 bc 13 b4 d7 14 b5 c8 15 ab1 f6 16 bc xc6 17 b5 a5 18 fc1 a8, and White agreed to a draw. Too early! The plan to improve the position lies on the surface: 19 a3! f8 (not possible is 19…a6 20 d6 xd4? because of 21 b7! a4 22 b4) 20 b4 g6 21 g3 e7 22 cb1. The only open file is in White’s hands, his pieces are also very active. One cannot talk about equality.

Perhaps grandmaster Dreev has penetrated the position the deepest: 7…a6!?

Sympathetic prophylaxis. Now White’s play to pin the knight c6 is rendered harmless, and the bishop e6 need not be given up. In the game Brunner – Dreev (Biel 1995) followed 8 h3 f5 9 e2 g6 10 0-0 g7 11 f4 0-0 12 e5 d7! 13 xd5 dxe5 and the opponents concluded peace.

6…g4 is the main continuation. The threat to the d4 pawn assumes an all the more tangible outline.

7 cd We are convinced that other moves are noticeably weaker than the capture on d5.

There is no point in playing 7 c5? – after 7…xf3 8 gf g6 Black’s position is better.

There is less benefit in 7 e3 e6 8 e2 e7 9 0-0 0-0 10 e5 xe2 11 xe2 dc! (a well-known device: the weakness of the c6 pawn after the exchange of knights is balanced out by the play against the the isolated d4 pawn) 12 xc6 bc 13 xc4 d7 (De Firmian – Christiansen, Key West 1994).

7 e2 – A trappy move. If Black is tempted by the bait – 7…dc?!, then after 8 d5 xf3 9 xf3 e5 10 0-0 he risks, as they say, not getting out of the opening. How serious it all is was shown if only by the game Mikenas – Flohr (Folkestone 1933): 10…d7 11 e2 xf3+ 12 xf3 0-0-0 13 b3! e6 14 bc ed 15 f4 d4 16 b5 c5 17 ab1 c6 18 h3+ d7? The second, and this time decisive mistake. On 18…d7 the result of the game is quite unclear, whereas now…

19 xa7+! xa7 20 a3!, and Black had to resign.

It is best for Black not to accept the Greek Gift, but calmly play 7…e6. There are no pawn weaknesses, also no problems with development – where is White’s advantage coming from?

7…xd5 8 b3 Yet again unfashionable is 8 e2 e6 9 0-0 e7 10 h3 h5. The only problem for Black is that he lags behind his opponent in development by one or two tempi. White can possibly try to exploit this by 11 b3!, but Black has sufficient defensive resources:

11…b6!? 12 e3 0-0 13 fd1 b4!? 14 d5!? 4xd5 15 xb6 xb6 16 xd5 ed 17 xd5 xb3 18 ab xf3 19 xf3 d8! 20 d7 b6, and the extra pawn plays no role at all;

11…xf3!? 12 xf3 d7!? (the most concrete way to equality) 13 xd5 ed 14 xd5 xd5 15 xd5 d8 16 xe7 xe7 17 e3 xd4 18 fe1 f6 19 ac1 c6.

In the famous game Nimzowitsch – Alekhine (Bled 1931) was played 8 b5 a5 9 b3 xf3 10 gf xc3. In this position the great chess inventor Aron Nimzowitsch got confused:

11 xc6+?! bc 12 b7? d5+! 13 d2 b6! 14 xa8+ d7 15 0-0 c7. The queen can still be saved (16 a5), but the game – already not.

A year later the 4th world champion himself showed the right way for White: 11 bc! e6 12 d5!, and after 12…ed?! he obtained a very strong initiative: 13 0-0 0-0-0 14 xc6 bc 15 b1 (Alekhine – Winter, London 1932). However the point in this theoretical debate, like the given game Keene – Roth (Aarhus 1976), is 12…a6! (instead of 12…ed?!) 13 dc ab 14 cb b8 15 b1 xb7 16 a4 b4 17 d2 e5+ 18 f1 d6 with equality.

However, is it the point?… Many chessplayers do not trust forcing variations, where it all hangs ‘on one nail’. Then in reply to 8 b5 they should play 8…c8!? This solid continuation allows Black to achieve equality without unnecessary worry. This is how events develop: 9 h3 h5 10 0-0 e6 11 e1 e7 12 e5 xc3 13 bc g6 14 xc6+ xc6 15 d5 xc3 16 e1 c7 (worth considering is 16…d3!? 17 b2 xd5 18 xd5 ed 19 xg7 g8, and if 20 f6, then 20…f8!) 17 de 0-0 18 ef+ xf7 19 g5 xg5 20 xg5 h6 21 xf7 Draw (Hasin – Bagirov, Baku 1961).

8…xf3 9 gf (once again 9 xb7? is not possible because of 9…db4! 10 gf b8 trapping the queen). On the board is the tabiya of the Four Knights system.

Black has two acceptable continuations: 9…b6 (I), leading to immense complications, and 9…e6 (II), after which the game is simplified and transfers to an endgame. The remaining possibilities are clearly weaker.

Thus, losing is 9…xd4? 10 b5+! xb5 11 xb5+ d7 12 xd7+ xd7 13 xd5 (Rantanen – Baljon, Valetta 1980).

Also poor is 9…db4? 10 e3 – the same knight cannot untie itself and there is no one to come to its aid. In the game Geller – Orev (Kislovodsk 1968) there followed 10…a5 11 d5 a4 12 fd1 b8 13 a3 4a6 14 b5+ d7 15 0-0 c7 16 xa4, and White’s advantage grew to decisive proportions.

Dubious is 9…xc3?! True, White (though it is now time!) had better forget about the b7 square – in the variation 10 xb7?! xd4 11 bc c2+ 12 e2 b8 13 c6+ d7 14 xd7+ xd7 15 h3+ c6 he does not win the knight, for example: 16 f4 e5! 17 ac1 ef 18 xc2 c5 with equality (analysis by Moiseev and Ravinsky). However simpler is 10 bc b6 11 d5 underlining White’s positional advantage.

I

9…b6

White is at a crossroads: 10 e3 (A) or 10 d5 (B).

A

10 e3 Let us say at once: not the strongest move. Although even in this case, as shown by practice, Black finds it quite difficult to refute it upon accurate play.

10…e6 As we see, Black has everything in order with his pawn structure. But here the lag in development can assume threatening proportions. Very much depends on White’s following move. Thus after 11 d1?! b4! 12 a3 a5 Black easily shakes off any fear – White will not succeed in carrying out d4-d5 in the near future. In the game Marin – Magem (Berga 1995) play continued 13 d3 c8 14 g1 0-0!? 15 f1 (the idea of castling is shown in the variation 15 h6 g6 16 xf8 xd4!) 15…xc3 16 bc g6 17 h6 e8 18 g5 d5 19 c4 a5!, and White’s attack is finally extinguished.

The break in the centre leads to unclear consequences: 11 d5 ed 12 g1 g6. For example, the game Plaskett – Wells (London 1991) led immediately to such complications that it is practically impossible to commentate on them: 13 0-0-0 d6!? 14 g5 d4! 15 e4 0-0 16 b1 e7 17 b5 d7 18 h6 fd8 19 a4 d3!? 20 xd3 d4 21 d1 xb5 22 xd7 xd7 23 e1 d4 etc.

Most often White chooses between 11 g1 and 11 0-0-0.

1) 11 g1 At first glance, the move does not require any particular explanation. With the attack on the g7 pawn White slows down the development of the bishop f8 and thereby gains some time to organise the break d4-d5.

None the less, as we see later, this is not all so simple…

a) First of all we mention that it is dubious to win a pawn – 11…xd4?! 12 xd4 xd4 in view of 13 b5+ e7 14 d1 e5+ 15 f1 g6 16 e4 f6 17 g2 f7 18 ge1 with a very strong attack.

For that reason we deal with the most logical and possibly the strongest continuation, but…not the most interesting!

b) 11…g6

12 0-0-0 In the game Zaichik – Dolmatov (Kutaisi 1978) White decided in general to do without the break d4-d5. The experiment ended unsuccessfully: 12 d1?! d6 13 h4 0-0 14 h5 b4! 15 a3 4d5 16 e4 f4 17 hg hg 18 h3 g7, and Black stands to win.

12…e7 Rejecting 12…g7?, and not without reason: after 13 d5! xd5 14 xd5 ed 15 c5! it becomes clear that the dark-squared bishop should guard the a3-f8 diagonal. In the game Sveshnikov – A.Ivanov (Leningrad 1976) this happened and Black did not succeed in rectifying the situation: 15…c7 16 e1+ e5 17 f4 e7 18 b5+ f8 19 xe5 c8 20 b4 b6 21 h3!

13 d5! The attempt to put off the pawn break ‘till later’ and play in a more refined way does not work, since after 13 h6 c7 14 h3 Black succeeds in hiding away his king: 14…0-0-0! There is a draw (and a very beautiful one) for White, but no more than that: 15 d5 (15 b5? xd4+) 15…xd5 16 xd5! (not possible is 16 xd5? xd5, and on 17 xd5 or 17 xd5 follows 17…d4+!) 16…xh2 17 g3 h4 18 f4! xg3 19 xd8+ xd8 20 xe6+! fe 21 xe6+ d7 22 g8+! with perpetual check.

After 13 d5 you get the feeling that White is about to embark on a ‘squeeze’ to hold the opponent’s king in the centre, but in actual fact everything turns to quite everyday equality. 13…ed 14 xd5 xd5 15 xd5 c7, and then:

16 c3 f6! 17 c5 (alas, the bishop is untouchable: 17 xf6? d4+ 18 d2 c2+ etc.) 17…e7 18 c3 f6 with repetition of position (Sveshnikov – Hodgson, Sochi 1986) or

16 b1 0-0 17 f4 ad8 18 g2 f6 19 c1 xd5 20 xd5 d8 21 b5 a6 22 a4 d6. Two bishops – this is a plus, but how will it be with a defective pawn formation on the king’s flank? Most likely, White will not manage to win (Lautier – Illescas, Ubeda 1997).

c) 11…b4!? Here it is! It seems that the g7 pawn can also be disregarded. This sacrifice was first made by Kasparov in a game against Ehlvest (Moscow 1977). We mention that playing 11…e7?!, with the same aim, is noticeably weaker in view of the forcing variation 12 xg7 xd4?! 13 b5+ f8 14 h6 f5 15 xh7+ g8 16 xh8+ xh8 17 f4 d4 18 d1 xb5 19 xb5 etc. (Illescas – van der Doel, Escacdes 1998).

After 11…b4 arises the most interesting moment in the whole variation with 11 g1.

If 12 xg7, then of course Black does not go for the win of the d4 pawn but concentrates his forces on the c-file: 12…d5 13 0-0-0 c8 14 b1 xc3 15 bc a6 16 c1 a5 17 a3 b6+ 18 a1 c4 with a menacing initiative.

Possible is 12 0-0-0, but then, by exchanging the knight c3, Black renders harmless the break d4-d5: 12…xc3! 13 bc f6!? This position was twice defended by Ukrainian master Peter Marusenko, and both times successfully:

14 g3 h6 15 d3 0-0-0 16 e4 d7 17 g4 hd8 followed by c6-a5-c4 (Nieminen – Marusenko, Port Erin 1999);

14 f4 0-0 15 f5!? xf5 16 h6 g6 17 xf8 xf2 18 d3 xf8 with obvious compensation for the exchange (Spanton – Marusenko, Port Erin 2000).

In the above mentioned game, Ehlvest – Kasparov, White preferred to wait a while with castling and play 12 b5!? d5 13 xg7 But even here after 13…b6!? Black found counterplay: 14 f1 xc3 15 xc6+ xc6 16 bc f8 17 g5 e7 18 b5 xf3 19 xb7 0-0 20 b1 f6 21 d1 h3+ 22 e2 h8

As before, he is a pawn down, but there is no coordination in White’s ranks. Perhaps in practical play his game is even more difficult than Black’s. Incidentally, in the further struggle Kasparov succeeded in gaining the upper hand.

There is nothing surprising in the fact that the idea 11 g1 has left the scene. If the threat to the g7 pawn does not trouble Black in the least, is it worth spending time on the rook move?

2) 11 0-0-0 White leaves the rook h1 alone and on the whole concentrates on preparation for the break d4-d5.

11…e7 It is difficult to say definitely if there is any benefit in the inclusion of the moves 11…c7 12 b1. One thing is clear: Black cannot now play 12…0-0-0 in view of 13 b5 b8 14 d5! xd5 15 h3! a6 16 xd5 xd5 17 xd5 ab 18 xe6+! winning.

12 d5 After the development of the bishop f8, the move 12 g1 looks particularly insignificant. But White played exactly this in the grandmaster games Nunn – Chandler (Bristol 1981) and Ehlvest – Oll (Riga 1995). This is how things continued: 13…0-0 13 d5 xd5 14 xd5 ed 15 xd5 c7, and now after either 16 b1 xh2 (Chandler’s move), or 16 c3 f6! (as Oll played) White, it goes without saying, cannot get organised.

After 12…ed arises a critical position for the 11 0-0-0 variation.

a) 13 xb6 brings no advantage but only if Black takes on b6 with the queen. However after 13…ab 14 xd5 0-0 15 g1 f6 16 g4! White remains with some chances, for example:

16…a5 17 b1 c5 18 xb6 d4 19 b4 c6 20 c4 Short – Miles, Brighton 1984);

16…d6 17 xb6 (it is worth waiting a while with this capture; 17 b1!?) 17…d4 18 gxd4 xd4 19 xa8 xa8 20 c4 f8 21 xb7 xb2+! (Hebden – Nunn, Marbella 1982).

Meanwhile here the endgame after 13…xb6 14 xb6 ab 15 xd5 is completely harmless for Black, which has been repeatedly confirmed in practice:

15…xa2 (possibly even stronger is 15…d8!? 16 b1 0-0 17 f4 e8 18 c4 f8 19 he1 xe1 20 xe1 c8 21 b5 e7 22 e3 c7, and Black’s position is even somewhat preferable, Zahariev – Kir.Georgiev, Corfu 1991) 16 b1 a5 17 b5 f8! Here it is already too late for 17…d8?!: 18 he1+ d7 19 b4! a3 20 d3 g6 21 e3 a7 22 b5 d4 23 f1. The position of the king on d7 seems inconvenient for Black (Morovic – Campora, Dubai 1986).

18 xe7 xe7 19 he1+ f6 20 d6+ g5!? 21 g1+ f4 White has got the most out of the position but there is no hiding the pawn weaknesses on the king’s flank. Black’s counterplay should be enough for a draw. (Potkin – Kazakov, Moscow 1998).

b) Black’s task is more complicated after 13 b5!? 0-0 14 xd5 xd5 15 xd5

In the game Onischuk – Kutsin (Nikolaev 1995) Black did not choose the best order of moves to transpose to the endgame: 15…c7?! 16 d7! fc8 17 b1 f6 18 xc7 xc7 19 xc6! Both 19…bc 20 b3, and 19…xc6 20 d7 b6 21 c1 leave no doubt: the endgame is highly unpleasant. Probably upon best defence Black is capable of defending this position, but he would not want to reach it again.

Therefore worth considering is 15…b4!? Here the transfer to an endgame does not give White the advantage: 16 xd8 fxd8 17 xd8+ xd8 18 a3 a6! etc. Winning the b7 pawn leads to a draw by perpetual check: 16 xb7 a5 17 a4 ac8+ 18 b1 c7 19 e4 a6 20 d7 xd7! 21 xd7 xa4 22 xe7 (not possible is 22 xe7? c8) 22…c2+.

This leaves 16 e4, but then 16…a517 c4 f6. The struggle is somehow imperceptibly concentrated around the white king. Possibly it is nothing serious but in any event Black directs the play.

18 a3 (also interesting is 18 b1 ae8! 19 g4 h5 20 f4 d8) 18…c7! 19 f4 (a double-edged move, but otherwise Black simply has a good game, for example, 19 ab?! b5 20 x5 fe8 or 19 b1 c6) 19…a2+ (the knight is in a very dangerous position but how can he concretely trap it?) 20 c2 (20 b1 b6 21 e5 ae8) 20…b6 21 e5 (21 b1 ac8 22 e3 b5 23 b3 c3) 21…xe5 22 xe5 ac8 23 b3 xf2+ 24 b1 b5, and Black is close to victory.

If these variations are correct, then 15…b4 removes all questions about the variations arising from 13 b5!? 0-0 14 xd5 xd5 15 xd5.

c) 13 xd5 xd5 14 xd5 c7 15 b1 0-0

Starting with 10 e3, both sides have made perhaps the most natural and logical moves. A position has been reached which is most important for the evaluation of the whole variation. The initiative is undoubtedly on White’s side but how great is it? In his favour he has the advantage of the two bishops plus chances of developing an attack on the g-file. For his part, Black is able for the present to defend his king against serious trouble, and if nothing comes of White’s attack, the weakness of the f3, f2, h2 pawns can outweigh all other nuances.

Meanwhile White has in prospect to solve a local problem: where to develop the light-squared bishop?

16 f4 Apparently the best. White intends to install the bishop on the h1-a8 diagonal. Other continuations promise less:

16 e2 ad8 17 hd1 xd5 18 xd5 d6 19 h4 f4 20 a3 (on 20 xf4 xf4 21 xb7 unpleasant is 21…d4) 20…xe3 21 fe e8. White’s attack is done with (Korneev – Izeta, Alcobendas 1994);

16 d3 b4 17 h5 xd3 18 xd3 g6 19 c1 b8 20 h6 d8 21 c3 f8 22 xf8 xf8, and Black again defends (Kharlov – Evseev, Kazan 2001).

16…b4 It is unclear how to improve White’s play in the variation 16…ad8 17 g2 xd5 18 xd5 f6 19 d1 g6 20 h4 d8 21 h5 gh!? 22 g1+ h8 (Narciso – Matulovic, Belgrade 2001), but the jump of the knight for some reason is more popular.

17 d4 A critical position for the variation.

Of course, it is possible to simply go back with the knight, thereby tacitly offering a draw: 17…c6. In the games below White rejected the draw, but we see nothing of substance to show there is a winning plan:

18 d1 ad8 (or 18…f6 19 g2 fd8 20 e4 xd1+ 21 xd1 d8, Anand – Karolyi, Frunze 1987) 19 g2 f6 20 a4 c8 21 c1 g4 22 xc6 bc 23 xc6 b8. The extra pawn is on hand, but the weak king b1 is really sick, and the pawn islands on the king’s flank are going nowhere (Shchekachev – Iruzubieta, San Sebastian 1996).

However the main thing is that the tempting idea 17…c6?! does not work. After 18 g1 whichever rook is placed on d8 – White, exploiting the poor position of the enemy king, will quickly aim his forces in the direction of the king g8:

18…ad8 19 g2 g6+ 20 f5! xf5+ 21 e4 a5 22 a3 c6 23 xd8 xd8 24 d4 g5 (Acs – Ruck, Paks 1996), and here it was possible to obtain a great advantage by force: 25 c3 c5 26 b4 e5 27 e1!

Also having its minuses is 18…fd8 19 g2 g6+ 20 f5! xf5+ 21 e4 a5 22 h6 g6 (22…f8 23 xg7! xg7 24 xb4 with a win).

23 xg6! hg 24 xg6+ h8 25 g7+ g8 26 f6+ f8 27 xe7+ xe7 28 e3+ f8 29 h6+ e7 30 e4+ d7 31 d6+ c8 (also losing is 31…c7 32 e7+ c8 33 c1+ b8 because of the quiet 34 f4! xa2+ 35 c1 a1+ 36 d2 xb2+ 37 e1) 32 xb4 f5+ 33 a1 xd6 34 xd6 e6 35 c5+ b8 36 c4 The heavy piece ending is completely hopeless for Black.

However success in a single analytical variation cannot hide the fact that on the whole Black is close to equality in the 10 e3 system. In the overwhelming majority of cases White does not succeed in developing an attack; it will all come down to an endgame in which White has purely academic chances of victory.

Parallel with the theory of 10 e3 our knowledge of 10 d5 has been increased. Here too it gradually becomes clear that quite frankly things are bad for Black…

B

10 d5 After 10…d4 White has a choice between two continuations.

The move 11 d1 leads to boundless complications (in which it is Black who will rather have the chance to confuse the opponent); 11 b5+, which suggests itself, allows a weighty advantage without any ‘ifs and buts’.

1) 11 d1 e5! The move 11…f5 is both illogical (why move away such a splendid knight?) and simply weak: after 12 b5+ d7 13 0-0 g6 14 e1 g7 15 g5 f6 16 d6! finishes things off. The bishop cannot be taken – 16…fg 17 de xe7 (or 17…c8 18 c1 with irresistible threats) 18 d5, while on 16…e5 follows 17 xe5+!

12 de On the other hand he cannot take the support away from the knight – 12 f4?! d6 13 fe xe5 14 e3 leads to a position in which it is only Black who has chances: 14…f5 15 xb6 xb6 16 a4+ d8 17 h3 xc3+ 18 bc e8+ 19 f1 f6 20 f4 e5 (Worley – Marusenko, Newport 2001).

After 12 de on the board we have the most critical position of the whole variation with 11 d1. If Black wants to ‘get to grips’ with the struggle, he needs to have a thorough think about the position reached.

We now have a choice of four continuations. It is possible to take the pawn (12…xe6 or 12…fe), or play a gambit: 12…f6 or 12…c5.

In fact the choice is between two moves. The rest are not very suitable:

a) 12…xe6 13 b5+ d7 14 e3 b4 15 f4 c7 16 xd7+ xd7 17 xd7+ xd7 18 0-0-0+ c6 19 d4 xc3 20 c4+ d5 (Balashov – Sveshnikov, Lvov 1973) 21 xc7. White’s advantage is measured by pawns, and this is only the start.

b) 12…f6?! 13 ef+ xf7 14 g2 b4 (or 14…e8+ 15 e4 b4+ 16 f1 e5 17 d3, Titz – Vizer, Graz 2001) 15 0-0 xc3 16 bc e6 17 f4 ac8 18 e3 hf8 19 e1 with the advantage (Ekstrom – Krizsany, Basel 1999). Of course, Black can improve his play but all the same 12…f6 looks suspect.

c) While here 12…c5!? might prove just the ticket! Let us see how harmoniously Black is ready to develop his pieces. The rooks will occupy the central d- and e-files, the queen jumps over to h4. The compensation may be worth more than the sacrificed pawn.

13 ef+ Refraining from an immediate capture does not solve the problem: 13 b5+ xb5 (also interesting is 13…f8!? 14 e3 xe6 with the better game, Eising – Kuijf, Amsterdam 1984) 14 ef+ f8 15 xd8+ xd8 16 xb5 xf7 17 0-0 c4 18 c3 d3 19 e4 d4. It is obvious that Black will not be struggling for equality (van Wely – Lautier, Monaco 1998).

13…xf7 14 e3 e8 15 d3!? Shirov’s recommendation – 15 e2 h4 16 e4 – cannot spoil Black’s mood. Incidentally, Shirov himself pointed out that Black has at least a draw in hand: 16…xe4!? 17 fe xe4 18 0-0 xe2+ 19 xe2 g6+ 20 h1 e4+.

15…h4! 16 e4 ad8!

The gambit has given up on glory! In the game Grinfeld – Shirov (Budapest 1996) White tried to curb the opponent’s initiative: 17 xd4 (or 17 0-0 d6 18 xd6+ xd6) 17…xd4 18 0-0 d6 19 b3+ f8 20 xd6 xd6, which only partially succeeded. Shirov won this fascinating game, indeed the move 12…c5 is possibly the main weapon against the variation 11 d1 as a whole.

d) 12…fe Not as thrilling, but a more popular continuation.

13 e3 The neutral 13 g2 has been repeatedly tried. Now the thrust (along the lines of the game Grinfeld – Shirov) 13…h4?! does not achieve its objective in view of 14 f4! d8 15 0-0 d7 16 e3 f5 17 b3 g4 18 h1 d6 19 g1 xe3 20 fe with an obvious advantage (Grinfeld – Kuijf, Munich 1992). But the simple 13…e7!? 14 0-0 0-0 15 f4 d7 16 d3 ad8 seems fully satisfactory. After 17 e4 b4! 18 h1 c6! the position is completely equal. (Al-Modiahki – Dzhumaev, Malaysia 1994).

13…c5 Now White has a choice: to continue his development with the move 14 g2 or to take the position to the edge by 14 b4.

d1) 14 b4!? Leading to forced play in which Black’s chances of equalising the game are higher than White’s chances of obtaining the advantage.

In short, even 14…xb4!?, apparently, cannot be refuted: 15 xd4 d5 16 b3 c8 or 15 xd4 xd4 16 xd4 d5 17 c1 c8 18 d2 0-0, and it is not clear how all this will end.

The natural choice after 14 b4 is between 14…0-0 and 14…f6.

d11) 14…0-0 15 bc xf3+ 16 e2 h4!? More often played is 16…f6, but then by transposition of moves we get into into d12.

17 cb It is clear that in positions of this kind the cost of a move increases again and again; correspondingly also the cost of a mistake increases. Thus, in the game Eilertsen – Henriksen (Norway 1990) it was enough to ‘miss’ just one thing – 17 g2?, and in an instant the position becomes difficult: 17…ad8 18 b3 d2+! 19 xd2 d4+ etc.

17…ad8 18 a4 Fainthearted is 18 xd8? xd8 19 d1 h4! 20 ba c4+ 21 d3 e5 with a decisive advantage (Arytunov – Marusenko, Kiev 1998).

In this very sharp position from the game Pisk – Pingitzer (Stockerau 1992), Black rushed to give check with the knight on d4. The attack was ruined. Meanwhile worth considering is 18…h5!? with the sample variation: 19 g2 h4+ 20 f1 (not possible is 20 f3 xf3) 20…xg2 21 e4 (on 21 xg2 strong is 21…g6+ 22 f1 d3+ 23 e2 xe3 24 h4 d2 25 e1 xb6, and White’s pieces are virtually stalemated) 21…h3 22 e2 (it is difficult to evaluate the position after 22 xg2 xe3 23 e4 xb6!?) 22…f4+. Alas, no good is 22…f4?! (hoping for 23 xf4? xf4+ 24 xf4 d3+ 25 e1 xc3+ 26 e2 d3+ with perpetual check) in view of 23 ad1! xd1 24 xf4 xh1 25 ba xf4+ 26 d2. The pawn, which has in a surprising way reached a7, is one step away from a complete triumph.

23 xf4 xf4 24 e3 f5!? 25 ac1 Also in the event of 25 f3 it is difficult for White to avoid perpetual check: 25…xf3!? 26 xf3 d2+ 27 xd2 xf3 28 ba f4+ etc.

25…d3!? How beautiful, also so forced. After this move White, in order to avoid perpetual check, has to give up two pawns (on f2 and b6).

The tempting 25…f8 does not work in view of 26 hf1 (only not 26 ba? xf2+ 27 e1 f3 28 e2 f4 or 28 e4 xc3!) 26…f3 27 d4 f4 28 d6!? xf2+ 29 xf2 xf2+ 30 d1 e3 31 c2 ab 32 d1, and White’s chances are to be preferred.

26 xf4 xf4 27 xd3 f3+ 28 d2 xf2+ 29 e2 xb6 30 hf1 h6 The material correlation is far from standard but we do not think there is any risk of Black losing this position with reasonable play.

d12) 14…f6 (more popular than the queen sortie to h4) 15 bc xf3+ 16 e2 0-0

In the game S.Polgar – Skembris (Corfu 1990) White played 17 g2?!, on which, in the opinion of grandmaster Skembris, 17…c4! was very strong. Obviously there are no alternatives to 17 cb.

17…ad8 It is necessary to add that bad is 17…xc3? 18 g2 c4+ 19 d3 with a great advantage (Zhuravlev – Gutman, USSR 1972).

18 c2 The time has still not come to give up the queen: 18 g2 xd1 19 axd1 xc3 20 xf3 c4+ 21 d3 xf3!, and a draw is not far off (Kuijf – Bersma, Hilversum 1987).

18…d4+ 19 xd4 xd4 20 e4 Not possible is 20 f3?? xf3 with a mating attack (Mayro – Ngyen, correspondence, 1983).

20…xa1 21 g2 e5 22 ba After 22 b1? xh2 23 f1 ab Black has more material and his king is better (L.B. Hansen – Kuijf, Grestel 1990).

Now however Black’s main problem is how to cope with the a7 pawn?

22…b5+ 23 e3

23…a6! The only move! Giving check on b6 was no use – White covers with the queen on c5. Now however everything ends pleasantly for Black. Thus, in the game Rozentalis – Lalic (Moscow 1994) after 24 b1 xa7+ 25 e2 c8 26 b3 a6+ the opponents agreed a draw. The French analysts Prie and Tirard propose as strongest 24 c1, but also here after 24…xa7+ 25 e2 a8!? Black’s chances are in no way worse.

It is clear that in the variation 14 b4, White, though he will obtain one, and then even two extra pieces, is risking slightly less than the opponent. This is why many prefer not to get involved in an exchange of blows, but quietly continue development – 14 g2. However there is simply no quiet life.

d2) 14 g2 h4 Black played the opening superficially in the game Romero – Bersma (Amsterdam 1987): 14…0-0 15 0-0 e5 16 e4 d7 17 xc5 xc5, and after 18 f4! the white bishops dominated.

15 0-0 d6 16 h3 f5 17 e4!? A move that calls ‘for a fight’. The game Malaniuk – Yudasin (Moscow 1991) ended peacefully: 17 b3 0-0 18 xe6+ h8 19 e4! f6 20 g4 f7! 21 f4 c4 22 c1 h6 23 f3 xf4, while the idea 17 b5?! d8 18 xd6+ xd6 in general is not worth considering: after 19 e2 f7! Black is already playing for a win (Winants – Adams, Wijk aan Zee 1995).

On 17 e4 there are four replies. We examine them, from the weakest to the strongest.

Unsatisfactory is 17… xe3?! 18 fe d8, as was played in the game T.Horvath – Hamdouchi (Hungary 1995). After 19 b3 e7 20 f4 d5 21 f5! White’s advantage had grown noticeably.

On 17…d5 White is forced… to exchange all the pieces, apart from the rooks! Let’s look at it: 18 g5 h5 19 f4! xd1 20 xd6+ xd6 21 axd1 h6 22 fe1 f7 23 xd5 ed 24 xd5 hg 25 xd6 gf 26 g2. In this endgame, unpleasant and truly ‘black’ work awaits the second player.

White can reckon on a minimal advantage after 17…e7. He will need to count on the light-squared bishop: 18 f4 0-0 19 xd6 xd6 (things are not essentially changed by 19…xe3 20 fe ad8 21 b3 xd6 22 ad1 fd8 23 xd6 xd6 24 c1) 20 xb6! ab 21 e1. There is not full equality; though it may be slight, everywhere White has a plus.

17…e7!? – this is the strongest move! The bishop must be retained for the attack. Black should not think about material losses, the main thing is to defend the e6 pawn and bring the king’s rook into the battle. Further events could swing about in the following way:

18 xb6 ab 19 b3 f7 20 xb6 hd8 21 fc1 (21 fe1 d4) 21…d7 22 c7 ad8 23 xb7 He could also take the pawn with the queen – Black’s attack is no weaker against this: 23 xb7 d1+ 24 xd1 xd1+ 25 f1 h6 26 a6 g6+ 27 h2 h6 and it is not apparent how White can consolidate his position.

23…d1+ 24 xd1 xd1+ 25 h2 f8! 26 b8+ f7 The most sensible thing is to agree to a draw after a repetition of moves: 27 b7. A few sharp moves – 27 c7?! g6 28 e5 (the threat was 28…d6+! 29 xd6 f4 mating) 28…f6 – and already it is Black who is playing for a win!

For example: 29 xe6 (or 29 c7 g5! 30 xg5 xf2 31 h4 d2) 29…f4+ 30 g3 xg3 31 e8+ h6 32 fg e5!!

Theory knows a great deal about the variation 11 d1, but it does not know the main thing: where is the clear advantage for White? In the variation 11…e5 12 de fe he has extra material, but not a quiet life; in the variation 11…e5 12 de :c5!? White can do little more than think about how not to lose.

The problem is that practice cannot wait until the theoreticians decide among themselves. It happens that to find a desired advantage in an individual variation – practical players there and then have switched to something else. And it turns out that the abandoned variation, as it were, is hanging in the air. This does not mean at all that it will always be bad – simply that at the present moment slightly better ideas are to be found elsewhere.

Today the variation 11 b5+ looks stronger than 11 d1, but who knows what tomorrow will bring?

2) 11 b5+!

11…d7 A forced choice. After 11…xb5 12 xb5 a6 (the threat was 13 f4 c8 14 xa7) 13 c3 Black has many such possibilities, but an acceptable one among them is not to be found:

13…c7 14 e3 d7 15 c1 e5 16 a4+ d7 17 e4 g6 18 0-0 f5 19 e6!, and Black cannot save himself (Bashkov – Magomedov, Chelyabinsk 1990);

13…c8 14 0-0 d7 15 xb7 g6 16 e1 b8 17 xa6 g7 18 g5 with an easy win;

13…d7 (relatively best) 14 xb7 g6 15 0-0 g7 16 e1 0-0 with some chances of continuing the struggle. However, after 17 g5! White’s advantage is still very great (Rozentalis – Adams, Hastings 1997).

12 a4 xb5 Also this move is forced – let’s investigate why.

The f3 pawn cannot be taken: 12…xf3+? 13 f1 e5 14 f4 a6 15 xd7+ xd7 16 d6 b5 17 d4 f6 18 e1 d7 19 g1 c8 20 xg7! (Bashkov – Marusenko, Polica 1992).

The main boost to the variation is the fact that Black is deprived of the defence 12…e5? 13 de xe6 in view of

…14 g5! This surprising blow finishes off the game: 14…xg5 15 0-0-0, and Black resigned (Bologan – Borges, Linares 1999).

13 xb5 g6 As shown by the game Alburt – Dorfman (Erevan 1975), weak is 13…e5? 14 de fe. White achieves an advantage in the most natural way: 15 e3 c7 16 c1 etc. 14 0-0 A healthy move. White does not need to provoke complications and, even more so, look for them. Thus, there is no need to take the pawn at once – 14 xb7?!, as after 14…g7 15 0-0 0-0 Black has sufficient compensation. The immediate 14 g5 looks more interesting, though White will hardly manage to save a tempo on short castling.

14…g7 15 e1! Feeling for the right idea: the e7 pawn ought to be attacked by the rook e1 instead of the bishop g5.

In the present position the move 15 g5 is trappy: 15…h6? is not possible in view of 16 xe7! xe7 17 b4+ e8 18 ae1+ e5 19 f4 h4 20 e4 winning (Von Gleich – Fette, Hamburg 1987). But after 15…0-0 15 e1 it all returns to the channels of the main variation.

15…0-0 16 g5 The tabiya of the variation 11 b5+.

It seems that all Black’s attempts to avoid trouble have been no help. Here are just a few paths of fruitless endeavour:

16…e5?! 17 e3 e8 18 f4 d7 19 xb7 – White has an extra pawn and the opponent has no compensation (Dvoretsky – Izeta, Terrasa 1996);

16…f6 17 xb7 (it is not clear how to react to 17 ad1!?) 17…b8 18 xe7 xb2 19 xd8 xd8 20 ad1, once again with a healthy extra pawn (anlysis by V. Chekhov);

16…f6 17 f4 e5 18 e3 c8 19 g3 g5 20 xe5 fe 21 b4 f7 22 d1 a5 23 g4, and it remains only to complain about the fate of the bishop g7 (Korneev – Moreda, Malaga 2001);

16…e8?! 17 d6 f6 18 d5! (it is obvious that Black has no available resources) 18…ed (no help are either 18…fg 19 de c8 20 ac1, or 18…e6 19 c7 a6 20 b3 fg 21 xe6!) 19 xe8+ xe8 20 c7 e5 21 xd7 xg5+ 22 h1 with the unstoppable 23 e6 (Stripunsky – Gershov, New York 2000).

16…f6 Even quite recently this move was considered relatively promising and in any case – acceptable. But now it is hopelessly out of date.

17 xf6 ef Now he does not have to defend the weakness on e7. True, in return Black presents the opponent with a passed d-pawn, hoping subsequently to blockade it. The alternative is 17…xf6, but after 18 xb7 e8 19 ad1 d6 20 e4 xe4 21 xe4 it is difficult to persuade oneself that there is real compensation for the material (Sanchez – Pablo, Barbera 1997).

After 17…ef, it seems that a convenient moment has arrived to finally gobble up the b7 pawn…

On the theme 18 xb7!? only one game is known: Cohen – Marusenko (Tel Aviv 2003). But one is soon convinced that it is still too early to take the pawn! This is how events developed: 18…e5 19 e3 c4!? 20 e2 e5 21 b4!? It is more natural to look at 21 f4, but after 21…b8 22 xa7 f3+ 23 g2 Black has available a complicated combination: 23…c8! 24 h3 xb2! Things are not bad for White, but also not brilliant: 25 xb2 xc3 26 b7 h4+ 27 h2 f3+ 28 h1 d2! or 25 e3 h4+ 26 h2 xe2 27 xe2 c4.

21…xf3+ 22 g2 d7 Here White cannot maintain the tension – 23 e7?, after 23…h4+! he would be forced to part with material and lose.

The right defence lies in 23 f4!? e5 (or 23…g5 24 e3 fe8 25 ae1, and White consolidates his forces) 24 d1 fe8 25 e4 Black’s threats have run dry, while it is not easy to defend the f6 pawn.

Here is a sample variation of the unfolding events:

25…g4 (or 25…g4+ 26 xg4 xg4 27 d6 f5 28 d7) 26 h3 f5 27 c5 b5 28 xe8+ xe8 29 hg xc5 30 gf b5 31 d6 xb2 32 d7 d8 33 c7 b6 34 xb6 ab 35 f6, and the white king heads for c7.

One cannot recommend this path for White – there are too many twists and turns. At any moment one could stumble. Far clearer is 18 ad1, counting on the d-pawn and the weakness of the f6 square.

18 ad1! e5 There is no solution to the problem in 18…e8 19 xe8+ xe8 20 xb7 b8 21 xa7 xb2 22 a3 b6 23 a4 c8 24 d4, and by now White has two passed pawns (Shchekachev – Bergez, San Quentin 2001).

19 e3 c8 20 d6 d8 It is a miserable endgame after 20…c6 21 xc6 xc6 22 d7 fd8 23 e4 g7 24 d6 (Gallagher – Krizsan, Lugano 1999). He has to give up the knight for the pawn or else the rook.

But how to break up the opponent’s defence after 20…d8 ? It is premature to play 21 d5 f5 and Black still holds on (Dolmatov – Dyachkov, Elista 1996).

21 d5! An excellent discovery by Romanian grandmaster Mihai Marin which effectively refutes Black’s whole system of defence. In the game Marin – Fressinet (Sitges 1999) there followed 21…xf3+ 22 f1! xh2+ 23 e1 g7 24 xf6! The very moment to lower the curtain.

Indeed, talking about the variation 9…b6, the ending has turned out sadly for Black. Nothing can be done about it – the variation is difficult. White needs only to refrain from trifles (10 e3) and firmly tread the smooth path – 10 d5! d4 11 b5+!

Sometimes it is useful to play dull, correct chess. With the black pieces go over to the endgame and make a laborious draw there.

II

9…e6

Reliable, solid, correct.

10 xb7 xd4 (but no way 10…db4?! 11 b5 c8 12 e3 e7 in view of 13 d5! ed 14 xd5 0-0 15 xc6 xc6 16 xe7+ xe7 17 d1 a5+ 18 b4, and Black’s prospects are bleak, Peng Zhaoqin – Stefanova, Wijk aan Zee 2002) 11 b5+ xb5

12 c6+! An important intermediate move, forcing the black king to occupy an uncomfortable position. Of course White can also play at once 12 xb5+, but without particular success: 12…d7 13 xd7+ xd7 14 xd5 ed 15 e3 b4+ 16 e2 hc8 17 ac1. The game Rozentalis – Bologan (Philadelphia 1994) continued 17…a6 18 d3 g6 19 d4 e6 20 h3 f5 with approximate equality, but possibly the most accurate was 17…d6!?, keeping in his sights the h2 pawn and intending to place the bishop on e5.

12…e7 13 xb5 Interesting but not too convincing is 13 xb5!? However there are more than enough ways for Black to go wrong.

Thus the game Vasyukov – A.Zaitsev (Berlin 1968) ended literally two moves later: 13…a6!? 14 d4 b4?? (14…c8, and White has nothing in particular) 15 g5+!

To put it mildly, Black’s play was dubious in the game Gulko – Ignatiev (Moscow 1969): 13…f6?! 14 g1 b4+ 15 f1 h6

16 b3! The best place for the bishop is the b2 square. It is essential that the a1-h8 diagonal is not covered over: on 16…c8 17 b2+ c3 there is the decisive queen sacrifice 18 xc3! xc6 19 xd5+ f5 20 e3+ f4 21 d1! White is going for a real hunt: 21…d6 22 c1! or 21…g6 22 xd8 xd8 23 e2 h5 24 f6 winning.

Ignatiev played 16…c8 17 b2+ e7, but after 18 d4 g6 19 b5 c3 20 c6+ d6 21 xc3 xc3 22 e5+ xc6 23 xc3+ b6 24 b4+ a6 25 g4! Black can save himself from mate only at too high a price.

In reply to 13 xb5 we can recommend 13…b8!?, for example: 14 d4 d7 15 0-0 c8 16 a6 e8 (interesting is 16…f6!? 17 b3 c3!?) 17 e3 b4 18 e2 d6 19 fd1 d5 20 b5 a6, and Black has nearly beaten back the opponent (Vorisek – Hollman, Czech Republic 1995).

After 13 xb5, on the board is the tabiya of the variation.

In the famous game Fischer – Euwe (Leipzig 1960) was encountered 13…xc3?! 14 bc. It seems that White has a small cartload of weak pawns, but Fischer showed his opponent that the weakest is the isolated pawn on a7: 14…d7 15 b1 d8 (or 15…xb5 16 xb5 d6 17 b7 f6 18 e2 c6 19 f7 a5 20 e3 with the advantage – analysis by Fischer) 16 e3 xb5 17 xb5 d7. Strongest here was an immediate 18 a5; however Fischer won the game also with the simple 18 e2.

Black also failed to equalise in the game Balashov – Hort (Buenos Aires 1980): 14…d5 15 b1!? d8 16 e3 f6 17 g1!?

Later an attempt was made to improve Black’s play by – 14…f6!? But, as shown by practice, this defence does not eliminate the problem: 15 a3+ f7 16 d1!? c8 17 d7+ g8 18 xf8 xc3+ 19 e2 xf8 20 hd1 h5 21 d8 a3

22 1d6! An extremely unpleasant move for Black. Until here 22 1d7 h6 23 xf8+ xf8 was played (Christiansen – Shamkovich, South Bend 1981), and the black queen succeeded in closely covering its king. Now however after 22…xa2+ 23 f1 a1+ 24 g2 the king g8 has to rely on its own agility: 24…f7 25 8d7+ g8 26 xe6 h6 27 f5! (Rantanen – Burger, Gausdal 1982).

13…d7 Essentially, here was White’s last chance to avoid the endgame. Whether it was necessary to avoid it is another question, but a second such chance will not present itself.

If he does not exchange queens, then it is necessary to choose between 14 a5 and 14 e2.

There is no advantage to be had in 14 a5?! xc3 (the modern treatment is – 14…f6!? 15 0-0 xc3 16 bc f7 17 a6 c5 18 f4 hc8, and in any event it is not Black who is fighting for equality, Onischuk – Dreev, Yalta 1995) 15 xc3 (or 15 bc f6 16 b1 f7 17 a6 e7 18 b7 d5 19 a3 he8 20 0-0 xf3, Gaprindashvili – Chiburdanidze, Pizunda 1978) 15…f6

16 e3 f7 17 0-0 e7 (M.Tseitlin – Kasparov, Daugavpils 1978). With queens, the weakness of the a7 pawn is an insignificant factor; it is his free development and prospects of attacking the f3 pawn that are more important.

More interesting is 14 e2!? – White prefers a direct attack on the king to positional niceties. The recipe for defence is already well known to us: the king should hide on f7: 14…f6! 15 xd5+ xd5 16 0-0 f7 17 d1. On principle, it is difficult to establish in this position that Black is threatened with real danger. But accuracy must never be relaxed. Thus, in the game Taeger – Rogozenko (Bad Weissee 1997) the Romanian grandmaster for some reason or other did not take under control the d7 square and after 17…f5?! 18 d7+ e7 19 e3 was forced to part with the a7 pawn. Instead of 17…f5?!, 17…b7 18 e4 b8 19 xb7+ xb7 20 b1 e7 21 e3 c8 looks more healthy. All Black’s problems are behind him (Nirosh – Bageri, Teheran 1998).

14 xd5+ xd5 For Black (with the king on e7) there is absolutely no reason to avoid the exchange of queens. And that is why 14…ed?! is foolish: 15 e2+ (Fischer recommends 15 b4+ e8 16 d4) 15…e6 16 e3, and the black king is again threatened with an attack.

After 14…xd5 arises the first serious fork in the variation 9…e6. White can choose between the immediate 15 xd5 (A) and the intermediate 15 g5+ (B).

A

15 xd5 ed

In this position there are three completely different plans of play for White: 16 e3, 16 f4 and 16 0-0.

1) 16 e3 e6 17 0-0-0 Appropriate, as it was for this reason that White played 16 e3. The attempt to switch ideas (refraining from castling queenside) brings no advantage:

17 c1 b4+ 18 hc8 19 d3 a5 20 xc8 xc8. The bishop endgame is not winning for White, but otherwise Black is left with the only open c-file (Pigusov – Dreev, Tallinn 1986);

17 g1 d6! 18 xg7 e5 19 g4 xb2, once again with equal chances (Rogers – Adams, London 1988).

17…b4 This is how the great specialist in the system – grandmaster Alexei Dreev, likes to play. However also in the event of 17…c8+ 18 b1 c5 White has no advantage: 19 hg1 g6 20 g4 hd8 21 a4 b6 (Gdanski – Adorjan, Polanica Zdroj 1992) or 19 he1 d6 20 d3 hd8 (Kavalek – Rogoff, Berlin 1975) with obvious drawing tendencies.

After 17…b4 the statistics of results of games are somewhere around the 50% mark. It is rare that one of the opponents manages to extract a full point:

Unpromising is 18 d4 f6 19 hg1 f7 and the bishop has to go back: 20 e3 hd8 21 g4 a5 22 d3 d7. White’s position is not worse (Garcia – Becerra, Matanzas 1995);

18 d3 hd8 19 a3 ac8+ 20 b1 x5 21 e1 xe3 22 exe3+ f6 23 d4 c4. White has achieved nothing (Franco – Dominguez, Mondariz 2002);

18 b1 hc8!? 19 d3 c5 20 e1 xe3 21 dxe3+ f6 22 e7 c4. The f2, f3 and h2 pawns feel the draught (Stripunsky – Dreev, Internet 2001).

18 a3 hc8+ 19 b1 c5! The defensive strategy is the same in any case: to force a bishop exchange, since it is well-known that nobody can win the rook endgame.

20 he1 xe3 21 xe3+ d6 22 f4 ab8 23 d4 with a draw (Vaganian – Dreev, Odessa 1989).

2) 16 f4 Far more interesting than 16 e3. Now forcing an exchange of bishops is considerably more difficult; Black can no longer allow himself to play ‘on general principles.’

16…f6!? The other choice is 16…d7!? 17 0-0-0 c6, and he is all ready for f8-d6.

17 0-0-0 d8 18 hg1 d7 19 e3! h6 In the game Kindermann – Lobron (Berne 1990) Black did not properly appreciate the bishop transfer and played 19…g8?! There followed 20 g4! and without delay the rook was swept over to a4 – White had an obvious advantage.

20 d4+ (also here it is perhaps worth thinking about the manoeuvre g1-g4-a4!?) 20…f5 21 xg1 xg7 22 xg7 c8+ 23 d2 f4 24 d3 xf3 25 e1 f4 26 h3 c4 27 e3 f5 28 f3+ e6 29 g4 xg4 30 hg b7 Though ‘more pleasant’ for White, the game did not leave the drawing zone (Onischuk – Dreev, Moscow 2002).

3) 16 0-0 On principle, the most logical move. The rooks combine for an attack on the a7 and d5 pawns (likewise they would not mind seizing the open c-file), while the king is brought over for defence of its own pawn weaknesses on the f and h-files.

Black ought to be able to defend the most vulnerable points in his position – besides a7 and d5, such should be considered all the squares on the 7th rank. From there comes the first link in the plan – the manoeuvre a(h)8-d8-d7! Then, by developing the bishop on f6, he is ready to blackmail the opponent with an advance of the d-pawn. On d5 this is a weakness, but if it gets as far as d3…

Moreover it is useful to fix the enemy pawns on the king’s flank by means of g7-g5.

16…e6 17 e1+ f5 White is at a crossroads: 18