Sergey Ivanov

The French in Linares, farther everywhere...

As is known, the match against the Hungarian grandmaster P. Leko was extremely unfortunate for the fourteenth World Champion A. Khalifman. Alexander lost three Black games, trying different openings for the invariable 1.e4, but none of them, neither 1...c5, nor 1...e5, nor 1...e6 was lucky for him.

And now during the preparation for the super tournament in Linares the choice was made, it fell on the French Defence! Moreover, Khalifman decided to play this opening not just to defend, he was ready to go deep into arising complications. The result speaks for itself: Alexander made three draws of four Black games in Linares and lost one to Anand, but the opening had nothing to do with it. The success was increased in the recent tournament in Indonesia, where Khalifman's score in the French was 2/2 (not counting the quick draw with J. Ehlvest).

Thus we can state that the World Champion has found an opening that agrees with his style and lets him win in highest rated tournaments. I invite the readers to consider more closely the last games, played by A. Khalifman in the French Defence: maybe they will help you to reappraise this interesting opening.

1. A. Shirov - A. Khalifman [C02]
Linares, 2000

1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Nc6 

If Black is not going to allow the structure that was introduced into practice by W. Kupreichik (5.Be3 Qb6 6.Qd2) to appear, he can play 4...Qb6 at first, and then get the needed position with rearranged order of moves after 5.Nf3 Nc6. By the way, Khalifman chose this very order in the game we are regarding: he still got the position he wanted, having restricted the resources of his opponent at the same time.

5.Nf3 Qb6 

The line with 5...Bd7 was regarded in detail in the comment of the game E. Alexeev - S. Ivanov (St.Petersburg, 1999) - ref. to "Chess Petersburg" No 2-1999.

6.a3

The chief ideologist of this system, grandmaster E. Sveshnikov, plays so at present. In the lines with 6.Bd3 and 6.Be2 there were found fairly sound ways to gain an equal play for Black.

6...Nh6!?

This is our counterblow. The black knight aims at the key square f5 in order to attack the pawn d4, yet it gets there by a not quite traditional route (because 6...Nge7 would be followed with 7.dc5 with the initiative). This plan was introduced into the modern practice by grandmaster S. Lputian, and the theory of this line is based chiefly on his games. Those who is interested to know more about this line can refer to my survey in the NiC, issue 52-1999. It is curios that this move is not even mentioned in the fundamental work "French Defence" by B. Zlotnik, published in 1982 in the "Chess Bulletin" (No 2 and No 4). In those years Black used to continue 6...c4, 6...a5 or 6...Bd7 without real equalisation.

7.b4

7.Bh6? Qb2 8.Bc1 Qa1 9.Qc2 c4! would be a mistake, to be followed by Na5, Bd7, Nb3, and the black queen breaks free. Neither 7.Bd3 cd4 8.cd4 Nf5 9.Bf5 ef5 10.Nc3 Be6= can give anything to White.

7...cd4 8.cd4

At the early stage of the development of this variant it was considered that White gains an advantage by mere 8.Bh6 gh6 9.cd4.

Still, later on it was proved that the gained opportunities for activating black pieces (the rook can enter the game by the way of Rh8-g8-g4, while the bishop brings pressure on White's centre after Bf8-g7 and f7-f6) give a sufficient compensation for the spoiled pawn "hair-do". It can be developed as follows: 9...Bd7 (An immediate 9...Rg8 10.Nc3 Rg4 11.h3 Rf4 12.Bb5 a5 13.O-O! ab4 14.ab4 Ra1 15.Qa1 is possible as well: White has initiative (Keitlinghaus - Blauert, Budapest, 1998)) A) Now the natural 10.Nc3? will be encountered with 10...Nb4! 11.ab4 (11.Rb1 Qc7-/+ (Yee - Akopyan, Los Angeles, 1999)) 11...Bb4 12.Qd3 Rc8 13.Rc1 Qa5 14.Kd2 O-O 15.Be2 Bc3 16.Rc3 Bb5-+; B) After 10.Be2 the move 10...Rg8!? looks good (10...Rc8 11.O-O Bg7 12.Qd2 O-O 13.Ra2 f6 14.b5 Na5 15.ef6 Rf6 16.Ne5 Be8 with complicated play is possible, too (Bosch - Lputian, Wijk aan Zee, 1999) 11.O-O Rg4 12.h3 Rf4 13.g3 Rf3 14.Bf3 Nd4 15.Bh5 Bg7 16.Re1 Rc8 with a compensation for the sacrifice by an exchange (Nikitin - Flor, Moscow, 1957).

8...Nf5 9.Bb2

Another opportunity is 9.Be3

Now it is hardly worth to recommend 9...f6 10.Bd3! (Black's calculation was based on 10.b5 Ne5! 11.de5 Ne3 12.fe3 Qe3 13.Qe2 Qc1 14.Qd1 1/2-1/2 (Romanishin - Lputian, Erevan. 1988)) 10...Ne3 11.fe3 fe5?! (better 11...Bd7+/= ) 12.b5 Nd4 (or 12...e4 13.bc6 Bd6 14.O-O O-O 15.Ng5!+/- (L. B. Hansen - Antonsen, Farum, 1993)) 13.ed4 e4 (Otero - Nogueiras, Santa Clara, 1999), because after the right 14.Be4! de4 15.Ne5 g6 16.O-O White gains a considerable development advantage that repays with interest the missing paw. 9...g6!? 10.Bd3 Ne3 11.fe3 Bh6 12.Qe2 Ne7 13.Nc3 Bd7 14.O-O Rc8 15.Rac1 Nf5 with equal play (Wall - Lputian, Montecatini Terme, 1999) appears to be more reliable for Black .

9...Be7

This is more precise than 9...Bd7, because 10.g4! forces the black Knight to retreat: 10...Nh6 (or 10...Nfe7 11.Nc3 h5 12.g5 Nf5 13.Na4 Qd8 14.Bd3 h4 15.Rc1, White having a noticeable space advantage (Shirov - Kramnik, Monaco, 1997) 11.h3 f6 12.Bd3 Nf7 13.Nbd2 fe5 14.de5 Be7 15.Rc1+/= (Sveshnikov - Lputian, Tilburg, 1992); 9...a5!? is interesting, too. In the game Sveshnikov - Bareev (Moscow, 1995) White obtained an advantage after 10.b5 a4 11.g4 Nfe7 12.Nc3 Nb8 13.Bd3 Nd7 14.O-O Ng6 15.Rc1 Be7 16.Na4. Later on Black managed to reinforce his play: 11...Nh6!? 12.h3 Be7 13.Nc3 Na5 14.Na4 Qd8 15.Rc1 Nc4!? 16.Bc4 dc4 17.O-O O-O 18.Rc4 b6 19.Nc3 Bb7 20.a4 f5 with sufficient compensation for the two pawns (Johnsen - Kallai, Budapest, 1999).

10.Bd3

Now 10.g4 is less convincing because of 10...Nh4; while 10.h4 appears to be too pretentious, because White is fairly underdeveloped. After 10...Bd7 11.g4 Nh6 12.Rg1 f6 13.ef6 gf6 14.Bd3 Nf7 15.Qe2 (Svidler - Belyavski, Novosibirsk, 1995) Black can favourably complicate the play with 15...e5!?. Of course, an unassuming development of the bishop 10.Be2 is possible, too, but then Black has simply no problems. There are several good plans for him:

10...Bd7 11.O-O

A) 11...h5!? 12.b5 (or 12.Qd2 g5! (struggling for the square d4!) 13.Rd1 g4 14.Ne1 Rg8 15.Nc2 Bg5 16.Qe1 Rc8-/+ (Relange - Berend, Brussels, 1993)) 12...Na5 13.Nc3 a6 14.Na4 Qd8 15.ba6 ba6 16.Bc3 Bb5= (Kozak - Dolmatov, Elista, 1997); 

B) 11...O-O 12.Qd2 (preventing Nc6-a5) 12...f6 (also 12...a5 13.b5 Na7 14.a4 Nc8 15.Nc3 Qd8 is possible to be followed with Nb6, Rc8, Bb4, etc.) 13.g4 Nh6 14.ef6 Rf6! 15.g5 Rf3 16.Bf3 Nf5 17.Rd1 Rf8 with a fierce initiative for the sacrifice by an exchange (Cools-Motwani, Vissingen, 1996).

10...a5!

This way only Black can count on an equal play. 10...Bd7 is worse because after 11.Bf5 ef5 12.Nc3 Black has to spend a tempo for 12...Be6

11.Bf5 

Black's idea is tactically based on the following line: 11.b5 Ncd4 12.Nd4 Nd4 13.Qg4 Nf5! 14.Bf5 Qb5! 15.Bd4 ef5 16.Qg7 Rf8 17.Nc3 Qc4 18.Ne2 (Haba - Glek, Stare Mesto 1992), now Black gains a decisive advantage with 18...Bd7! 19.Rc1 Qd3! 20.Rd1 Qb5 21.Nc3 Qc4! (pointed out by I. Glek).

11...ef5 12.Nc3 Be6 13.b5 a4! 

This is an important moment: Black wants to gain more space on the queenside and reserve the square a5 for manoeuvres of his pieces.

14.bc6

A new idea. Continuations that occurred previously not only promised no advantage to White, they often couldn't ensure a parity: 14.O-O Nb8 (14...Na5!? 15.Qa4 O-O with the idea of Rfc8, Nc4 with a compensation is interesting, too) 15.Bc1 Nd7 (15...h6!?) 16.Rb1 (16.Bg5!?) 16...O-O 17.Na4 Qa5 18.Nc5 Nc5 19.dc5 Bc5+/= (Campora - Dokhoyan, Wijk an Zee, 1989); 14.Qd3 Nb8 15.Bc1 h6! 16.Ng1 Nd7 17.Nge2 Qc7 18.Bd2 Nb6+/= (Sveshnikov - Moskalenko, Norilsk, 1987); 14.Ra2 Nb8 (14...Na5!?) 15.O-O Nd7 16.Na4 Qb5 17.Nc3 Qc4 18.Qb1 O-O 19.Rc1 Rfc8= (Grosar - Tabernig, Montecatini Terme, 1997)

14...Qb2 15.O-O

White doesn't stop at the sacrifice of a piece. He had yet another interesting opportunity: 15.Na4 Ra4 16.cb7 The white pawn that has got to b7 is very strong. The play should be forced now with 16...Qc3 (16...O-O? 17.O-O! is a mistake, as there is no defence from 18.Rb1, winning the game) 17.Nd2 O-O 18.Rb1 Rd4 19.b8Q Re4 20.Kf1 Qd3 21.Kg1 Rb8 22.Rb8 Bf8 brings to a position where Black, having lost by an exchange, has a good compensation, consisting in active stands of his pieces.

15...bc6

Why does Black not take a piece - 15...Qc3? It proves that after 16.cb7 Rb8 17.Qa4 Bd7 18.Qa7 A) 18...O-O 19.Rfc1 Qb2 20.Rcb1!? (20.Rab1 Qa2 21.Ra1 forces to repeat moves) 20...Qc3 21.a4 with initiative.; B) 18...Qc7 White has an important resource of 19.e6!, introducing the knight into the play. (19.Rab1 Bc6 20.Rfc1 Rb7!-/+ would be a mistake).

Now White has a serious initiative in all lines: 19...Be6 20.Rab1 O-O 21.Rfc1 Qd6 22.a4 with compensation; 19...fe6 20.Rab1 Bd6 (20...Bc6 21.Ne5 O-O 22.Qa6 Bb7 23.Qe6 Kh8 24.Rfc1) 21.Rfc1 Qd8 22.Qa6 Bf4 23.Rc5 O-O 24.a4 with compensation; 19...Bb5 20.Rfb1 Qb7 21.Qa5 Bd8 22.Qe1 with initiative.

With the move in the game Black secured himself against any troubles and got even a more pleasant play.

16.Na4 Qb5 17.Nc3

17.Nc5!? Bc5 18.dc5 Qc5 19.Nd4 with a positional compensation for the pawn is interesting.

17...Qc4 18.Ne2 O-O

After 18...Ra3 19.Ra3 Ba3 20.Qa1 O-O 21.Qa3 Qe2 Black should not count on an advantage.

19.Rc1 Qa6 20.Rc3 Rfc8+/= 

A draw was achieved in this position. Owing to the advantage of two bishops and the weakness of the white pawn a3 Black has minimal advantage, but no real grounds to play for a win. Thus we can conclude that Shirov's novelty has not enfeebled Black's position, and we can continue to consider this variant to be reliable enough.

2. P. Leko - A. Khalifman [C17]
Linares, 2000

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Ba5 

Very long this line stayed in the background of the theory. We owe its revival to M. Botvinnik who began to apply it in the forties and had several convincing wins. In 1954 M. Botvinnik succeeded in upholding this system in the match with V. Smyslov for the title of the World Champion. At present the main apologists of the move 5...Ba5 are Armenian grandmasters R. Vaganian and S. Lputian, this is why I have called the concerned system "Armenian Variant" in my survey, published in the NiC (issue 43-1997) 

6.Qg4 Ne7 7.dc5

This is not the basic continuation, yet it is rather malicious. 7.Qg7 Rg8 8.Qh7 cd4 9.b4 leads to a position that will be regarded below; 7.b4?! is unfortunate in this redaction because of 7...cb4! 8.Nb5 b3! 9.c3 Nf5 10.Bd2 O-O 11.Qh3 a6 12.g4 Qh4!+/= (Timman - Vaganian, Horgen, 1995).

7...Bc3 8.bc3 Ng6

Black can chose now from a variety of opportunities:

A) 8...Nd7 9.Nf3 (9.Qg7 occurred already in one of founder games of the variant: 9...Rg8 10.Qh7 Ne5 11.Be2 Qa5! 12.Bd2 Qc5 13.Nf3 Nf3 14.Bf3 e5 15.Bh5 Bf5!-/+ (Reshevski - Botvinnik, Moscow, 1946))

A1) Yugoslavian Encyclopaedia recommends 9...Ng6 10.h4 h5 11.Qg3 Nc5 with unclear play. In the game Spiriev-Mueller (Budapest,1991) 12.Ng5 Qc7 13.Be3 Nd7 (13...Ne4!?) 14.Bd4 Nde5 15.Bb5 Kf8 16.O-O f6 17.Rae1 followed, White having good attack prospects. However, White can reduce to the main line with 10.Bd3.;

A2) 9...O-O 10.Bd3 f5 11.ef6 Nf6 12.Qh4 The control of the key square e5 ensures White's opening advantage. Further opportunities: 12...Nc6 (or 12...Qc7 13.Bf4 Qc5 14.Be5 Nf5 (King - Lputian, Dortmund, 1988) 15.Qb4+/- - D. King) 13.c4!? Qa5 14.Bd2 Qc5 15.O-O dc4 16.Bc4+/= (Lau - Lputian, Altensteig, 1989);

A3) 9...Qc7 Recommended by P. Keres. After 10.Qg7 Rg8 11.Qh7 Ne5 12.Qh5! (12.Bb5 Bd7 13.Bd7 Nd7 with unclear play is weaker) 12...Nf3 13.Qf3 Bd7 14.Bf4 Qc5+/= White's chances are some better owing to the possession of two bishops.;

B) 8...Qa5!? Attacking white pawns, Black contains his opponent's forces and prevents him from a quick development of his initiative. 9.Bd2 Ng6 10.h4! (10.Nf3 Nd7 11.c4 Qc7! 12.cd5 O-O! 13.d6 Qc5 14.Qb4 Nge5= (Tseshkovski - Lputian, Podolsk, 1990) 10...h5 11.Qg5 (11.Qb4 Qb4 12.ab4 Ne5 13.Bf4 Nbc6 14.b5 Ng6 15.Bc7 Nce5 with unclear play (Filipenko - Tondivar, Groningen, 1993)) 11...Nd7 12.c4

B1) 12...Qa4 13.cd5 Nde5 ((another opportunity is 13...Qe4 14.Ne2 Qd5 15.f4 Nc5 16.Nc3 Qd4 17.Rb1+/= (Short - Lputian, Manila, 1990)) 14.Be2 (Short - Timman, Amsterdam, 1991) 14...ed5!? 15.Nf3 (15.Bh5 Rh5! 16.Qh5 Bg4 17.Qg5 Qc2 with compensation - L. Ftachnik; 15.f4 f6 16.Qg3 Ng4) 15...f6 16.Qe3 O-O with a complex play;

B2) 12...Qc5!? 13.Nf3 dc4 14.Bb4 Qd5 with counterlpay 15.Rd1 Qe4 16.Be2 f6 17.ef6 gf6 18.Qe3 Qe3 19.fe3 Nb6 20.Nd2 Ne5 21.Ne4 Kf7 22.O-O Nd5, and White's initiative, at most, makes up for the missing pawn (Kruppa - Lputian, Moscow, 1991)

9.Bd3 Nd7

9...O-O 10.Nf3 f5 11.ef6 Qf6 12.Qd4 deserves attention (Savon - Legky, Odessa, 1989) (12.O-O e5 13.Qh5 Bf5! with counterlpay) 12...Nc6!? 13.Qf6 gf6 14.c4 d4 15.Be4 Rd8 with counterlpay - V. Savon

10.Nf3 Qc7 11.O-O

White is forced to sacrifice the central pawn: after 11.Qg3 there is 11...b6 (11...Nc5 is not bad, too) 12.h4 Nde5 13.Bg6 Nf3 14.gf3 Qg3 15.Bf7 Kf7 16.fg3 bc5, equalising the game (pointed out by S. Atalik)

11...Nde5

All the previous play of Black was based on the attack of the e5-pawn, but maybe it was worth to change and play 11...Nc5 ? Of course White has initiative after 12.Be3 (or 12.h4 ) 12...Nd3 13.cd3 Qc3 14.d4, but Black has an extra pawn and a solid position.

12.Ne5 Qe5 13.Bb5 Bd7 14.Bd7

Both 14.Rb1 Bb5 15.Rb5 O-O! and 14.Qa4 Qc7 are weaker, White gains nothing.

14...Kd7 15.Qa4!

This natural move proved to be a novelty, and a very unpleasant one, this time.

15.Rb1 occurred previously, and Black had time to introduce the rook h8 into the game: 15...Rhb8 16.f4 (16.c4 Qe4 17.Qh3 d4 18.Qh7 Rh8 19.Qg7 Rag8 20.c6? Kd6!-+ Fogarasi - Atalik, Budapest, 1998) 16...Qf5 17.Qe2 Ne7 18.g4 Qe4 (Hachian - Lputian, Panormo, 1998) 19.Qe4 de4 20.f5 with unclear play.

15...Ke7

Probably 15...Kc7 16.Rb1 Rhb8 deserved to be chosen, though now Black's way is not all roses, too. Black wants to exchange the queens, while White wants to open up files and get to the opponent's king. S. Atalik recommends 17.Bd2 (17.Be3 Qc3; 17.c4 Qd4 18.Qa5 b6) 17...Qe2 18.Rfd1 Qc4 Black seems to be holding, though it's not excepted that White's play can be reinforced.

16.Qb4! Rab8

Now 16...Rhb8 does not help already because of 17.c6 Qd6 18.cb7 Qb4 19.ba8Q! Qb6 20.Be3 with White's win. 16...Kf6 17.Be3 a5 (17...Qe4 18.Bd4 e5 19.Rae1 Qc2 20.f4!ќ) 18.Qb7 Qc3 19.Rad1 with a dangerous attack by White appears to be dubious, too.

17.f4

White acts rather vigorously. Of course, there is no sense now to play 17.c6?! Qd6, because Black has no particular problems in the endgame. A quiet move 17.Be3!? deserves attention, too, making it not easy for Black to coordinate his pieces.

17...Qe4 18.f5 Qb4

An exchange of queens is usually not in Black's favour in this redaction: White's pawn structure improves, while his rook begins to work on the opened a-file. 18...ef5 19.Bg5 Kf8 looks rather dangerous, but White seems to have nothing concrete here.

19.ab4 ef5 20.Rf5

White has a clear advantage after 20.Ra7 Ke6 21.Re1 Ne5 22.Bf4 f6 23.h4+/- as well.

20...Ke6

21.Rf2 

Missing a considerable part of his advantage. 21.g4 f6 should have been played (21...a6 is bad because of 22.Bf4 Nf4 23.Re1! - White is deprived of this opportunity after the move in the game) 22.Ra7 Ne5 23.Bf4 g6 24.Be5 fe5 25.Rf1+/-

21...a6 22.Bf4 Nf4 23.Rf4 f5 24.Rd4 Rbd8 25.Re1 Kf6 26.Red1 Rhe8

Black has activated his forces and managed to bring the opening to a draw owing to the precise defence. We can conclude that this game has brought up a number of problems to Black.

3. G. Kasparov- A. Khalifman 
C17
Linares, 2000

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Ba5 6.Bd2

A quiet continuation with ideas like in the line after 5.Bd2.

6...Nc6

The most precise order of moves. 6...Ne7 is not good (being the chief reaction to 5.Bd2) because of 7.dc5! Bc3 (7...Nbc6 8.b4 Bc7 9.f4+/-) 8.Bc3 Qc7 9.Nf3 Nd7 10.Bd3 with a positional advantage by White (V. Damianovich - Stoyanov, Belgrade, 1993; Encyclopaedia of Chess Openings recommended 6...cd4 7.Nb5 Nc6?! (7...Bc7 is better), which practically has been disproved in the game Milos - S. Ivanov (Capelle la Grand, 2000): 8.Ba5 Qa5 9.b4! Qd8 10.f4 Nh6 11.Nd6 Kf8 12.Nf3 f6 13.Bd3! fe5? (better 13...Nf7 14.Nf7 Kf7 15.O-O+/-) 14.fe5 Ne5 15.Ne5 Qd6 16.Qh5 Kg8 17.O-O Qe7 18.Rf3 Bd7 19.Raf1 Nf5 20.Nd7 Qd7 21.Bf5 ef5 22.Rf5+-.

7.Qg4

Curiously enough, this natural move has not occurred in this position before. The usual continuation is 7.Nb5, yet after 7...Nd4! 8.Nd4 (8.Nd6?! Kf8-/+; 8.Ba5 Qa5 9.b4 Qb6 10.Nd4 cd4 11.Nf3 Ne7 12.Qd4 Qd4 13.Nd4 Ng6= (Tseshkovski - Budnikov, Vladivostok, 1990) 8...cd4 9.Bb5 (9.Ba5?! Qa5 10.Qd2 Qd2 11.Kd2 f6 12.f4 fe5 13.fe5 Nh6 14.Nf3 O-O+/= (Diaz - Arencibia, Cuba, 1990); 9.Nf3?! Bb6 10.Bd3 Ne7 11.Ng5 h6 12.Qh5 g6 13.Qh3 Nc6 14.f4 Qe7 15.O-O Bd7 16.Nf3 O-O-O+/= (Richmond - Hug, Debrezen, 1992) 9...Bd7 10.Bd7 Kd7 11.Nf3 Bb6! 

(stronger than 11...Bd2 12.Qd2 Qb6 13.O-O Ne7 14.Nd4 Rhc8 15.c3 h6= (Rodgers - Korchnoi, Biel, 1986) Black can face the future fearlessly, notwithstanding the "centralised" king, because his pawn d4 (which is an extra pawn, in addition) hampers the development of White's initiative rather seriously. Further opportunities: 12.O-O (12.c3 dc3 13.Qa4 Kc7 14.Bc3 Rc8 15.Qg4 Ne7 16.Qg7 Qg8! 17.Qg8 Rhg8+/= (Chabrilo - Dimitrov, Cavala, 1990) 12...Ne7 (12...Rc8 13.Rc1 Kc7 14.Bg5 Ne7 15.Nd4 Kb8= is good, too (Abdullah - Hug, Biel, 1990) 13.a4 a6 14.Ng5 Qe8 15.Qf3 Bc7 16.Qb3 (16.Qf7 Qf7 17.Nf7 Rhf8 18.Ng5 h6 19.Nf3 Nc6 20.Rfe1 Rf3!? 21.gf3 Ne5 with compensation (Madl - Luether, Hastings, 1994) 16...h6 17.Nf3 Nc6 with mutually critical play (King - Neverov, Baku, 1986).

7.Nf3 was tested, too, for example 7...Nd4 (7...Nge7 8.dc5 Ng6 9.Bb5 Bd7 10.Qe2 Qb8 11.Bc6 bc6 12.O-O with unclear play (Gurgenidze - Bronstein, Leningrad, 1960) 8.Nd4 cd4 9.Nb5 Bc7 10.f4 Nh6 11.Bd3 O-O 12.Qh5 f5 13.Bb4 Rf7 14.h3 Bd7 15.Nd4 Bb6 with a complex play (Ehlvest - Lputian, Minsk, 1987).

At the same time, it is hardly worth to recommend 7.dc5 Ne5 8.Nb5 Bc7! 9.Bc3 (better 9.Nc7 Qc7 10.Bf4 Nd3 11.Qd3 Qf4=) 9...Nf6 10.Nf3 Ned7!+/= (Zlotnik - Legky, USSR, 1988).

7...Kf8

The move Kf8 is seldom good in "French" positions, but here it is an exclusion from the rule. The point is that White is forced to surrender the centre now, and Black pieces (except the rook h8, naturally) can take very convenient stands.

In case of the natural 7...Nge7 there is already 8.dc5 (but no 8.Qg7 Rg8 9.Qh7 cd4 with an initiative) 8...O-O 9.Nf3 Ng6 10.Bd3, and Black has problems.

8.dc5 Ne5 9.Qg3 Ng6 10.O-O-O

10.Nf3 suggests itself, but after 10...Bc7 11.Qg4 e5! it is not so easy for White to find a safe position for his queen. After the concerned game was played 10.Bd3 Bc7 11.f4 Nf6 (11...Qf6!? 12.Nge2 Bd7 13.O-O N8e7) 12.Nge2 Bd7 13.h3 Bc6 14.Qf2 has occurred (Mitkov - Zaja, Pula, 2000) 14...Ne4!? with unclear play.

10...Nf6 11.f3 

White takes the square e4 under his control and prepares a retreat to f2 for his queen. After 11.Bd3 there is again the unpleasant 11...Bc7, while 11.Nf3? just loses because of 11...Bc3 12.Bc3 Ne4-+

11...Bd7 12.Nge2 

12.h4 Bc7 13.Qf2 h6 14.h5 can be parried with 14...Nf4.

12...Bc7 13.Qf2 b6

The hampering pawn c5 should be exchanged.

14.Be3 Ne7

14...Kg8 deserves attention, to be followed with h6 (or h5) and Kh7.

15.g4 bc5 16.Bc5 Bb6 17.h4 Bc5 18.Qc5 Qb6 19.Qb6 ab6

The game has proceeded to the ending. Black has a strong centre, but the rook h8 is still out of play. The chances are approximately equal. Thus, White's novelty (7.Qg4) is rather interesting, but Black has good counterchances after the right answer 7...Kf8.

4. V. Anand - A. Khalifman [C17]
Linares, 2000

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Ba5 6.b4

This basic method for struggling against the system 5...Ba5 was introduced by A. Alekhin in the twenties.

6...cd4 7.Qg4

The most critical way, involving a mutual destruction of flanks. The game becomes a tactical character, and Black can hope for a good counterplay if he defends successfully. Recently 7.Nb5 Bc7 8.f4 occurred most frequently. One of the latest examples is 8...Bd7 9.Nc7 Qc7 10.Nf3 a6 11.a4 Ne7 12.Bd3 Nf5 13.O-O Nc6 14.Qe1 Qb6 15.Rb1 Nce7 16.b5 ab5 17.ab5 Ne3 with unclear play (Anand - Lputian, Wijk aan Zee, 2000)

7...Ne7

7...Kf8 seems to be illogical in this position: 8.ba5 dc3 9.Nf3 Ne7 10.Bd3 Nd7 11.Qb4 Qc7 12.O-O Nc5 13.Qc3+/- (Unziker - Botvinnik, Amsterdam, 1954).

8.ba5

An alternative continuation 8.Nb5 is regarded below.

8...dc3 9.Qg7 Rg8 10.Qh7 Nbc6

10...Nd7 has been out of use for a long time, because the knight is not active here, hindering the development of pieces only: 11.f4 Qa5 12.Nf3 Nf8 13.Qd3 Bd7 14.Ng5 Rc8 15.Rb1 b6 16.h4+/- (Gligorich - Dueckstein, Zagreb, 1955)

11.f4

Black can get more chances from 11.Nf3 Qc7 12.Bf4 (12.Bb5 Bd7 13.O-O Ne5! 14.Ne5 Qe5 15.Bd7 Kd7 16.Qd3 Rac8! 17.Qb5 Nc6 18.g3 d4 19.Rb1 Qb5 20.Rb5 e5!-/+ Wach - Kindermann, Ptuj 1995) 12...Bd7 13.a6 (13.Bd3) 13...O-O-O!?

14.ab7 (14.Bd3 f5!? 15.ef6 Qf4 16.fe7 Rde8 17.Bg6 Ne5!? 18.Be8 Rg2!‚ (Dolmatov - S. Ivanov, Kazan, 1995) 14...Kb8 15.Qd3 Rg4 16.Be3 (better 16.g3 Ng6 17.h3 Nf4 18.Qc3 with unclear play) 16...Nf5 17.Bc5 Re4 18.Kd1 Rc4 19.Qf5 Rc5 20.Qf7 d4 21.Bd3 Ne5 22.Ne5 Re5 with compensation (Ricardi - Forster, Elista, 1998)

11...Qa5 12.Nf3

12.Rb1 Black can meet with 12...Nd4!? 13.Qd3 Nef5 14.Nf3 Nf3 15.Qf3 Bd7 16.Rb7 Bc6 17.Rb4 Qc5 18.Qf2 d4, getting a strong counterplay (Sax - Vaganian, Wijk aan Zee, 1989)

12...Bd7

12...d4 with the idea of 13.Ng5 Rg5 14.fg5 Qe5 with compensation was not proved in the practice.

13.Rb1

There is a very keen play after 13.Ng5, as Black can chose any of the following lines: 13...Rg5 14.fg5 O-O-O15.Bd3 Qc5 16.Qh4 d4 17.O-O Ne5, 13...Rf8 14.Rb1 O-O-O 15.Nf7 Rf7 16.Qf7 Be8 17.Qe6 Bd7 18.Qf6 Bf5 or 13...O-O-O 14.Nf7 Nf5 15.Nd8 Qd8 16.Qh3 Ncd4 17.Qc3 Kb8 with unclear consequences in all three cases. Those who is interested to study nuances of this fascinating position can refer to a survey in the NiC, issue 33-1994.

13...O-O-O 14.Qd3 Nf5

Black used to continue with 14...d4, for example: 15.g3 Nf5 16.Bg2 Qc7 17.O-O (Shirov - Romero, Spain, 1998) 17...Nce7! 18.Ng5 Bc6 19.Nf7 Bg2 20.Kg2 Rd5 with compensation (A. Shirov). With the move in the game Black attempts to weaken the position of his opponent, yet he loses several tempos at that. The question is, what will prove to be more important.

15.Rg1

15.Rb5 Qa4 16.Qc3 Rg2 17.Bg2 Qb5 18.Bh3 d4 19.Qd3 Qc5 with counterlpay is weaker (Rudolf - Vaganian, Germany, 2000). In case of 15.g3 Black can turn the game to a theoretical position with 15...d4 or prefer 15...Nce7.

15...d4 16.g4

A natural reaction: the knight f5 is hanging over White's position, and he takes an opportunity to chase it away.

16...Nfe7 17.Rg3

A demonstratively firm move. 17.Ng5?! Rg5 18.fg5 Ne5 with initiative. is dangerous, but 17.Qe4, to be followed with Bd3 or 17.h4 with an advance of the h-pawn in the prospect, deserved attention, too.

17...Be8

An interesting plan: Black can play f7-f6 and then Bg6 in some lines.

18.h4

18.Kf2 can be met with 18...Nd5, planning f6 or Ne3.

18...Nf5 19.Rg1 Ne3

Black sacrifices the pawn, but the white king will feel uncomfortable after the files open up.

20.Be3 de3 21.Qe3 Ne7!

Contemplating Nd5 and Bc6.

22.Bc4

22...Bc6?!

Now Black loses the most important pawn c3. 22...Kb8! should have been played, to be followed with 23.Rb3 Rc8, Black having a sufficient compensation this way.

23.Rb3 Kb8 24.Qc3+/-

White has scaled down Black's initiative and achieved his material advantage. But it should be taken into account that Black's position was very promising after the opening, that is, the concerned variant appears to be quite viable.

5. J. Timman - A. Khalifman [C17]
Bali, 2000

1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Ba5 6.b4 cd4 7.Qg4 Ne7 8.Nb5

Another possible continuation that used to be popular in the eighties.

8...Bc7 9.Qg7 Rg8 10.Qh7 a6!

Weaker is 10...Be5 11.Nf3 Rh8 12.Qd3 Bf6 13.Bf4 Ng6 14.Nc7 Kf8 15.Bg3 e5 16.Na8 Nc6!? (16...e4 17.Qb3 ef3 18.O-O-O+/-) 17.b5!? (17.Qb5 e4 with counterlpay (Sion Castro - Arencibia, Leon, 1991)) 17...Na5 18.Qd2+/= (P. Keres)

11.Nc7

Black has a good play after 11.Nd4 Be5 12.Bb2 Qc7 13.Ngf3 Nbc6 14.O-O-O Bf6 with compensation (Deev - Luether, Longby, 1990).

11...Qc7 12.Bb2

12.Ne2 Qe5 13.Bb2 changes the order of moves, but an original development is possible, too: 13.Qd3 Nbc6 14.Bb2 Bd7!? (14...Nf5 15.O-O-O Qf6 16.Nd4 Nfd4 17.Bd4 e5 18.Bb2 Be6 19.Qd2 O-O-O 20.f4!+/= (Liberzon - Vaganian, Haifa, 1989)) 15.O-O-O (15.Bd4 Nd4 16.Qd4 Qf5 with compensation) 15...Qf6! 16.Nd4 Qf2 17.Nf3 Nf5 18.Qd2 Qe3= (Chandler - Vaganian, Manila, 1990)

12...Qe5 13.Ne2 Qc7!?

A fresh idea.

13...Nbc6 14.O-O-O Qc7 15.Nd4 Nd4 16.Bd4 e5 17.Bc5 Bf5 gives no absolute equalisation to Black, yet the theory insists that he may not be afraid after 13...Qf6! (with the idea e5). Further opportunities: 14.f4 Nbc6 15.Qd3 Nf5 16.O-O-O Qh6 17.Re1 Bd7 (17...e5!?) 18.Kb1 Rc8 19.g3 Kf8! with mutual chances (Yudasin - Lputian, Kiev, 1986)

14.Ng3?!

Having encountered something unexpected, White chooses not the very best continuation.

The following variants demonstrate Black's opportunities: 14.Nd4 e5 15.Nf3 Bf5 16.Qh6 Nd7 17.Rc1 O-O-O 18.c4 d4 19.c5 Rg6 or 14.Bd4 e5 15.Bc5 Bf5 16.Qh4 Nbc6 17.Ng3 Be6 18.Bd3 O-O-O 19.O-O e4 with very keen play for both sides. The strongest for White is 14.f4 Nbc6 15.Qd3 Nf5 16.O-O-O (both 16.g3 e5! 17.fe5 Ne5 18.Qb3 d3! 19.cd3 d4 with initiative. (Johannesson - Karlsson, Gausdal, 1990) and 16.c3 dc3 17.Bc3 Bd7 18.g3 Qb6+/= (Lau - Haugli, Nestwed, 1988) are weaker) 16...Bd7 17.Re1. In the game Nunn - Kinsman (London, 1993) an approximately even position was achieved after 17...Rc8 18.Kb1 Nce7 (18...Ne3!?) 19.Nd4 Nd4 20.Bd4 Qf4 21.g3 Qf5.

14...e5

A mighty pawn fist in the centre ensures an efficient play for Black.

15.Be2

15.Nh5 can be parried with 15...Qd6.

15...Be6 16.f4

Now if 16.Nh5, then 16...Nd7 with the idea 17.Ng7 Rg7 18.Qg7 Qc2 is good, providing an excellent position for Black.

16...ef4 17.Nh5 Qe5 18.O-O?

This move loses the game.

What he should have done was to play 18.Rf1 Rg2 19.Nf4 Bg4 20.O-O-O, and White is holding.

18...Bf5!

White's calculation was based on 18...Qe2? 19.Nf6 Kf8 20.Qh6 Rg7 21.Nh7 Kg8 22.Nf6 with a perpetual check.

19.Qh6 Rg6 20.Qf4 Rg2 21.Kg2 Qe2 22.Kg3 Qh5-+

As a result, White has misses material and his king is under attack. It was all over in several moves.

23.Bd4 Nbc6 24.Rae1 Nd4 25.Qd4 Qh3 26.Kf2 Qh2 0-1

6. Zaw Win Lay - A. Khalifman [C07
Bali, 2000

1.e4 e6 2.d4 d5 3.Nd2 c5 4.ed5 Qd5

In my opinion this is the most reliable way to struggle against Tarrasch system.

5.Ngf3 cd4 6.Bc4 Qd6 7.O-O Nf6 8.Nb3 Nc6 9.Nbd4 Nd4 10.Nd4 a6 11.Re1 Qc7 12.Bb3 Bd6 13.Nf5

The most fundamental line. A detailed analysis of this variant can be found in my article that was published not long ago in the NiC (issue 52, 1999).

13...Bh2 14.Kh1 O-O 15.Ng7 Rd8 16.Qf3 Kg7 17.Bh6 Kg6 18.Rad1?

After this game a question mark should accompany this move. Mean while, the position on the diagram occurred twice in grandmaster practice! The main continuation is 18.c3 Nh5 19.Be3 (or 19.Bc1 etc.)

18...Rd1 19.Rd1 e5!

Now the disproof comes! Black abandons his dark-squared bishop, but the engaging of the light-squared one proves to be decisive. Previously Black did not manage to cope with this position: 19...Bd7 20.Rd4! Qe5 21.Be3 Bc6? (better 21...h5) 22.Rg4! Ng4 23.Qg4 Kf6 24.Qh4 Kg7 25.Bd4ќ (Korneev - Vakhidov, Ubeda, 1997); 19...Qe5 20.Be3 Qh5 21.Qh5 Kh5 22.Kh2 Ng4 23.Kg3 Ne3 24.fe3 b5 25.Kf4+/= with a better endgame (Ponomariov - Sadvakasov, Lausanne, 1999). In my article I regarded 19...Be5 20.Qe3 Ne4! to be the best variant, but in this case White can only arrange a perpetual check. And why should he make a draw when he can win!?

20.Kh2

White's choice is not wide: 20.c3 Bf4 (20...Bf5 21.Bc1 Bf4-+ is good, too) 21.Bf4 ef4 22.Bc2 Kg7-+; 20.g3 Ng4 (but no 20...Bg4 21.Qe3!) 21.Be3 (21.Bf8 Bf5 22.Bd6 Qc6-+) 21...Bf5 22.Bd5 (22.c3 e4 23.Qe2 Rd8-+) 22...Qc2 23.Bf7 Kf7 24.Qb7 Kg6 25.Rd6 Kh5 26.Qf7 Bg6-+

20...Ng4 21.Kg1 Kh6

(21...Nh6?! 22.Qg3 Bg4 23.f3 is not so clear). White, stunned with what had happened, stopped the clocks in this position (0-1). The surrender cannot be called untimely: Black has an extra piece, and White's temporary activity cannot compensate it. There are possible variants:

22.Bf7 e4 23.Qh3 Kg7 24.Bh5 (24.Qh4 Qe5 25.Rd8 Nf6-+) 24...Nf6 25.Qh4 Qe5-+

22.Qg3 Bf5 23.Rd4 Kg5!-+;

22.Bd5 Kg7 23.c4 (23.Qg3 Qc2-+) 23...a5 24.Qg3 Ra6 25.f3 Qc5 26.Kf1 Qe3-+;

22.Qh3 Kg7 23.Qh5 (23.Qg3 a5!? with the idea of Ra6-+) 23...h6 24.Rd6 (24.Rd3 e4) 24...Qe7-+;

22.Qf7 Qf7 23.Bf7 Bf5-+