Beruflich Dokumente
Kultur Dokumente
Svetozar Gligoric
Translated by
Biljana and Zoran !lie
London
member
of ChrysalifBooks pic
Author's Foreword
This book has sp rung by chance from my w ri tten tex t fo r a lectu re, las ti ng
many hou rs, which I ga ve to the mos t talen ted young playe rs of my
coun try. This lec tu re, which took place a t the Chess School of the Yugosla v
Chess Fede ra tion in Belg rade 1 999 , made a deep imp ression upon my chess
colleagues a t home and, wi thout any furthe r help from me, the in te rna tional
mas te r Sini sa Jok si c though t i t was wo rth making a li ttle book ou t of i t and
published this in Se rbo -C roa t in 2000 , while a second edi tion appea red in
German in 200 1 .
Now has come the time to exp ress my g ra ti tude also to in te rna tional
mas te r Mr. Zo ran Ili c, who o n his own ini tia ti ve comple ted the job by
making sugges tions fo r an enla rged edi tion in English, and to his wife, Mrs.
Bil jana Ili c who helped in the transla tion of my o riginal manusc rip t.
Svetozar Gligoric
July, 2002
Contents
List of lllustrative Games
Historical Introduction
The
11
22
24
35
54
55
58
62
67
91
f4
.id2 tt'lf6
.id2 fxe4 1 2 tt'lxe4
f5 1 1 f3
Part Three:
Variation
Variation
Variation
Variation
Variation
9 tt'le 1 tt'ld7 1 0 f3 f5 1 1 g4
9 tt'ld2
9 .id2
9 .ig5
9 b4-"Bayonet Attack"
Index of Variations
91
97
1 15
121
1 27
1 60
Illustrative Games
Page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
12
16
19
24
29
31
35
37
39
41
45
50
55
58
63
61
71
14
78
85
87
91
93
95
91
98
1 04
1 05
1 10
1 15
121
128
135
136
141
1 46
151
1 55
A Historical Introduction
In chess opening theory it is common practice to name variations after the
places where they have been played for the first time, instead of naming
them, more justly, after their creators. Thus, a principal opening variation in
the Semi-Slav Defence, played for the first time at the international
tournament in Meran in 1 924, is not called the Rubinstein Defence (after
the famous grandmaster who invented it), but the Meran Defence.
There have also been errors of a different nature when naming openings.
Alekhine did not invent the so-called Alekhine Defence, although he did
secure it wide recognition by being the first to play it in great competitions.
Najdorf did not find a new variation in the Sicilian Defence (it was found
by Czech masters) but, as a top-level player, he was the first to understand
the importance of this idea while playing at the international tournament in
Prague in 1 946. As a matter of fact, the greatest contribution to the Najdorf
variation was made by Robert Fischer. Likewise, a variation of the King' s
Indian Defence, played by Karpov a s White i n his match against Kasparov
in New York 1 990, in the official American bulletin of the match was given
my name, although I was not the one who invented this early excursion of
the queen's bishop (perhaps it was Reshevsky?); true, I played it
successfully-but only to avoid, after I d4 tiJf6 2 c4 g6 3 lLlc3 g7 4 e4 d6
5 tiJf3 0-0 6 ..te2 e5 7 ..te3 , having to play against my own variation,
because if 7 . . . lLlc6? ! White is ready for a favourable regrouping of the
pieces with 8 d5 lLle7 (now 9 . . . ltJd4 doesn 't work) 9 tiJd2, as in the game
Gligoric-Kraidman, Tel Aviv 1 966.
One of the things I really did invent, however, is the Mar del Plata
Variation-a whole plan of development for Black, which I employed for
the first time at the international tournament in Mar del Plata in 1 953
against Najdorf, and two rounds later also against Eliskases, winning both
games. Shortly afterwards, in Europe, Najdorf copied my method as Black
in a Candidates tournament game against Taimanov, scoring an efficient
and classic victory, after which the variation, as fast and furious as a forest
fire, occupied a significant place in worldwide tournament practice, living a
full life for nearly half a century and well into the present day. In bygone
times, however, communications being far from what they are today, games
from far-away Argentina remained unnoticed for a long time and so it took
many years for me to 'lose my patience ' and announce the name of the
variation's ' anonymous ' creator. In fact, perhaps too modestly, I was
hesitant about claiming my authorship for three decades . . .
Part One :
The Development of Ideas that led
to the Mar del Plata Variation
If we ignore the different strategic concept in the Siimisch variation (5
f3), the Fianchetto line, with its pronounced space control, used to be the
most popular method for White. Then, shortly after the Second World War,
Boleslavsky and Bronstein, and later Geller as well, found dangerous
tactical chances for Black by abandoning the central pawn stronghold on e5
in order to exert pressure along the e-file and the b6-g l and g7-a l
diagonals-and combining this with various associated tactical options.
Aware that maintaining the pawn tension in the centre is a strong weapon,
because in such an unresolved and tight situation Black cannot find suitable
squares for all his pieces, White players turned to the Classical variation
with its quick kingside development: 1 d4 liJf6 2 c4 g6 3 liJc3 .tg7 4 e4 d6
5 liJf3 0-0 6 i.e2 eS 7 0-0. Black ' s first reaction was the natural move
7 liJbd7 but, by manoeuvring his rooks to e l and b l , White retained both
the tension in the centre and his spatial advantage.
. . .
h
8
8
7
2
1
a
Then, around about 1 950, the Soviet grandmaster Aronin promoted the
new move 7 liJc6! in order to increase the pressure on the d4 square and
force White to commit himself to a definite pawn formation in the centre.
This idea was immediately adopted by the top Soviet players, who had
...
10
a lready contributed enor mous ly to the popu larity of the King 's Indian
Defence. White p layers tried 8 .i.e3 !? in order to hinder B lack 's desire to
b lock the centre (after which he has a c lear p lan for coun terp lay by prepar
ing the pawn break . . .t/-f5), but the Argentinian grand master Na jdorf soon
found the right answer in 8 . . . :e8! when , if 9 d5 , there now fo llows 9 . . .lDd4 !
1 0 lDxd4 exd4 1 1 .i.xd4 lDxe4 with an e qua l ga me in a si mp li fied position.
Nor did Larsen 's atte mpt to p lay for an endga me advantage by 9 dxe5 dxe5
10 'i1Fxd8 pro ve to be e ffective enough to p lay for a wi n.
In a short period after 1 950- 1 952 White p layers fma lly accepted that after
7 0-0 lDc6 8 dS lDe7 (here the knight is for the present poor ly p laced , but it
is on the way to the batt le zone where B lack is gathering as much force as
he can for the impending attack on the kingside) there is a direct con fronta
tion of two initiatives : White 's on the queenside , and B lack 's on the king
s ide. White 's first reaction was to continue 9lDe1 (to deny the b lack knight
on f6 the active h5 and g4 s quares and to prepare a stab le white pawn cha in
leading towards the centre) , after which B lack 's best response was 9...lDd7 !
(making way for the b lack f-pawn and o ffering better contro l than 9 . . .lDe8!?
of the c5 and e5 s quares -where White can atte mpt a breakthrough by
c4-c5 or f2-f4).
Rea li zing that a ll this was the pre lude to a dra matic race between the two
p layers , each p lanning to be first to penetrate their respective flanks , and
that e ach tem po added to the gu lf be tween victo ry and defeat , the lead ing
So viet grandmasters were in a hu rry to launch B lack 's attack as quickly as
possib le a long the 'g 'or 'h '- file by a manoeuvre of the king 's rook from f8
to f6-g6 or h6 , after the se quence 10 lDd3 (or 1 0 .i.e3) 10.. f5 11 f3 f4 12
b4 gS 13 cS l:[f6!? etc.
.
a t the
I had
than
1
a
2
a
Now I saw that I could try 1 6... g4! without further ado, because the pawn
at e4 was not sufficiently well protected ( 1 7 fxg4? lll xe4 ), while against the
'terrible' 1 7 lll x a7 I noticed the decisive blow 17 ... g3!. This convinced me
that my new idea was not only worth the risk but appeared also to be
logical . Even today, I wonder why no one had come up with the idea
before.
In our game, however, the experienced Najdorf played in slightly
different fashion and I discovered a further plan which enabled Black to
manoeuvre quickly and easily over the board. This broader-based set of
moves was a fresh and powerful weapon for Black in the newly born Mar
del Plata Variation, and here is the matrix game with annotations:
Game 1
2
a
At the time this move (in place of 7 . . . l'Llbd7) was brand new, having just
hcen discovered by Aronin and seen in the game Taimanov-Bronstein,
I JSSR Championship 1 952.
8 d5 !De7 9 !Del ti::Jd 7
I
regarded this move as more precise than 9 . . . l'Lle8, which also opens the
for the f-pawn, because now White will have to spend one more tempo
order to achieve the thematic queenside breakthrough c4-c5 .
way
in
2
a
10 ti::Jd3
lO f5 1 1 f3 f4 1 2 ..id2 ti::Jf6 ! ?
...
1 3 b4! ?
Optimistic about gaining space, Najdorf corrects my mistake in the move
order, and wastes a tempo instead of exploiting the opportunity presented to
continue with an immediate 1 3 c5 ! . Now everything proceeds normally as if
the correct sequence 1 2 . . . g5 1 3 b4 l'Llf6 had been played.
A consistent idea. What should stand here is not the rook but the suddenly
activated, previously passive knight.
1 6 llcl
Expecting the continuation 10 e3, my "preparations" went only as far as
the 1 5'h move. I realized that my critically sensitive point was c7, and, over
the board, without much thought, I found the following move.
1 6 ...l:lt7!
This excellent manoeuvre both defends and attacks.
1 7 cxd6 cxd6 1 8 a4
The game is a classic example of a situation where the opponents
consistently pursue their respective operations on opposite flanks. However,
the more important flank is the one where the kings are, because Black can
even allow himself to lose the battle on the queenside and still deliver
tactical strokes which would enable him to win the decisive battle on the
opposite wing.
1 8 ... iH8!
Black consistently carries out his plan. In good time, he protects his king
from being disturbed along the sh rank via the opposite flank, while
additionally covering the weak pawn on d6 and releasing the g7 square for
the rook.
19 aS l:.g7
At this moment, the Mar del Plata Variation was born !
15
2
1
a
20 h3 l0h8 !
Preparing the decisive breakthrough . . . g5-g4.
21 l0b5
White tries to weaken his opponent on the queenside, but Black can more
or less ignore it.
2 l . .. g4 22 fxg4 hxg4 23 hxg4 a6 24 l0a3 i.d7 25 l0c4 l:.c8 26
li'lb6 llxcl 27 .i.xc l .i.e8
a
2
a
White 's only chance is to get to c8 with his rook, but this is a long way
and the black king is well-protected from everything that is happening
on the opposite side.
off
2
b
. .
Eliskases - Gligoric
Mar del Plata 1 953
\Tars
ttl'W
IItts
of the 20'11 century, thanks to the efforts of Korchnoi and others, were
options found for White. which will be dealt with in a separate part of
book.
IO f5 1 1 f3 f4
...
2
a
13 tbd3 tbf6 !
In the earlier game Taimanov-Bronstein, USSR 1 952, Black was not suc
ll'ssli.d with 1 3 . . . .l:r.f6, on which followed 14 c5 .l:r.h6 1 5 cxd6 cxd6 16 ltJb5
t_;\f'X 1 7 .te l a6 1 8 ltJa3 b5 1 9 ltJc2 ltJd7 20 a4 ! etc.
Black
llt 'ii b3
1-'111
...
Now
Iu; best
18
If 27 -.xa7 c6.
2
a
33 ....:dxd7 ! !
This move destroys all White ' s plans.
34 xd7 'it'g6
Although at the moment he is a piece up, White cannot defend himself
against the mating threats along the g-file without material losses.
lecture for young p layers -and this was in Be lgrade , not until the year
1999!
When , seve ral months after Mar de l P lata , Na jdorf, as Black agai nst
Taimanov , adopted the s ystem I had p layed against him , he pr oduced his
uwn model example of how to p lay the line in a game which has since
hccome a classic in chess history.
Game3
Taimanov- Najdorf
l/)f6 14 c5
Bronstein , in his famous tou rnament book , wrote : "In the 20th USSR
championship 1 952 , Taimanov , as White , won two games with this same
method , i.e. a pawn break a long the c- file , progress on the quee nside and
deep penetration of pieces into enemy ter rito ry along the d- , e- , f- and
g-files -after which the press wer e left with the impression that Black had
lost both games because of the o pening. However , there wer e pla yers who
practised the "re futed variation " with success. For instance , in Mar del Pla ta
the victim of fashion was Na jdorf who lost to Gligori c and d rew with
Trifunovi c. But , at the sta rt of the Zurich tou rnament , these games were not
known to Taimanov . Thus both p layers began this game with great ho pes.
Na jdorf, self-taught by practical ex perience , had studied Yugoslav anal ysis
which went up to at least the 2 1'1 move , whi le Taimanov was encouraged b y
his previous two victories. "
The r eader will know by now that I have to co rrect Bronstein 's descri p
tion a li ttle b y stating that the idea of Black 's system was entirel y mine and
that there was no pre limina ry Yu gos lav ana lysis whatsoever.
The first time that this promising pawn sacrifice, introducing attacking
tactics, was applied here.
1'1<'\TS
. \-t
Wh ite loses the queen, because if it tries to run away, Black's knight
i'IVl'S a winning check on g3 .
.\7 'Wxg7+ xg7 38 l:tg2+ h8 39 tLlel tiJf4 40 .:tg3 i.. f2 4 1 .l::tg4
Wtd 42 tiJd2 h5
line,
the game was adjourned (the good old days ! ) , and White sealed the
After 43 . . . l:.g8 44
Nevertheless, the most frequent and first played continuation was 9 ll:\e1
which is worthy of particular attention.
.'. c17,
111 I
In the
l.avonct
last few years, special attention has been given to the so-called
attack ' , in its improved version, with 9 b4,
a
1
a
ll'lllll" White
"mg th a t
,, .. , , ... hut
..
lyrical performance
"This draw has the charm of perfection. Each move is interesting and, to
this day, appears flawless. With 17 . . . c5 Fischer launches an intricate
double-Pawn sacrifice which involves exact timing. Gligoric rises to the
occasion, returning material in an attempt to wrest the advantage. The
economy and ingenuity displayed by both players produces a harmonius
flow of movement, remarkable in its esthetic appeal. The effect is of a pas
de deux in which each partner contributes equally to the total symmetry."
Game 4
Gligoric - Fischer
Bled 1 96 1
IO ... f5
I I cxf5
l'l't rosian-Tal, in this same tournament, continued (with Black 's tiJ on
q I I t"t exf4 1 2 i.xf4 fxe4 1 3 tiJxe4 tiJf5 14 i.g5 tiJf6 1 5 g4 tiJd4 1 6
: . \dl'2 "fie7 = (I chose this simple pawn exchange in our game for 'safety
I o'ol.\'0/1.1'1, having been impressed by the effectiveneSS of the pawn avalanche
:' .t:5-g4 in my own previous wins as Black-here I wanted to avoid such
ol /oil<' ()/1 the other side of the board!).
ll ...liJxf5
lu this
'"IJ'.dic.
14 ... lbh5
White has the c4-c5 lever; Black has the dynamic break with . . . g5-g4.
Chances are roughly even.
1 5 J.g5 'it'd7
Keeping an eye on the d6-pawn so that . . . c5 becomes possible.
16 g3
(I could not let the knight come to f4, although I was not too happy at
having to weaken the light squares around my king).
1 6 ... h6
In a later round Gligoric (as Black) played against Tal 1 6 . . . c5?. But after
1 7 0.b5 ! 0.xb5 1 8 cxb5 White obtained a bind.
17 .te3
a
2
a
1 7 ... c5 !
I was informed that Gligoric thought I had blundered a Pawn (correct-!
admit!), but it is a deliberate sac. On 1 7 . 0.xe2+ 1 8 'i!kxe2 g5 1 9 c5 White
has it all his own way.
..
18 .txd4
Not 1 8 0.b5 0.f5 1 9 ..td2 a6, etc.
1 8 ... exd4 1 9 lbb5 a6
Not 1 9 . . . i.e5? 20 f4 .
20 tDbxd6
Apparently Black has lost a Pawn without any visible compensation. His
Jlll:o:es, which are now so awkwardly placed, soon spring to life, however.
20 d3 ! 21 'ifxd3
A double-edged game would result from 2 1 i.xd3 i.d4+ 22 h l tl:Jxg3+
.'I tl:Jxg3 'i!Vxd6 24 'i!Vc2 i.h3 .
..
2 l ... i.d4+
The combination requires intricate footwork. A mistake would be
' l i.xb2 22 tl:Jxc8 i.xal 23 ti:Jb6 and it' s all over (23 . . . i.d4+ 24 'ii'xd4).
. . .
22 g2
After 22 'it>h l tl:Jxg3+ 23 tl:Jxg3 'i!Vxd6 White is weak on all the squares
and his K-side looks like Swiss cheese. Chances would be even.
22 lDxg3 !
.
2
1
a
28
23 lbxc8 !
Best. Not 24 hxg3 (or 'it>xg3) 'ii' h 3 mate. On 24 !Dxg3 'ii'xd6 again is
good.
25 l:.xfl 'ii' x b6 26 b4 !
The saving clause.
26 .. .'iVxb4
I saw the draw coming but felt the position was too precarious to play for
a win. On 26 . . . cxb4 27 c5 ! xc5 28 !Dxc5 'ii'x c5 29 'it'xg6+ 'it>h8 3 0
'ii'x h6+ 'it>g8 3 1 'it> h 1 wins. The only other try is 2 6 . JH7 2 7 bxc5 xc5 28
l:.b 1 followed by d5-d6 with tons of play.
.
3
2
Game 5
: W(,
7
6
3
2
1
a
12lbe4 ! ?
White i s in too much o f a hurry to occupy his stronghold. A more natural
\\,IY would be to bring another knight to this square after 1 2 f3 and 1 3 0,2 .
12 . .. lbf6 13 f3 lbd4 1 4 i.e3 c6!
Since there is no white knight on c3, Black emphasizes the sensitivity of
1lw d5 square.
15 l2Jxf6+ 'ifxf6 1 6 lDf2 cxd5 1 7 cxd5 Vllif7 1 8 i.c4 b5
lllack has obtained satisfactory counterplay.
19 .txd4 exd4 20 .txb5 'ii'x d5 21 'ii' a4 'ii'h 5 22 'ifb3+ h8 23
.:._,.(, i.e5
"'
I , ,,
'"' ,, I)
lik I li'Jd7
I()
li:Jd3
24 f4?
White is too greedy to win material and rushes into a hidden trap. He
should have continued 24 g3 .1La6, with chances for both sides.
27 'ifd5
White had calculated that this manoeuvre would destroy Black' s attack.
Not 27 .1Lg2 .1Lxg2 28 <li>xg2 'iff3+ 29 Wh3 .l:tf5 .
27 ...'ii'e 2 ! 28 'it'xd6
h
8
1
a
28 ...g7 ! !
The German grandmaster could not get over the shock when, after this
quiet move, he realized that he had no adequate response and that, even a
rook up, he was lost. 29 i.g2 .l:txf2 etc. is also no use.
29 'ii'x f8+ xf8 30 i.g2 i.xg2 31 xg2 'i'xb2 32 a4 'i'c2 33 h3
g7 34 l2Jg4 'i'c8 35 h4 g5+ White resigned.
After 9 li:Je 1 li:Jd7 1 0 li:Jd3 f5 White players later found a new finesse to
save an important tempo for the breakthrough c4-c5. This was by the inter
mediate development 1 1 i.d2 (instead of 1 1 f3) when, after l l . . .li:Jf6 1 2 f3
f4 1 3 c5, White will open the c-file a move earlier without spending a
tempo on the superfluous b2-b4 (as in Najdorf-Gligoric).
Game 6
Larsen -Tal
Game Five, Candidates match,
Eerse/ 1 969
2
a
1 5 tl'lb5
1 5 cxd6 cxd6 1 6 llJb5 is simplest because it avoids the unclear complica
lions of 1 5 . . . a6 1 6 cxd6 ! ? .
1 5 :n
...
8
15 a6 (see diagram) 1 6 cxd6 ! ? is sharp 8
(passive is 1 6 llJa3 g4 1 7 cxd6 cxd6 1 8 llJc4 7
7
g3 1 9 h3 .i.xh3 20 gxh3 b5 2 1 l:te 1 bxc4 22
6
6
.!:hc4 llJh4 23 i.. fl h5 24 i.. g2 llJh7 25 l:tfl 5
li1g5 with the attack, Hoeksma-Riemersma, 4
4
llolland 1 987) 1 6 ... axb5 1 7 dxc7 'ii'd 7? !
3
3
( II uzman recommends as better 1 7 . . 'iVe8 ! ? 1 8
2
2
'iib 3 g4 1 9 i.. b4 l:tt7 20 l:tfd l gxf3 2 1 i.. x f3
li.Jh4 with counter-chances) 1 8 'ifb3 llJ e8 (if 1
a, b
c
d c f 8 h
1 8 . . . g4 1 9 i..b4 l:tf7 20 d6 ! g3 ? ! 2 1 l:tfd l
gxh2+ 22 <ifi>xh2 b6 23 i.. a 3 , the outcome favours White, Sariyadzanov-Klimov, St. Petersburg 1 997) 1 9 llJc5! 'ii'd 6 20
li.Je6! 'ii'b 6+ 21 <ifi>h1 .i.xe6 22 dxe6 llJxc7 23 e7+ l:tti 2 4 i..c4 llJh8 2 5
J:Ifd 1 lieS and in the game B iryukov-Solovyov, St.Petersburg 1 999, White
accepted a draw without noticing the road to victory after 26 i.. x f7+ llJxt7
27 'ii'c 2llJa6 28 'ii' c 8llJd6 29 'ife6+ <ifi>h8 30 .te l etc.
..
1 6 cxd6 cxd6
3
2
a
1 7 'Wc2!
The threat ti::Jb 5-c7-e6 would annihilate Black's chances of an attack, so
now he has to lose two tempi to prevent it. Less precise is I 7 ti::J 2 i.f8 I 8
'ifc2 1i.d7 ! ? I 9 'ifc7 ti::J e 8 2 0 'ii'x d8 l:txd8 2 I ti::Jx a7 l:ta8 22 ti::Jb 5 l:txa2 23
1i.c3 l:ta8 24 l:ta I l:tb8 25 i.b4 i.e7 and B lack managed to hold the draw in
a passive endgame, Korchnoi-Gligoric, Leningrad I 973 .
1 7 ...g4
Later on, Black players turned to the more cautious continuation 17 ti::J e8
(see diagram) 18 a4 h5 1 9 ti::J fl 1i.f8 20 h3 (if 20 ti::Jx a7 l:tc7 2 I 1i.a5 l:txc2
22 1i.xd8 l:txe2 23 ti::Jx c8 l:txa4 24 ti::Jd 3 g4 25
.l:t2
l:te3 26 ti::Jei l:ta8 27 l:tfc2 l:tb3 28 'it2
8
8 l:ta2 29 l:tb I 'itf7 30 'ite2 1i.e7 3 I ti::J x e7 li::J x e7
7
7
32 ti::Jd 3 l:ta8 33 1i.xe7 rl;xe7 with equal play,
6
6
Novikov-Glek, Lvov I 98 5 ; or 20 'ifb3 l:tg7 2 I
5
5
h 3 ti::Jh4 22 l:tc2 a6 2 3 ti::Ja 3 1i.d7 24 l:tfc 1 l:tb8
4
4 25 li::J c 4 b6 26 a5 g4 ! 27 fxg4 ti::J f6 28 ti::J x b6
3
3 hxg4 29 hxg4 ti::J x g2 ! and instead of 30 'ii'h 3 ? !
2
2 ti::J e 3 3 1 i.xe3 l:th7 ! which favours Black,
lvanchuk-Timman, match, Hilversum 1 99 1 ;
-....a 1
Timman recommends as better 3 0 'itxg2 li::J xg4
3 1 i.xg4 i.xg4 32 'it>fl i.d7 ! with chances for
both
sides) 20 ... l:t g7 (see diagram) 2 1 a5 (if 2 I
8
8
ti::J x a7 l:tc7 [less clear is 2 1 . . .1i.d7 22 ti::Jb 5 ti::J h4
7
7
23 'ifb3 'ith8 24 a5 g4] 22 1i.a5 l:txc2 23 .txd8
6
6
l:txe2 24 ti::J x c8 l:txa4 25 ti::Jd 3 g4 26 .l:t2 l:te3
5
27 ti::J e 1 g3 28 l:tfc2 l:tb3 29 'itfl l:ta2 30 l:tb 1
4
4
'itf7 3 1 <iti>e2 l:ta8 32 ti::Jd 3 1i.e7 33 1i.xe7 ti::Jx e7
3
3 34 ti::J x e7 'itxe7 with equal play, Averkin2 Kasparov, USSR 1 979) 2I. .. Ji.d7 (or 2 I . . .ti::J h 8
22 Vb3 ti::J f7 23 l:tc2 ti::J h6 24 l:tfci a6 I,h- Ih,
1
iS!b...a
Drasko-Gligoric, Sarajevo 1 983) 22 'ii'b 3 ti::J h 4
..
!\ .!tel i.. e 7 with chances for both sides, Polugaevsky-Tal, match, Alma
1 980.
:\Ia
19 .th3 20 tDxa8
.
Or
20 li:Je6 'i'b6+ 2 1 lt2 J.xe6 22 dxe6 lte7 23 ,.a4 li:Jf8 24 i.. a 5 'We3
i.. fl li:Jxe6 26 J.d2 'it'b6 27 i.. a 5 ,.e3 28 i.. d 2 Ih-Ih, Podgaets
M.Gurevich, Sverdlovsk 1 984.
.
' 'l
2
a
22 .. .'iVg2+? !
The correct continuation i s 2 2 . . ...h4+ with a draw, Averkin-Tal, USSR
1 969.
However, after
13 i.f3
If 1 3 f3 t;:)f5 1 4 ti:)d2 ti:)xe4 1 5 lt:Jxe4 c5 1 6 .td3 ti:)d4 1 7 l:tb l a5 IJS-th,
Glig_oric-Szabo, Dallas 1 957, while also playable is 14 . . . . c6 1 5 ti:)xf6+ .txf6
1 6 tDe4 cxd5 1 7 cxd5 i.. g 7 1 8 g4 ti:)d4 1 9 'lti>g2 'ii'b 6 20 b4 i.. d 7 2 1 a4
ti:)xe2 22 'ii'x e2 l:tac8 with chances for both sides, Robatsch-Giigoric, Lone
Pine 1 97 5 .
O r 1 3 ti:)x f6+ .txf6 1 4 .tc3 t;:) f5 1 5 i.. f3 i.. g7 1 6 i.. e4 ! 'ii'h4 1 7 l:t e 1 b 6
1 8 g 3 'ii' g 5 1 9 Vel 'ii'x c l 2 0 l:taxc l a5 IJS-IJS, Polugaevsky-Tal, match,
Alma Ata 1 980.
Game 7
Piket - Kasparov
Tilburg 1 989
I d4 lDf6 2 lbf3 g6 3 c4 i.. g 7 4 lbc3 0-0 5 e4 d6 6 i.. e 2 eS 7 0-0
ll'\c6 8 dS lbe7 9 lDe1 lDd7 10 i.e3 fS 1 1 f3 f4 12 i.f2 gS
a
1
a
20 ... g3 !
8
2
a
2 1 li)xa8?
If 2 1 hxg3 fxg3 ! 22 i.xg3 i.h6 ! 23 CDxa8 CDh5 24 i.2 CDgf4 25 CDd3 !
l:tg7 26 CDxf4 i.xf4 27 g4 ! i.xc l 28 'ifxc l tDf4 29 'i1Ve3 h5 ! with the
initiative.
2 1 . .. li)h5 ! 22 <i1th l
The only move. If 22 i.xa7 'i1Vh4 23 h3 i.xh3 24 gxh3 'ifxh3 25 l:t2
gx2+ 26 'ifo>x2 CDh4 27 i.fl 'ii' h 2+ 28 CDg2 l:tg7 and Black wins.
h
8
8
7
6
Game 8
This is more ambitious than 1 5 cxd6 cxd6 as was played in the above
g i ven game, Piket-Kasparov. It is still possible for White to exert maximum
pressure on the queenside with c5-c6.
15 ... h5 16 c6 ! ?
2
a
White blocks the queenside where he is active, but the blockade is only
partial because on 16 . . . b6 he has the strong response 17 a5 .
20 ... hxg4 2 1 g3 ! :n
After 2 l . . . f3 22 i.d3 , with the kingside blocked, it would be easier for
White to defend himself.
22 i.d3 h7 23 h1 gS 24 l:.a2
White makes useful preventive moves for defending the king.
3
2
2
1
4 1 ...'ii'g6
Or 4 l . . .'ii'e 3 42 'ii'x e3 .i.xe3 43 .l:.f6 .l:.f7 44 .l:.xf7 'it?xf7 45 .i.e2 i.d4 46
i.fl .i.g4 47 i.xh3 i.d l 48 c7 winning.
42 :n
Covering the gl square, and parrying the threat 42 . . . .i.f4 .
h2+
Korchnoi - Kasparov
A ms terdam 1 991
13 lDg6
..
14 lbd3
28 g5? !
Atalik
Gufeld
3
2
1 1 ...liJf6
In Havana in 1 997, the same opponents (who also played in Beij ing in
1 996, when Black won) played this line and after 1 l . . .f4 12 i.. 2 g5 13 Ac l
2
b
1 8 ..1xa7
Up to this moment, White played all his moves very quickly. If 1 8 ltJxa7,
a
b
c
(see diagram) then Black, in an unclear
d e r g 11
8
position,
would obtain an initiative for the
8
pawn.
7
7
6
6
5
2
a
20 tbd7 !
...
lo
A bolt (a prepared one though ! ) from the blue. Black gains a vital tempo
carry out a devastating attack.
21 h3
The only sensible answer. If 2 1 lt:Jxa8 lt:Jxb6 22 lt:Jxb6 'ifh4 23 h3 i.xh3
24 gxh3 'ii' x h3 25 .l:tf2 gxf2+ 26 'it>xf2 White exposes himself to an unpleas
a nt attack, where Black has the choice between repeating moves and search
ing for a mating attack.
2l tbxb6 !
..
23 tbh8
..
24 i.c4
If 24 tiJe6, then Black proceeds with 24 . . . i.. x e6 25 dxe6 'i'xe6 with equal
chances, while 24 . . . tiJf7 ! ? is unclear.
35 e2
5
4
1
a
Piket - Kasparov
Linares 1 99 7
f
8
8
7
6
3
2
1
1
a
13 l:tcl
This simple developing move (instead of 1 3 a4) is the introduction to a
modem plan with a pawn sacrifice and a gain of time to exert pressure
along the c-file, which ties the black forces to the defence of the c7 square,
instead of allowing them to attack. In case of 13 a4 l:i:Jg6 14 a5 'it>h8 1 5 l:i:Jb5
l:i:Jf6 16 l:i:Jxa7 i.d7 17 c5 g4 ! 1 8 c6 g3 1 9 hxg3 bxc6 20 l:i:Jxc6 i.xc6 2 1
dxc6 l:i:Jh5 ! 2 2 gxf4 exf4 2 3 i.c4 l:i:Jg3 Black has chances for a strong
attack, Ziegler-Shulman, Goteborg 1 999 (24 i.e6 'iif6 ! 25 i.xg3 fxg3 26
i.h3 l:i:Jf4 27 'ifd2 'iih4 28 l:i:Jc2 .:tfb 8 etc.).
a
13 li.Jg6
..
Variation 9 lD e i lDd7 1 0 i. e3 4 7
Ilf2 i.d7 2 1 Ilfl i.f6 2 2 i.d2 'ii'h 2+ 23 'ifr> f2 'it'g3+ 2 4 <iitg 1 l:.h2 25 Ilf2
'fi'h4 0- 1 Ramos-Bologan, Las Palmas 1 993, are weaker) 1 6 'ii'e8 17 'ith1
'ii'h 5 1 8 i.g1 'ii' h4 1 9 lDd3 ! b5 20 cxb6 cxb6 21 'ife l 'ir'h5 22 lDf2! with
advantage for White, Korchnoi-Nataf, Cannes 1 998
Also worth mentioning is White ' s thematic idea of sacrificing a pawn (as
in the main game) by playing 14 c5, trying to obtain pressure on the c-file
as quickly as possible. In the game Vera-Nataf, Havana 2002, White suc
ceeded in that after 14 ... lDxc5 15 b4 lDa6 16 lDd3 Ilh6 17 a4 'it'e8 18 'ith1
'iVh5 (More consistent than 1 8 ... c6 19 b5 lDc7 20 dxc6 bxc6 21 lDb4 c5 22
li.Jbd5 lDexd5 23 lDxdS lDe6 24 b6 with an initiative for the pawn sacrifice,
Lopez Martinez-Volokitin, Leon 200 1 ) 19 i.gl i.d7 20 lDa2 Ilc8 21 lDf2
'tieS 22 lDg4 l:lg6 23 b5 lDc5 24 i.xc5 dxc5 25 l:lxc5 b6 26 Ilc3 h5 27 lf::, f2
'ot?h8 28 lDb4 ltlg8 29 lDa6 'it'd8 30 h3 i.f8 31 'ii'c 2 i.d6 32 Ilcl with
strong pressure on the c-file, Vera-Nataf, Havana 2002 .
.
14 c5 !
Kozul was the first to come up with this idea, which was later analyzed
and tested in practice by Piket and Korchnoi. This pawn sacrifice enables
White to speed up his action on the queenside.
..
6
5
6
5
4
3
1
3
2
2
1
10
i.. e3
20 i.e1 !
20 'ii'd2 g4 2 1 llfc 1 h7 22 lt:Jxc7 lt:Jxc7 23 llxc7 g3 is double-edged.
Not 24 . . . i.xa4? 25 b5 .
25 fxg4
a
2 S Ilg7?
29 hxg4?!
32 g3 ? ?
White could have won after 32 l:lfxf3 'ii'xg2+ ( i f 32 . . . l:lxf3 3 3 l:lxf3 'ii'g4
34 i.g3) 3 3 'iVxg2 l:lxg2+ 34 'it>h l l:lxf3 3 5 l:lxf3 .
32 ...l1f4! 33 tbe3 ?
Again correct was 33 l:lfxf3 l:lxh4 34 CiJ2 h2+ 35 'it>h l l:thS 36 l:tc8+
Wh7 37 l:lf8.
33 .lbe4 !
.
45 i.e7+! g8
If 45 . . . xe7 46 1i'e6+ f8 47 'i!fc8+ etc.
Piket - lvanchuk
Wijk aan Zee 1 999
1 d4 f6 2 c4 g6 3 c3 i.g7 4 e4 d6 5 f3 0-0 6 i.e2 e5 7 0-0
c6 8 d5 e7 9 el d7 10 i.e3 f5 1 1 f3 f4 12 i.f2 g5 13 l:c l
g6 14 c5 xeS 1 5 b4 a6 1 6 d3
a
3
2
1 6 . l:t7
.
1 7 b5
After 17 a4 White was wary of the possible transfer of the passive knight
from a6 via b8 and d7 in order to strengthen the delayed attack on the king
s ide, so, by threatening to capture directly on a7, he forces Black to occupy
t he d7 square with his bishop.
1 7 ... .i.d7 1 8 a4
A slower alternative was 1 8 l?:Ja3 , intending l?:Ja3-c4-a5 .
2
a
1 8 ...'ife8 !
The most efficient way for Black to discourage White from taking on c7.
The alternatives are
1 8 . . . i.xb5 1 9 axb5 l?:Jb8 20 'ic2 h5 2 1 'ifilh1 i.h6 22 l:.g 1 l:.g7 23 l:.a 1
l?:Jd7 24 l:.a3 l?:Jf6 25 l:.ga 1 g4 26 l:.xa7 l:.xa7 27 l:.xa7 g3 28 .i.e 1 b6 29
'ii'c 6 1h- 1h, Krivoshey-Dobrowolski, Koszalin 1 999, and
1 8 . . . h5 1 9 l?:Jxa7 i.f8 (or 1 9 . . . i.h6 20 l:.c3 l:.g7 2 1 l?:Jb5 l?:Jf8 22 h3 l?:Jh7
23 i.e 1 l?:Jf6 24 l?:Jt2 'iti>h8 Te.Johansen-Djurhuus, Norway 1 998) 20 l?:Jb5
l:.g7 2 1 l:.c4 l?:Jh8 22 g4 ! hxg4 (22 . . . fxg3 23 hxg3 h4 24 g4 is in White' s
favour) 23 fxg4 l:.h7 2 4 'iti>g2 'if6 25 h 3 c 6 2 6 l?:Jc3 l?:Jc7 2 7 i.f3 and White
stands better, Piket-Nijboer, Rotterdam 1 999.
1 9 b2
Here we have perhaps a critical position. After 19 l:.c3 h5 ! 20 'iib 3 g4
Black' s attack on the kingside might be faster than White ' s pressure on the
queens ide.
1 9 b8 ! ?
..
20 'ilc2 ! ?
White threatens to take o n c7 . I f 2 0 li:,c4 li:,xb4 or 2 0 'i!Vd2 h5 2 1 .l:tc4 g4
Black's attack would advance quickly .
a
2
1
a
20 ... i.e8?
A more promising concession would be the one with 20 . . . i.xb5 2 1 axb5
li:,xb4 22 'ifb3 (or 22 'i!Va4 a5 23 .l:tc4 b6 24 .l:ta l .l:tf8 25 i.e l with unclear
chances) 22 . . . a5 23 b6 i.f6 24 li:,a4 i.d8 25 .l:tc3 li:,e7 26 .l:tfc l li:,c8 27
i.b5 ! with the initiative in return for the material investment, or the more
enterprising 20 . . . c6 2 1 dxc6 bxc6 22 li:,xd6 etc .
24 ctJd3 !
Not so dangerous here for Black is 24 1i'c3 c6 25 dxc6 bxc6 26 l:txc6
.Jil.xc6 27 'iVxc6 liJxb4 28 'ifxd6.
1
a
28 ... gxh2+?
With this move Black deprives himself of any counterplay. Better
resistance would have been offered by playing 28 . . . i.xa4 29 l:txf7+ xf7
30 'ii'c 7+ liJe7 3 1 1i'xb7 with a slight advantage for White.
9 ti:le8 1 0 f4
...
2
1
a
l O exf4
..
1 1 xf4 h6
a
1 2 ti:Jc2
8
8
7
6
5
....
r
3
2
1
12 . . . f5 1 3 exf5 g5
a
3
2
a
14 i.d2
Or I4 i.e3 tiJxf5 I 5 i.f2 ( I 5 'i'd3 ! ? 'i'e7 I6 i.d2 .ite5 1 7 l1f2 tlJeg7 1 8
llafl Gavrilov-Navrotescu, Eoforie I 996) I 5 . . . i.e5 1 6 ..td3 'i'e7 I 7 'ife2
"iig 7 I 8 h i tiJf6 I 9 i.g i tiJh4 20 ..id4 ..ig4 2 I 'ifd2 tiJd7 22 i.e2 1h-1h,
Sosonko-Van der Wiel, Dutch championship I 994.
1 0 . . . f5 1 1 f4
a
3
2
The idea of this system is to prevent the activity by Black on the kingside
and, by opening the position, to exploit the weakness of the e6 square.
l l . exf4
. .
. . .
1 2 lbxf4
The possible continuation 1 2 ii.xf4 (see
diagram) is lel)S frequently played:
6
s
7
5
3
2
1
a
12
. . .
fxe4
6
4
3
2
a
8
7
5
4
3
2
.'1
1:1
"
e .:..
f
g 1...1..a
,
8
7
6
5
1:1
...a
8
7
2
"
-....a
3
2
1
1 5 i.d3 i.e5 16 a4
PreP.aring the activation of White ' s queen ' s rook via a3 . If 1 6 g3 l1Jeg7 1 7
l1Jg2 'i.fe8 1 8 l:lb 1 i.d7 1 9 b3 c5 20 l:le 1 'ii'c 8 2 1 i.f4 i.xf4 22 l1Jxf4 'i!lc7
23 b4 b6 24 bxc5 dxc5 25 l1Jg5 l1Jd4 lh- lh, Yennolinsky-Rogozenko, Bad
Zwesten 1 997.
19 tbxg6 !
LaUe - McShane
Redbus Knockout, Southend 2000
1 d4 tiJf6 2 c4 g6 3 ltJcJ i.g7 4 e4 d6 5 tiJf3 0-0 6 i.e2 e5 7 0-0
ltJc6 8 d5 ltJe7 9 ltJe1
Another order of moves has also been tried: 9 .td2 to provoke 9 . . . tbe8
10 tbe 1 f5 1 1 tbd3 and reach the same position.
. . .
ll
. . .
tiJf6
Now, the same position has been reached as in the game Larsen-Tal,
Eersel 1 969 (compare the earlier comments on it in this book, Game 6).
Playable is the waiting method (imitating White's opening tactics) with
.. h8, (see diagram) but White can also
b
1
d e r 8 '
"
c
insist on improving his pieces with 12 Act
8
(less consistent is 1 2 f3 f4 1 3 c5 h5 1 4 Ac l g5 8
7
1 5 cxd6 cxd6 1 6 l:i:lt2 Af6 1 7 h3 l:tg6 1 8 a4 7
6
ll:lg8 1 9 l'l:lb5 l'l:lh6 20 Ac3 a6 2 1 ll:la3 ll:lf6 22 6
5
5
lC!c4 b5 23 l:i:laS g4 24 ll:lc6 1i'e8 25 fxg4 hxg4
4
26 hxg4 l'l:lhxg4 27 ll:lxg4 i.xg4 28 i..xg4 4
3
Axg4 29 Ah3+ g8 30 i. e l i.f8 3 1 i.h4 3
2
iVhS 32 iVe l 1i'h7 33 l:tff3 iVxe4 34 iVxe4 2
lC!xe4 and Black won in 58 moves, Piket- 1
J.Polgar, Monte Carlo 1 994) 12 . . . l:i:lg8 13 f4
a
b
c
d e r 8 h
fxe4 14 lC!xe4 i.. f5 1 5 i.. f3 lLlef6 16 Ae1 l:i:ld7
17 fxe5 lC!xe5 18 ll:lxe5 i.xe5 19 i.. c3 Ae8 20 1i'd2 i.. x e4 21 Axe4 i.xc3
22 1i'xc3+ 1i'f6 23 Af4 iVxc3 24 Axc3 g7 25 c5 Ae7 26 cxd6 cxd6 27
i.g4 Af8 28 Axf8 xf8 29 i.. e6 with the better endgame, Shirov-J.Polgar,
Monte Carlo (blindfold) 1 995.
11.
1 2 f3
a
2
1
1 2 f4
...
6
5
4
3
2
1
11
1 3 c5
The reader has seen 13 :ct g5 14 c5 I1Jg6 1 5 cxd6 cxd6 16 11Jb5 :n 1 7
11J t2 i.. f8 1 8 'ifc2 i.. d7 1 9 'ifc7 (Korchnoi-Gligoric, Leningrad 1 973) in
comments on the line 9 i.. d2
A cautious plan is 13 g4 g5 1 4 .te l h5 1 5 h3 11Jg6 1 6 'it>g2 :n 1 7 :h i
i.. f8 1 8 .tf2 i.. d 7 1 9 c5 as in Hoeksma-Van Wely, Dutch championship
1 994, and White won after 1 9 . . . 11Jh4+ ! ? 20 .txh4 gxh4 2 1 'ifb3 etc. , in 5 8
moves.
13
. . .
g5 14 cxd6
14
. .
cxd6
a
1
a
1 5 lDfl
Playable is 1 5 l:tc l (the other rook would stand even better here)
1 6 t'i'Jf2 l:tf7 1 7 t'i'Jb5 i.f8 1 8 c2 t'i'Je8 19 a4 h5 20 b3 i.d7 2 1 h3
l:tg7 22 a5 t'i'Jh4 23 d l g4 24 fxg4 a6 25 t'i'Ja3 t'i'Jf6 26 .te l l:tc8 27 t'i'Jc4
hxg4 28 hxg4 e7 29 l:tc3 i.e8 30 l:th3 t'i'Jh7 3 1 i.b4 f6 32 t'i'Jh 1 t'i'Jg6 3 3
t'i'Jg3 t'i'Jg5 34 l:th 1 l:th7 3 5 .l:Ixh7 'it;xh7 3 6 t'i'Jf5 t'i'Jxe4 37 'ifd3 t'i'Jc5 3 8
'ir'h3+ '1i>g8 39 g 5 'ii'd 8 4 0 t'i'Jfxd6 i.xd6 4 1 t'i'Jxd6 Black resigned, Podgaets
Loginov, Tashkent 1 97 3 .
1 5 . . . tllg 6
1 5 . . . h5
b
c
d
e
r g h
Similar is 15 . . . t'i'Jg6 (see diagram) 1 6 'ii'c 2 !
8
8
%If7 1 7 l:tfc 1 a6 (or 1 7 . . . i.f8 1 8 t'i'Jb5 t'i'Je8 1 9
7
a4 h5 20 h3 t'i'Jh4 2 1 l:ta3 a6 22 l:tc3 i.d7 23 7
6
t'i'Ja3 l:tg7 24 'ifd l l:tb8 25 i.e l 'iff6 26 t'i'Jc4 6
b5 27 axb5 axb5 28 t'i'Ja5 'it'g6 29 t'i'Jc6 l:ta8 30 5
l:ta3 l:txa3 3 1 bxa3 t'i'Jf6 32 t'i'Jb8 i.e8 3 3 l:tc8 4
4
and White was quicker, Kozul-A. Srebmic, 3
3
N o va Gorica 1 999) 18 a4 h5 19 h3 g4 20 fxg4 2
2
hxg4 2 1 hxg4 i.f8 22 a5 b5 23 axb6 'it'xb6 24 1
t'i'Ja4 a7 25 i.a5 l:tb8 26 g5 t'i'Jg4 27 i.f3
"
b
c
d
e
f g h
t'i'Je3 28 e2 l:ttb7 29 l:tc6 l:tg7 30 l:tac1 i.d7
3 1 lba6 'it'd4 32 t'i'Jb6 t'i'Jh4 33 t'i'Jxd7 t'i'Jxf3+ 34 xf3 l:txd7 35 i.c3 'ii'c4
36 l:tc6 'ifb5 37 t'i'Jg4 t'i'Jxg4 38 xg4 l:tg7 39 'ii'e6+ h7 40 ..tr>f2 Black
resigned, Kozul-Shchekachev, Linares 1 996.
1 8 l:tfcl a6 1 9 a4 l2Jh4
More passive is 19 ... i.f8 (see diagram) 20
a5 g4 (20 . . . b5?! 2 1 axb6 xb6 22 tll a4 a7
23 i. a5) 2 1 fxg4 hxg4 22 hxg4 b5 23 axb6
'ifxb6 24 t'i'Ja4 "fia7 25 i.a5 l:tb8 26 g5 ! t'i'Jh7
( 2 6 . f3 ? ! 27 i.xf3 t'i'Jg4 28 d2 ! l:th7 29 i.b6
l::!:. x b6 30 llxc8 was superior for White,
Spassov-Danailov, Pamporovo 1 98 1 ) 27 t'i'Jb6
f3 28 i.xf3 l:r.xb6! 29 i.xb6 'iVxb6 30 xeS
. .
8
7
3
2
3
2
4
3
2
a
llJxg5 3 1 l:lxa6 'ife3 ? (Black' s only chance was 3 1 . . .'ii'xb2 32 'ii'c 2 ! _.d4 3 3
l:la3) 32 l:la3 'iib6 33 'ii'g4 i. h 6 34 l:lc2 <Jo>h7 35 g3 ! and White won,
Rogers-Sznapik, Olympiad, Thessaloniki 1 988.
a
2
a
20 'iid 1
An improvement o n 2 0 a S g4 2 1 fxg4 hxg4 2 2 hxg4 b 5 2 3 axb6 'ifxb6
with an unclear position, Kozul-Arapovic, Porec 1 998.
Akopian - Dimitrov
Linares 1 996
1 d4 l2Jf6 2 c4 g6 3 l2Jc3 i.. g7 4 e4 d6 5 l2Jf3 0-0 6 i.. e 2 e5 7 0-0
l2Jc6 8 d5 l2Je7 9 l2Je1 l2Je8 10 l2Jd3 f5 1 1 i.. d 2 fxe4 1 2 l2Jxe4
Black has refrained from the more ambitious retention of pawn tension in
the centre and is now looking for simplification and potential pressure on
White's less well-protected d5-pawn.
a
3
2
12
. . .
c6
8
7
4
3
3
2
2
1
6
5
-....u 1
f
3
2
1
a
1 3 dxc6 !
Alternatives are less efficient:
.'1
8
7
5
4
2
a
l 5 l:tel !
Not 1 5 0Jxe5? i.xe5 1 6 i.xe5 'i!i'e7 and White would lose a piece.
15 . . . lbc7
White is better after 1 5 . . . 0Jd4 16 i.xd4 exd4 1 7 i.g4 .
l 6 'ii a4 !
8
2
a
1 6 c5
..
1
a
2 1 liJf3 ! 'ii f4
If 2 1 . . .'iVe7 22 xg7 xg7 (or 22 . . . lLlxg7 23 'i'xc5) 23 'i'd3 ! lLlf4 24
'ii'xd6 lLlxe2+ 25 fl .
l l ... c5 ! ?
This i s a less frequently played move to try and block the queenside first.
12 dxc6
White may agree with Black' s conception and continue 12 li)d3 f4 1 3
.i.f2 h 5 1 4 b4 b 6 1 5 a4 a5 ! ? 1 6 bxc5 bxc5 1 7 J:lb 1 g 5 1 8 li)xc5 ! dxc5 1 9
.i.xc5 li)g6 2 0 .i.b6 ! (The passed d-pawn will even compensate for White ' s
1 2 . bxc6
.
a
c
15 b4
8
15
Ild8 16 b5
White can _play this move after removing his queen from d-file. For
example, 1 6 'ifc2 I:Df6 1 7 b5 &DeS 1 8 bxc6 'i!Vxc6 1 9 I:Dd5 l:.d7 20 a4 a5 2 1
c 5 &Da7 2 2 cxd6 'ii'x c2 2 3 &Dxc2 fxe4 24 I:Dxf6+ .l:txf6 2 5 fxe4 l:.xfl + 26
i.. xfl with a clear advantage for White, Finegold-Aitounian, Las Vegas
1 994.
a
2
a
1 6 . . . d5? !
Better was 1 6 . . . fxe4 ! 1 7 I:Dxe4 (Black is fine after 1 7 fxe4 .l:txfl + 1 8
i.. x fl I:Df6) 1 7 . . . I:Df5 1 8 bxc6 &Dxe3 1 9 'ii'x e3 'ii'x c6 20 &Dc2 a5 ! with
chances for both sides.
24 :c t ! l:.c8
White has the better game after 24 . . . i.. xc4 25 .l:txc3 i.. x d3 26 .l:txf8+
i.. x f8 27 .l:txd3 &Dxe4 28 i.. e 3 .
27 .txc4 J:.xc4
27 . . . i.xc4 28 li::ld 3 li::lb 5 29 .l:.cc2 li::l d4 30 .l:.cd2 is also better for White.
28 l:.xc4 .txc4 29 .:tb2 ! aS 30 f2 i.a3 3 1 l:tb8+ f7 32 ltb7+
e6
lf 32 . . .<itg8 33 .i.f6 White wins easily.
36 lbo lbxo
1 2 a4
1 2 . . . ltJgS 1 3 aS
a
b
c
d e
f g h
In the earlier encounter Gelfand-J.Polgar,
8
Dos Hermanas 1 995, the game continued 13 8
t'iJd3 (see diagram) 13 ...f4 14 i.f2 hS 1 S cS gS 7
7
1 6 cxd6 cxd6 ( 1 6 . . . t'iJxd6 1 7 t'iJc5 ! ) 1 7 t'iJbS 6
6
a6 1 8 t'iJa3 t'iJh6 19 t'iJc4 g4 20 h1 l:tb8 2 1 s
s
l:tcl ! ? l:tg8 (unclear is 2 l . . . g3 22 i.g 1 gxh2 4
4
23 i.a7 l:ta8 24 i.f2) 22 t'iJb6 i.f6 ? ! 3
3
(22 . . . i.d7 is solid, while 22 . . . g3 allows 23
2
l:t x c 8 ! l:txc8 24 t'iJxc8 'ifxc8 (24 . . . gxf2 25 1
'ii'c 2) 25 i.b6) 23 t'iJxc8 l:txc8 24 'ii'b3
a
b
c
d e f g h
(stronger was 24 i.b6 ! 'ii'd 7 25 l:txc8 'ifxc8 26
'irb3 and if 26 . . . h4? ! 27 t'iJ f2 ! with the better
game) 24 . . . l:tb8! 2S i.b6 d7 26 l:tc4 i.d8 27 l:tfc1 t'iJf7 28 t'iJf2 i.h4 29
'ii'd 1 t'iJh6 30 b3? (30 t'iJd3 i.d8 3 1 i.g l ! ?) 30 . . . i.d8! 3 1 i.xd8 l:txd8 32
t'iJd3 'ifh7 33 l:tc8 (time pressure) 33 ...l:td7 34 l:ta8 l:tdg7 3S l:tcc8 g6 36
lLlf2 t'iJf6 37 l:txg8+ t'iJfxg8 38 'ii'c 2 h7 ? ! (3 8 . . . 'ii'g 5 ! ?) 39 'ifc8 'ifgS 40
l:tb8 'ii' h 4 4 1 t'iJd3 'ife7? (4 l . . . gxf3 ! 42 i.xf3 t'iJg4) 42 b4 t'iJf6 43 'We6!
g6 and a draw was agreed, but White missed 44 'ii'x e7 ! l:txe7 45 l:tc8 !
with a clear advantage.
6
5
4
3
2
1
8
7
2
1
E J.oi):S
12
Ttr
-- --i
- - - - - - - - ,r
, ... 6
.1.%
''jf
v;r
-
ft
/////h
W///h
p"-
- ---&
-g
ft :if{
'
- --
a
1 5 ttJd3
s
7
- -
1 5 . . . ttJh5 1 6 c5 ttJef6 1 7 b4
If 17 c6 ! ? a6 ! with counter-chances.
17 . . . a6!
17 . . . 'ile7 would allow 1 8 b5.
6
5
23 . . . gxf5
Also playable is 23 . . . i.xf5 24 :e 1 g5.
24 b5 i.xb5 ? !
More promising i s either 2 4 . . . lt:Jxe2 25 'i!Vxe2 ii.xb5 2 6 lt:Jxb5 axb5 27
'Wxb5 :g8 or 24 . . . axb5 25 lt:Jxd7 (25 ..txb5 i.xb5 26 lt:Jxb5 :g8)
25 . . . :xd7 (25 . . . lt:Jxe2 26 'i!Vxe2 'iVxd7 27 :ab 1 with even chances) 26
i.xb5 :c7 with active play.
39 :8a7
More consistent was 39 : 1 a7 'i!Vg5 40 lt:Jd7 h6 (or 40 . . . i.e7 4 1 xg8+
'lftxg8 42 'i!Vc8+ 'iti>fl 43 a8 lt:Jf6 44 lt:Jxe5+ dxe5 45 d6 h6 46 dxe7 and
White wins) 4 1 'iti>g 1 i.e7 42 xg8+ 'iti>xg8 43 a8+ 'iti>fl 44 'iVc8 'iVd2 45
ii.fl with a clear advantage.
. .
Korchnoi - Kasparov
Debrecen 1 992
1 d4 lbf6 2 c4 g6 3 lbc3 .tg7 4 e4 d6 5 .te2 0-0 6 llJtJ e5 7 0-0
lbc6 8 d5 lbe7 9 lbel lbe8 10 i.e3 f5 1 1 t3 f4 1 2 i.f2
a
h
8
8
7
1 2 ... h5
b
c
d e r g h
i..d 7 20 i.. b6 'ii'e 8 2 1 'Wb3 g3 with an unclear
game, Teplitsky-Bacrot, Mennaid Beach Club
1 999) 14 a4 (playable is 14 cxd6 tlJxd6 1 5 tiJd3 l:tf6 1 6 l:tc l h5 17 tlJc5
tiJf8 1 8 tiJb5 lL:Ixb5 19 i.. x bS l:tg6 20 'ife l g4 2 1 i.. h 4 i.. f6 22 .i.xf6 'ifxf6
23 tiJd3 ltg7 24 "ii'a 5 with advantage for White, Ivanchuk-Van Wely,
8
1 3 c5 g5
a
1 4 a4 !
b
c d e r g h
14 llcl (see diagram) (White intends 1 5
-..
common continuations 1 4 . . . llf6 or 1 4 . . _ &[jg6
a
b c
d e f g h
are better options.) 1 5 .i.xc5 llf6 1 6 'ii'b 3
(safer is 1 6 lDd3 .l:r.g6 1 7 t:{jf2) 16 . . . h7 17 &[jd3 .l:r.g6 18 &[jb5 a6 19 t:{ja7
llxa7 20 .i.xa7 b6 21 .l:r.c2 .i.d7 22 llfc1 &[jc8 23 j_b8 &[jcd6 24 j_xc7
&[jxc7 25 llxc7 &[jb5 with counter-chances, Korchnoi-G.Hernandez, Merida
1 996.
"
l>
14
tbg6
. . .
li:'Jf2 'ii'e2 39 'ii'x e2 fxe2 40 li:'Jd3 e4 4 1 li:'Je1 i.. h 6 42 l'Je7 i.. d 2 White
resigned, Shirov-J.Polgar, Dos Hermanas 1 99S.
a
15 aS
a
2
1
1 5 J.h6 ! ?
...
16 b4
Stronger than 16 c6 bxc6 1 7 dxc6 a6 ! . Another possibility is 16 &iJb5 a6
1 7 tlJa3 'i!;lh8 1 8 tlJc4 :g8 19 :a3 &iJf6 20 cxd6 cxd6 2 1 &iJb6 (playable was
2 1 it'b3 ile7 22 it'b4 ! ..tf8 [or 22 . . . :d8 23 &iJb6 :b8 24 h3 with the better
game] 23 &iJb6 :b8 24 :c3 i.h6 25 &iJd3 g4 26 :rei) 2 1 . . .:b8 22 l:tc3 g4 !
23 fxg4 tlJxe4 ! ? 24 l:txc8 l:txc8 25 tlJxc8 fixeS 26 ..tb6 tlJe7 27 gxh5 &iJf6
with unclear chances, Korchnoi-Shirov, Buenos Aires 1 993 .
1 6 . . . h7 1 7 cxd6
Unclear is 1 7 a6 b6 1 8 i.b5 .
1 7 . . . cxd6 1 8 lDb5 g4
After 1 8 . . . a6 1 9 tlJa3 &iJf6 20 l:tc l g4 2 1 i.b6 'fke8 22 l:tc7+ l:tf7 23 tlJc4
White had a somewhat preferable position in Zuger-Gallagher, Switzerland
1 993 . In the present game Kasparov wants to accelerate his attack by a
pawn sacrifice.
2
1
22 'i'O
22 . . .'ifd7! 23 lDxa7!
Passive is 23 tlJc3 l:tac8 with pressure.
26 lDxf3 'iig4 27 b6
There is no time for 27 .l:.ae i t:Df4 with many threats.
30 l:.ad l
White rejects the possible 30 .l:.ed i 'ii'e 6 3 I t:Db5 e4 32 t:Dc7 'ii'g4 3 3
t:Dxe8 exf3 34 t:Dxd6 h8 ! with a n extremely sharp position. Both rivals
stopped counting material in their quest for the initiative.
30 . . . Wxa5 31 'i'c7+
Of course not 3 I .l:.xd6? 'ii'x e I + winning.
34 ltJxg6
40 . . . ..th6
Or 40 . . . .l:.a8 4 I t:Db5 .l:.aa2 42 .l:.g5 .l:.a5 43 .l:.ff5 lla i + 43 .l:.fl .l:.a5 with a
draw.
Game 20
Gelfand - Topalov
Wijk aan Zee 1 996
1 ctJfJ ctJf6 2 c4 g6 3 ctJc3 i.g7 4 e4 d6 5 d4 0-0 6 i.e2 e5 7 0-0
ctJc6 8 d5 ctJe7 9 ctJe1 ctJe8 10 .ie3 f5 1 1 f3 f4 12 i.f2 h5 13 c5 g5
14 a4 dxc5 ? !
More consistent i s 1 4 . . . l:r.f6-see the next game, Shirov-Nunn, Amster
dam 1 99 5 . The idea of the move in the game is to spend a tempo forcing the
opponent' s bishop to the c5 square, which ideally White' s knight would like
to occupy.
4
3
4
3
1
a
More ambitious than 1 7 ti:Jf2 ! ?. White vacates the best square c5 for his
knight, and prepares to meet an eventual incursion of the opponent' s queen
by occupying the h4-d8 diagonal first.
17
. . .
g4
. .
18 i.h4 a6
The alternative was 1 8 . . . ti:Jd6 1 9 ti:Jc5 'i'f8 20 ti:Jb5 ! ? (or 20 ti:Je6 .i.xe6
19 g3 ! 'i!Vd6
24
..
2
1
a
28 ti:Jg4?
White misses the strongest line: 28 ti:Jd3 ! ! .l:txg3 29 'ii'f2 + 'ii'f6 3 0 .l:txg3 ,
or 28 . . . .l:tgh6 29 .l:tafl + e8 30 .l:tt2 with the better chances .
28
. .
39
..
41 c;a;,o i.. x c3 42 bxc3 <it>e5 would leave a winning p awn endgame for
Black.
Variations after
I 1 j3
87
Game 2 1
Shirov - Nunn
Amsterdam 1 995
1 d4 tLlf6 2 c4 g6 3 tLlc3 i.g7 4 e4 d6 5 i.e2 0-0 6 tLlf3 e5 7 0-0
tLlc6 8 d5 tLle7 9 tLle1 tLle8 10 i.e3 f5 1 1 f3 f4 12 i.f2 h5 13 c5 g5
14 a4 l;lf6
a
2
1
a
15 a5
Playable is 1 5 cxd6 t:l:Jxd6 (or 1 5 . . . cxd6 16 aS t:l:Jc7 [Black prevents t:i:Jb5
but thereby also wastes precious time; nevertheless 1 6 . . J 1g6 17 ll'lb5 a6 I 8
..ib6 d7 I 9 ll'la7 is also preferable for White] I 7 h i l::th6 1 8 ll'lc2 ll'lg6
1 9 ll'la3 g4 20 fxg4 hxg4 2 1 J.xg4 'ikg5 22 .i xc8 l::tx c8 23 ll'lc4 ll'le8 24
t:i:Jd2 ll'lf6 25 ll'lf3 1i'h5 26 h3 and Black could not strengthen his attack,
Kozul-Fercec, Pula I 993) 16 ll'lb5 ( 1 6 ll'ld3 ll'lg6 1 7 ll'lc5 is another plan)
16 a6 1 7 ll'lxd6 cxd6 and now:
. . .
8
7
6
1
3
2
Jl.f6 23 ti::l f2 ti::l h 6 24 .l:r.c3 1i'g8 25 ii'c2 g4 26 fxg4 Jl.h4 27 .l:r.cl Jl.d7 28
.l:r.c7 .l:r.g7 29 Jl.f3 hxg4 30 ti::l x g4 ti::l x g4 3 1 i.xg4 i.xg4 32 hxg4 .l:r.xg4 33
'ife2 .l:r.f8 34 'iff3 i.g3 35 'iti>n .l:r.h4 3 6 'iti>e2 'ii'g 6 37 .l:r.e7 .l:r.h2 38 .l:r.e6 'ifg5
39 'iti>d3 i.h4 40 .l:r.c2 .l:r.g8 41 .l:r.xd6 .tel 42 .l:r.e6 .l:r.h6 43 .l:r.xe5 ii'g7 44 .l:r.h5
1-0 Grabliauskas-McShane, Roskilde 1 99 8 .
8
"
=.!0-.....a
3
2
17 lbd3
8
t!
4
3
Alternatives are:
1 7 h3 (see diagram) (too cautious) 1 7 . . . h8
18 ti::l d 3 ti::l g 8 1 9 ti::l c S ti::l h 6 ( 1 9 . . . 'ife7 led to
messy play after 20 .l:r.a3 ti::lh6 2 1 ti::lb 5 g4 22
fxg4 hxg4 23 ti::lxd6 cxd6 24 ti::l e6 i.xe6 25
dxe6 f3 in Korchnoi-J.Polgar, Monaco
blindfold 1 994) 20 ti::l e 6 (a standard move
blocking the i.c8 's diagonal and thereby reducing the power of . . . g5-g4; however now it
..
1 7 . . . g4 1 8 i.h4 'tif8
1 8 . . . g3 is premature because of 1 9 hxg3 fxg3 20 'ife 1 ti::Jc4 2 1 ti::J d 1 .i.h6
h
8
8
7
2
1
a
2 1 liJxf4?
A wrong combination. Correct was 2 1 .i.xf6 'ifxf6 22 l:lf2 (Nunn' s
suggestion i s very interesting: 22 gxf4 g 3 23 Ji.h5 ! 'ii'h4 2 4 'i!Ve2 gxh2++ 2 5
'it>h 1 .l:f.g 1 + 2 6 .l:f.xg l hxg 1 ='if+ 27 <Ji>xg 1 .i.h3 28 ti::Jx e5 <Ji>h8 29 f5 ! 'ifg3+
30 <Ji>h 1 'ifxe5 3 1 l'fh2 l'fxh2+ 32 xh2 i.xf5 33 exf5 ti::Je xf5 with an equal
90 Variations a
game) 22 . . f3 2
fl i.. d7 24 Ci:Jc5 .l:.d8 25 'ifb3 a6 26 Ci:Jd 1 i.. c 8 27 Ci:Je3
q;g7 28 .l:.c 1 .l:.f8 with complex play, Kiriakov-Sotnikov, Moscow 1 995.
.
23 . . . .U.xf6 24 l:xf6
If 24 e5 .l:.xf4 25 exd6 .l:.f2 26 xf2 'iVxh2+ 27 e 1 (or 27 fl Ci:Jf5)
27 . . . 'ilxg3+ 28 d2 'ikf4+ (equally good is 28 . . . 'iVxd6) 29 c2 Ci:Jf5 3 0
b 1 Ci:Je3 with a strong initiative.
28 :n 'ilid4+ 29 g2 g7
a
8
7
6
5
4
3
30 l:tf4 ? !
Or 3 0 h4 lLlc4 3 1 .l:.e 1 .l:.f8 ! 32 h5 l:tf2+ 33 h3 (if 33 h 1 'ii'd2 34
'ii'xg6+ <itt f8 wins) 33 . . . Ci:Jce5 34 'ikd 1 (34 'iig 5 l:.h2+) 34 . . . .l:.d2 with a
mating attack.
30
. .
Part Three :
Variation 9 etJe 1 etJd7 1 0 f3 f5 1 1 g4
This, so to speak, prophylactic vanat10n has long enjoyed sporadic
popularity. Before Black tightens his grip around the white king by advanc
ing the g-pawn to g3 , White is quick to grab space in front of his king with
g2-g4, thereby reducing the risk of being exposed to a mating attack and
potentially giving him a free hand to increase his queenside initiative
undisturbed. The continuation has both advantages and drawbacks-the
latter being White' s problem of how to successfully maintain the kingside
blockade and not run into trouble from any opening of the h-file. As with
the other variations discussed here, due to the abundance of material we
shall select only a few games which characteristically represent the ideas of
the line.
Game 22
Pinter - Nunn
Olympiad, Thessaloniki 1 988
1 d4 tL!f6 2 c4 g6 3 tLlc3 .i.g7 4 e4 d6 5 i.e2 0-0 6 tL!fJ e5 7 0-0
tLlc6 8 d5 tL!e7 9 tLle1 tLld7 10 f3 f5 1 1 g4 <it>h8
a
1 2 tLlg2
This move shows that White's primary concern is a successful blockade
of the kingside. Two years earl ier, at the Olympiad in Dubai, the same
I I g4
1 2 ... a5
Preparing to exert pressure on the e4-pawn with the knights on c5 and f6.
13 h4
Or 1 3 h3 t'Llc5 1 4 .ie3 b6 1 5 b3 .id7 ! with a good game, Miladinovic
lstratescu, Olympiad, Elista 1 998.
1 9 tLlbS tt:lf6
Exerting pressure on both e4 and g4 .
Lesiege - Fedorowicz
New York 1 993
14 ..tf2 h5! ?
More solid is 1 4 . . . g5 1 5 b4 (If 1 5 h4 h5 1 6 gxh5 ltJdf6 1 7 hxg5 ltJxh5
with a very good game) 1 5 . . . h5 16 h3 Af7 with equal chances.
1 5 g5 !
Stronger than 1 5 h3 hxg4 1 6 hxg4 i.f6 1 7 i.e 1 with an equal position.
20 ctJxa8 .:txa8
2 1 hl ?
A wrong move order. The correct one was 2 1 c 5 g4 22 h l ! 'i'g5 2 3 ltg 1
dxc5 24 .1Lxc5 h4 25 b4 with a good game.
2 1 . .. h4 !
Now Black is better.
3 2 .ib4
a
8
7
6
5
4
3
2
The next, very interesting game has a different move order, but reaches a
position quite typical of this variation.
Game 24
Gelfand - Kasparov
Reggio Emilia 1 991/92
I d4 tLlf6 2 c4 g6 3 tLlc3 i.g7 4 e4 d6 5 i.e2 0-0 6 tLlf3 e5 7 0-0
13 g4 g5 1 4 b4
Worth mentioning is 14 e 1 h5 1 5 h3 l:i:!.g6 16 c5 (or 16 g2 l:tf7 1 7
f2 f8 1 8 b4 l:th7 1 9 l:th 1 e7 20 'iib 3 g7 2 1 c5 .id7 22 l:tac 1 a6 !
with an unclear game, Lukacs-Grivas, Budapest 1 993) 1 6 . . . l:tf7 1 7 cxd6
r xd6 1 8 l:tc 1 .if8 1 9 .i.f2 l:th7 20 'iib 3 hxg4 2 1 hxg4 l:i:!.h4 22 ..t.xh4 l:txh4
23 l:i:!.t2 a6 ! 24 a4 l:[h7 25 g2 d7 with equal chances, Khalifinan-Spasov,
Manila 1 990.
1 4 h5 15 h3 fi
.
18 c5
More solid was 1 8 .if2 a5 1 9 a3 hxg4 20 hxg4 l:i:!.h4+ 2 1 .ixh4 l:[xh4 22
l:th 1 'ii'h 8 with an equal game.
1 8 hxg4 1 9 hxg4
..
1
a
1 9 ... ll'lh5 !
White would be slightly better after 1 9 . . . l[jh4+ 20 .ixh4 l:hh4 2 1 .l:th l .
20 l:h 1
lf 20 gxh5 l:txh5 (20 . . . .ih3+ 2 1 g l ! , but not 2 1 xh3 l:txh5+ 22 g2
'ifh8 23 .th4 l[jxh4+ 24 f2 l[jg2 ! ! 25 l:th1 l:th2) 2 1 l:th 1 (or 2 1 l[jf2
l[jh4+ ! 22 g 1 'ii' h 8 with the attack) 2 1 . . .l:txh 1 22 <iti>xh 1 'ii'h 8+ 23 <itl>g2
'ii'h 5 24 .tfl g4 with a strong attack.
of
26 b5 g5 27 'ii'e l e3
More precise is 27 . . . <iti>g7 ! 28 c6 .ic8 29 a4 a6 etc.
28 c6 c8
Better than 28 . . . bxc6 29 bxc6 .ic8 30 l:tb l .
38 ... c8
1/z-1/z.
Variation 9 l2Jd2
This continuation was the result o f a search for a plan t o develop White ' s
queenside initiative a s actively as possible and i t has enj oyed great popular
ity in grandmaster practice. White is not happy with the d3 square for the
king ' s knight and instead wants to send it to its most efficient post-the c4
square.
One of Black' s first ideas was to take the opportunity to exchange the
dark-squared bishop with 9 . . . i.h6, but this makes his kingside vulnerable
and speeds up the development of White ' s queenside after an exchange on
cl.
Game 2 5
Gligoric - Littleton
Zonal, Praia da Rocha 1 969
1 d4 lbf6 2 c4 g6 3 lbc3 g7 4 e4 d6 5 l2Jf3 0-0 6 e2 e5 7 0-0
lbc6 8 d5 lbe7 9 lbd2 h6 10 b4 lbe8
a
1 1 lbb3
In the game Gligoric-Vukic, Saraj evo 1 969, I played the riskier 1 1 c5 f5
1 2 exf5 gxf5 1 3 ll':lc4 i.xc l 1 4 l:lxc l ll':lg6 1 5 i.h5 'ile7 1 6 i.xg6 hxg6 1 7
f4 exf4 1 8 l:lxf4 dxc5 1 9 bxc5 'fixeS+ 2 0 h i 'iie 7 2 1 'ii'd2 ll':ld6 22 ll:lxd6
cxd6 23 l:le l 'iff6 24 l:lf3 i.d7 and won after 25 ll':le4 fxe4 26 l:txf6 l:txf6 27
h3 l:le8 28 'ifd4 etc.
98 Variation 9 1:iJd2
1
a
Korchnoi - Gligoric
Havana 1 969
1 d4 ttlf6 2 c4 g6 3 ttlc3 ..tg7 4 e4 d6 5 ttlf3 0-0 6 .tel e5 7 0-0
ttlc6 8 d5 ttle7 9 ttldl !
Several months earlier, this move was played in the game Taimanov
Tatai, Venice 1 969 (9 l:iJd2 .i.h6 1 0 b4 a5 1 1 bxa5 .l:ba5 12 l:iJb3 i.xc 1 1 3
Variation 9 /:i)d2 99
9 ... d7 1 0 b4 f5
l l b3
White prepares c4-c5, but more energetic is the sacrifice 1 1 c5. Now
accepting the challenge by taking the pawn 1 l . . .dxc5 1 2 bxc5 /:i)xc5 1 3
i.a3 b6 1 4 i.. x c5 bxc5 1 5 /:i)b3 'iti>h8 1 6 'i!fd2 fxe4 1 7 /:i)xc5 /:i)f5 1 8 /:i)5xe4
i.. h6 1 9 'ii'd 3 is slightly better for White, Lputian-Akopian, Erevan 1 98 8 .
A more complex game arises after 1 1 . . ./:i)f6 1 2 f3 f4 (Also bad i s 1 2 . . . a 5
1 3 bxa5 dxc5 1 4 /:i)c4 Geller-Zaitsev, USSR
1 969) 13 /:i)c4 g5 (see diagram)
8
b
c
d e f g h
players have to an equal extent increased the
pressure on both the queenside and the centre with the continuations 1 4
i.. a 3 and 1 4 a4 .
14 i.. a 3 /:i)g6 15 b5 /:i)e8 (Stronger than 1 5 . . . dxc5 although with this con
tinuation too Black players have been successful in practice. For example
1 6 .i.xc5 l:[f7 1 7 a4 h5 1 8 a5 g4 19 b6 [ 19 i.. 2 i.. f8 20 'ii'd2 g3 2 1 hxg3
l:[g7 ! with counterplay, Lputian-Gadj ily, Dubai 2000 (0- 1 , 37)] 1 9 . . . g3 20
'iti>h 1 /:i)h7 2 1 d6 'iih4 22 i.. g 1 i.h3 23 bxc7? [23 gxh3 ! 'ilfxh3 24 .l:[2 gx2
25 .i.x2] 23 . . . .i.xg2+ 24 'it.?xg2 'it'h3+ 25 xh3 lLlg5+ 26 g2 /:i)h4+ 0- 1
Ftacnik-Cvitan, Germany 1 997) 1 6 b6! cxb6 ! ? ( 1 6 . . . axb6 1 7 cxb6 cxb6 1 8
'ii'b 3 Shirov-Akopian, Daugavpils 1 989) 1 7 /:i)xd6? ! (Better is 1 7 cxd6 ! ?
1 00 Variation 9 0.d2
l l . . . fxe4 !
The motive for this move lies in the fact that one of the white knights has
travelled a long way from the centre and the e4 square. The exclamation
mark relates more to the fact that the opponent 'did not understand ' the
move than to giving any advantage to Black. In any case, the exchange in
the centre decreases White 's pressure on the queenside, because the pawn
on d5 is now not so solid, and at the same time it frees the f5 square for the
activation of the passive knight on e7.
In fact, I applied a similar idea against Korchnoi at the Interzonal tourna
ment in Sousse 1 967, after 9 li:Je 1 tt:\d7 1 0 d2 f5 1 1 l:tc 1 (White expected
1 l . . .li:Jf6 1 2 f3 f4 when he could play 1 3 c5 gaining a tempo without the
preparatory b2-b4, as in Larsen-Najdorf, Santa Monica 1 966). I played
1 l . . . fxe4 ! 12 li:Jxe4 li:Jf6 1 3 .i.d3 li:Jf5 14 .i.g5 c5 ! 1 5 .l:r.c3 'ilfc7 16 'ilfd2
li:Jd4 and a draw was agreed here, but Black's position is really very good.
Variation 9 CDd2 1 01
1 2 tbxe4 tbf6
a
13 tbg3 ! ?
This was the first move i n the game on which White spent any amount of
time. It was more natural to play 1 3 i.f3, intending to maintain the strong
hold on e4. Perhaps Korchnoi remembered our game in Sousse where he
had not achieved much with such a strategy and therefore preferred to try
something else in order to surprise his opponent. But th ' surprise' is not
unpleasant, because the manoeuvre actually enhances Black's development.
White probably wanted to free some space and remove the tension in the
centre so as to gain a free hand on the queenside. Perhaps he also reckoned
that it was better to exchange the second black knight, which was threaten
ing to become active.
15 i.e3 h5!
During the game this move too was assessed as weak by White, who sim
ply took it as a weakening of the black kingside. In fact, the move enables
the black bishop to take up a strong position on f5 , and, more importantly,
threatens 1 6 tl:Jg4, which provokes White to start moving one of his pawns
on the kingside.
. . .
16 f3
The move seems necessary, but it also makes the kingside (inaccessible
while all the pawns were on the second rank) liable to be opened up-and
Black's whole strategy is actually based on the sharpening of the game on
this part of the board and in the centre. The pawn on f3 is the target of the
. . . e5-e4 breakthrough.
1 02 Variation 9 ti:ld2
1 6 'iVd7 1 7 'ii d 2
.
White provides a better protection of his pieces along the e-file, because
Black can ignore the aggressive 1 7 Ci:la5 and reply with 1 7 . e4.
.
1 7 ... b6
On the direct 1 7 :ae8 Black has to reckon on 1 8 .i.xa 7 b6 1 9 c5. The
preventive move in the game restricts the activity of the white knight and
the dark-squared bishop.
...
The alternative was to attack the white centre with 17 c6, but this would
be more favourable for White since his pieces are well-placed for exerting
pressure along the queenside files.
...
18 a4 ltae8
While White plays on the queenside and attacks over there, Black is com
pletely ready to react with a counter-action in the centre which is the more
important part of the board.
1 9 aS
On 1 9 c5, threatening 20 i.. b 5, Black can easily respond with 1 9 .'Wt7.
..
1 9 ... e4
This is the main point of Black's strategy. The number of squares through
which the black pieces can penetrate to the kingside increases more than
enough to neutralize the pressure that White exerts on the queenside.
20 axb6
20 f4 Ci:lg4 2 1 i.. xg4 hxg4 ! also provides a good game for Black.
20 ... exf3
Black exploits his chances to open the position in the centre of the board
as quickly as possible.
Variation 9 ltJd2 1 03
White constantly plays under the wrong impression that he is the one who
has the initiative. Instead of this move, he should have tried to maintain the
balance in the centre, which is more important for the outcome of the battle.
But after the better 22 ltJd4 ltJe4 23 .ixe4 xe4 or 23 'iic l .ig4 24 .ixg4
xfl + 25 'iix fl 'iixg4 White is the one who has to defend himself, which is
the result of his incorrect conception, starting with the 1 3th move, as well as
of his underestimation of Black's counter-chances in the centre.
23 i.g4!
..
Black doesn 't lose a single moment on the threat 24 .ixb6, because just
one passive move could reduce the impact of his attack on the white king.
24 i.xg4
Obviously 24 .ixe4 xfl + 25 'iixfl xe4 26 .ixb6 'iie 8 27 .i2 .ie2
cannot appeal to White, nor can 24 .ixb6 .ixf3 25 gxf3 xf3 ! .
24 ...'ifxg4 25 l!xc7
It would be too late now for White even if he had realized the impending
danger. 25 h3 'iig 3 26 .if4 was no good because the knight on b3 is
hanging.
a
1
a
This is the climax in the Rashomon story-where every side has its own,
different truth. White believes he is winning, when he is, actually, lost! He
took the pawn on c7 quickly enough, thinking that the direct threat 25 . . . ltJc3
was not dangerous because of 26 xg7+ and that Black's whole action was
merely a reflection of his despair.
25 ... i.e5!
Only now, when the white rook has become useless on the vacated
seventh rank, did White exactly appreciate his predicament. There is no
1 04 Variation 9 &i::ld2
26 i.d4
White plays the only move to prevent the sacrifice on h2. But Black has
other ways as well to continue his deadly attack.
28 .. .'ii f4 ! 29 g3
29 i.xe5 is no good because of 29 . . . 'ii' f2 + and mate on the back rank.
Larsen - Fischer
Game Four, Candidates match, Denver 1 9 71
1 c4 g6 2 ttJO i.g7 3 d4 ttJf6 4 tDc3 0-0 5 e4 d6 6 i.e2 e5 7 0-0
ttJc6 8 d5 tDe7 9 ttJd2 c5
This move has additional justification because the knight on d2 has
obstructed the d-file, so Black doesn ' t have to worry about pressure on the
d5 and d6 squares.
a
10 .l:.b 1 !
2
a
Variation 9 !Cd2 1 05
1 0 ... ttJe8
This is better than 10 . . . a5 ! ? (Black should not tamper too much with the
side where he is weaker! ) 1 1 a3 !Ce8 12 b4 axb4 1 3 axb4 b6 14 bxc5 bxc5
1 5 !Cb3 f5 16 f3 h8 1 7 i.g5 f4 1 8 .l:.a l .l:.b8 19 i.h4 ..tf6 20 i.e ! g5 2 1
!CbS .l:.g8 22 .l:.a7 .l:.b7 2 3 .l:.a8 'iWd7 24 !CaS and the two open files on the
queenside made the penetration of the white pieces easier and this led to
victory on the 45th move, Gligoric-Zuckerman, Los Angeles 1 974.
1 1 b4 b6 1 2 a4
No better is 12 'ir'a4 f5 13 bxc5 bxc5 14 !Cb3 fxe4 15 !Cxe4 !Cf6 16 f3
!Cxe4 1 7 fxe4 .l:.xfl + 1 8 ..txfl h8 1 9 'ii' a 5 'iWd7 20 ..te2 !Cg8 I_h - I.h ,
Gligoric-Ciocaltea, Hastings 1 97 1 172.
Or 12 bxc5 bxc5 13 !Cb3 'it>h8 ! 14 i.d2 f5 15 !CbS ! ? fxe4 ! 16 ..ta5 'ifd7
1 7 !Cd2 'ii' f5 ! 1 8 !Cc3 !Cf6 with equal play, Shirov-Lanka, Torey 1 990.
24 ttJe2 f4 !
Tightening the noose around the white king.
2
a
1 06 Variation 9 li)d2
1 0 a3 lL!d7
Black tries to activate the pawn mass on the kingside as quickly as poss
ible. Instead of this frequently played move perhaps more precise is
10 . . . d7 with the idea of gaining a tempo for development by threatening a
blockade with . . . a5-a4 (see the examples below). In case of 1 0 . . . li)e8 1 1
l:tb l f5 1 2 b4 axb4 1 3 axb4 li)f6 1 4 c5 h8 1 5 f3 f4 1 6 li)c4 li)e8 1 7 li)a4 !
l:tb8 1 8 b5 b6 1 9 cxb6 cxb6 20 i.d2 d7 2 1 lL!cxb6! White is clearly
superior, Sakaev-Tseitlin, St. Petersburg 1 997.
1 1 l:.b 1 f5 1 2 b4 'it>h8
Passive is 12 . . . b6 13 f3 f4 1 4 li)a4 axb4 1 5 axb4 g5 1 6 c5 li)f6 1 7 cxd6
cxd6 1 8 b5 d7 1 9 li)c4 li)c8 20 .ta3 li)e8 2 1 g4 ! fxg3 22 hxg3 g4 ! 23
.te l ! gxf3 24 .txf3 li)f6 25 .tg5 ! with the better game, Kasparov-Smirin,
USSR 1 98 8 .
Variation 9 lDd2 1 0 7
13 f3
2
a
1 3 ttJg8
...
An interesting plan, starting from the 1 51h move, is 13 ... axb4 14 axb4 c6
lDa5 g5 1 9 lDc4 lDg6 20 lDb5 :a6 2 1 i.d2 :g8 22 :c 1 i.f8 23 'ifc2 i.e7
1 08 Variation 9 ti:Jd2
24 tl::J a 5 g4 25 tl::J c 7 gx3 26 .i.x3 .lia7 27 tl::J e 6 .txe6 28 dxe6 ti:::J h4 29 .te l
ti:::J x3 30 gx3 .lia8 with unclear chances, Chuchelov-Nataf, France 2000)
15 ti:::J f6 16 ti:::J fl cxd5 17 cxd5 f4 18 .i.d2 g5 19 .tel .lig8 20 .lial .lixal 2 1
'ii'x al g 4 22 'i'a5 b 6 2 3 'i'a7 .t f8 24 .i d l tl::J g 6 25 .lic2 gxf3 26 .i.xf3 .i.g4
27 ti:::J d 2 'i'c8 28 'i'a2 'ii' d 7 29 'ii'c 4 'ii'g7 30 ti:::Jb 5 'ii'h 6 31 'ii'e 2 tl::J h 4 32
.txh4 'ii'x h4 33 .lic6 i-d7 34 .lixb6 tl::J g4 35 ti:::J fl 'ii'd 8 36 .lib7 'ii'c 8 37 .lic7
'ii'b8 38 .lixd7 'ii'b 6+ 39 tl::J e3 tl::J x e3 40 hl .lig7 41 .lixg7 xg7 42 g3
i.. e7 43 gxf4 exf4 44 tl::J c3 'ii'x b4 45 ti:::J d l tl::J x dl 46 'ii'x dl 'ii'c 5 47 .tg2
'ii'f2 48 'Wg4+ 1/2-1/2, Beliavsky-Antic, Yugoslavia 200 1 .
.
1 4 'ifc2
a
8
7
6
14 . . . liJgf6
In the game Chernin-J.Polgar, Groningen 1 993, Black tried 1 4 . . . ti:::J h6 1 5
ti:::J b 5 axb4 1 6 axb4 ti:::J f6 1 7 c5 ti:::J h 5 1 8 cxd6 cxd6 1 9 tl::J c4 .lia6 20 g3 with
unclear chances.
1 5 i.. d3
8
7
6
4
6
5
4
b
"
d e f g h
did not succeed, K.amsky-K.asparov, Dortmund
1 992, or 1 9 . . . tl::J x g3 ? ! 20 hxg3 ti:::Jh 5 2 1 f4 ! Karpov-Kasparov, Tilburg 1 99 1 ,
and Epishin-Piket, Wij k aan Zee 1 992) 1 9 fxe4 .ih3 2 0 .lif2 'ir'd7 with
equal chances, Epishin-Van Wely, Wijk aan Zee 1 992, or 20 . . . tl::J g4 2 1
2
Variation 9 tiJd2 1 09
1 5 ... f4 1 6 i.e2
Also playable is 16 tiJb5 b6 17 c5 dxc5 1 8 bxa5 xa5 19 t?Jc4 (premature
is 1 9 d6? ! t?Je8 ! 20 dxc7 t?Jxc7 2 1 tiJd6 a8 etc.) 1 9 . . . a8 20 a4 tlJ e8 21 a5
i.a6 22 i.d2 .ixb5 23 .ixb5 tiJd6 24 ltb2 with compensation for the pawn,
lvanchuk-Kasparov, Linares 1 992.
2 0 g 5 2 1 b5 .1Ixa4 ! ?
2
a
25 i.xc5
Correct was 25 h3 ! 'ir'e8 (either 25 . . . b6 26 .ixc5 bxc5 27 b6, or 25 . . . 'iff8
26 .ixc5 .ixh3 27 gxh3 g2 28 f2 is winning for White) 26 t?Jxe5 ! dxe5 27
i.xc5 with the better chances.
25 ... gxh2+
Not 25 . . . "ile8 26 h3 'ifh5 27 .id 1 ! dxc5 28 b6 etc.
26 xh2 'fi'f8
If 26 . . . tiJh5 27 lth 1 ! .
1 1 0 Variation 9 ti)d2
27 lLlxe5?
Correct was 27 :th i ! dxc5 28 <ifi>g i with the advantage.
29 l:tfc l llJhS 30 fl
Not 30 'ikxc7? Wh4+ 3 I g i :txg2+ 32 xg2 h3+ 33 h i ti)g3+ and
Black wins.
Sakaev - Nevednichy
Olympiad, Elista 1 998
1 d4 llJf6 2 c4 g6 3 lLlc3 g7 4 e4 0-0 5 lLlfJ d6 6 e2 e5 7 0-0
lLlc6 8 d5 lLle7 9 lLld2 aS 1 0 a3 d7 1 1 b3
8
3
2
2
a
8
3
2
=s-.....a
Variation 9 c!Lld2 I l l
..
l l c6
...
...
1 1 2 Variation 9 l?Jd2
7
5
2
1
a
12 i.b2
An interesting idea is 12 :a2 (see diagram) 12 ...iih8 (alternatives are
1 2 . . . 6 1 3 :c2 l?Je8 1 4 dxc6 l?Jxc6 1 5 l?Jd5 'iid 8 1 6 l?Jf3 M.Gurevich
Stryjecki, Pardubice 2000; 1 2 . . . l?Je8 1 3 l?Jdb l cxd5 14 cxd5 f5 1 5 a4 ::r.c8
1 6 i.d2 l?Jf6 1 7 l?Ja3 fxe4 1 8 l?Jc4 l?Jf5 1 9
"
c
e
1'
d
f
1'
.::t e l Portisch-Nijboer, Olympiad, Istanbul
8
8
2000; 1 2 . . . i.h6 1 3 'iic 2 c5 1 4 l?Jf3 i.xc 1 1 5
7
7
'ifxc 1 l?Jh5 Guseinov-A.Kuzmin, Dubai 1 999;
6
6 1 2 . . . c5 1 3 l?Jdb 1 f5 1 4 l?Jb5 'itr>h8 1 5 l?J l c3
5
5
l?Jg8 1 6 f4 Portisch-Babula, Olympiad,
4
4 Istanbul 2000) 13 :c2 :cs 14 i.d3 i.e8 1 5
3 l?Jdb1 b5 1 6 cxb5 cxb5 1 7 b 4 l?Jh5 1 8 g 3 f5
3
2 1 9 bxa5 l?Jf6 20 ::r.b2 with a strong initiative
2
=-..a 1
on the queenside, Portisch-Iordachescu, Berlin
1 997.
a
b
c
d
e
f g h
Variation 9 d2 1 1 3
...
8
7
6
5
1 2 . c5
..
1 3 b5 lbe8
1 4 b4
6
5
14 b6
...
1 1 4 Variation 9 Ci::Jd2
15 bxc5 bxc5 1 6 f4 !
Emphasizing the usefulness of the bishop on b2. Less dangerous is 1 6 a4
f5 1 7 l:.a3 i.h6 1 8 f3 i.c8 1 9 :e 1 Ci::J f6 20 i.d3 fxe4 2 1 l"i::Jxe4 Ci::J f5 22 i.e 1
i.f4 23 Ci::Jxf6+ 'ii'x f6 24 i.xf4 exf4 25 i.xf5 i.xf5 I/2-Ih, Piket-Gelfand,
Tilburg 1 996.
20 'ilb3
25 h2 f6 26 lbf3 lbxfJ+ ? !
Correct was t o retain the knight b y 26 . . . Ci::Jg 6 27 g 3 etc.
Variation 9 d2
This continuation adds t o the development o f the white pieces, retains the
' status quo ' and doesn' t make premature commitments regarding whether
the knight will go to el or perhaps to g5.
Game 3 0
Gligoric - Quinteros
Linares 1 981
1 d4 lbf6 2 c4 g6 3 lbc3 J.g7 4 e4 d6 5 ltJO 0-0 6 J.e2 eS 7 0-0
lbc6 8 dS lbe7 9 J.d2
a
9 ... lbd7
1
a
4
3
1 1 6 Variation 9 i.d2
premature at a moment when White was better developed, his maj or pieces
connected, and when he was not weakened on the long diagonal (as after 9
b4).
On 9 ...tLlh5 White can also _play 1 0 l:tcl f5 1 1 tLlg5 (or I I exf5 tLlxf5 I 2
tLle4 tLlf4 1 3 l:.e I tLlxe2+ I 4 'ifxe2 b6 I 5 b4 h6 I6 i.c3 i.d7 I 7 'ii'd3 g5 I 8
tLlfd2 g4 I 9 tLlfl 'ike7 20 c5 'ii' f7 2 I cxd6 cxd6 22 b5 tLld4 23 i.xd4 exd4
with good counterplay, Taimanov-Spassky, Moscow I 973) 1 1 ... tLlf4 1 2
3
2
4
3
8
7
6
4
Variation 9 i.. d2 1 1 7
h
8
7
5
3
2
8
7
6
5
4
3
1f:--l--J'.o;!,ll
a
1 1 8 Variation 9 J.d2
1 0 b4
This seemed to me more appropriate for a qu eenside advance than I 0
llc l .
1 0 ... h6
The opponents fight for the initiative on opposite sides and this prophy
lactic move means a loss of a tempo in the race, but Black did not like
I O . . f5 because of I l lbg5 .
.
1 1 'ii'b3 h8
Another, perhaps unnecessary, loss of time on taking precautions, now on
the b3 -g8 diagonal, and another 'point' for White in the battle for the
initiative.
1 2 ltac l
Sounder was to play 1 2 lbe l first in order to strengthen the pawn chain as
quickly as possible.
12 f5 13 lLle1 tiJf6
.
14 0 f4 1 5 c5 g5 16 lLlb5 lLle8
Black has to defend the c7 point from the forays of the white knight, but
this is after taking ' one step forward-two steps back' .
1
a
Variation 9 i..d2 1 1 9
Compared to the continuations 9 ltle l and 9 ltld2, this i s a great gain for
because he has managed to include even the king's rook in exerting
pressure on the queenside. Black is now on the defensive.
White
23 ..td7 24 a4
.
2
a
Also possible was 2 6 . . J::tg 7. With his next move, White, at the right
moment, parries the threats along the h-file, dealing with the foray of the
queen to h4.
27 .tel ! l:r.b8
Black has been stopped on the kingside and now he tries to prevent the
threat 28 cxd6 cxd6 29 aS and 30 ltlb6.
28 b5 axb5 29 axb5 b6
He mustn 't allow 30 b6 cxb6 3 1 cxd6 ! .
30 c6!
Black's position would be satisfactory after 3 0 cxb6 cxb6.
30 .-tcs 31 'ii a 3 !
..
The double threat of 32 'iVa7 and 32 ltlxeS spells the beginning of the
end.
3 1 . l:tg7
.
To meet 32 'iVa7 with 32 . i.. xg4, but he cannot parry the other threat as
well.
. .
Variation 9 g5
Unlike the previous variation, here the white bishop acts much more
aggressively on the 9'h move. Because of the prejudice that the bishop will
have nothing to do on a diagonal crowded with black pieces, it was only
later on that the continuation attracted the attention of White players. The
bishop at g5 nevertheless hinders Black in his attempt to carry out
efficiently the thematic breakthrough . . . f7-f5 . According to the available
data, some 1 50 tournament games were played in which White players were
quite successful, so the variation has been sporadically accepted in grand
master practice as well.
Game 3 1
Yermolinsky - Topalov
Wijk aan Zee 1 999
1 f3 f6 2 c4 g6 3 c3 j_g7 4 e4 d6 5 d4 0-0 6 j_e2 e5 7 0-0
c6 8 d5 e7 9 i.g5
a
h
8
8
7
2
1
a
In
9 h5
...
1 22 Variation 9 .i.g5
a
1 0 t'Llel
More accurate than 1 0 g3 f6 (if 10 ... f5 1 1 exf5 .i.xf5 12 li)d2 i.. h 3 1 3
.U.e 1 li)f6 1 4 li)de4 with the better position, Legky-Milos, Aosta 1 990;
White also held the advantage after 10 ... h6 1 1 .i.d2 .i.h3 1 2 Ae 1 f5 13 exf5
li)xf5 14 li)e4 'Wd7 1 5 .i.c3 li)f6 1 6 li)fd2 li)xe4 1 7 li)xe4 N.Nikcevic
Brustman, Koszalin 1 998, or 1 3 li:)h4 li)f6 14 exf5 gxf5 1 5 'ii'c 1 f4 1 6 gxf4,
Legky-Mestel, Groningen 1 989) 1 1 .i.d2 fS
a
b
c
d e f g h
1 2 exfS li)xfS (see diagram) 13 li)e4 (Black
8
8
has counterplay after both 1 3 li)h4 ! ? .i.h3 1 4
7
7
li)g2 lt:)f5 1 5 .i.d3 li)d4 1 6 f3 li)f6 1 7 li)e4
6
li)xe4 1 8 .i.xe4 .i.f5 1 9 .i.xf5 Axf5, Timo- 6
shenko-Vitolinsh, Podolsk 1 990, and 1 3 i.. d 3 5
4
li)f6 1 4 li)g5 li)d4 1 5 f3 c6, Eingom-Kuzmin, 4
3
Moscow 1 99 1 ) 13 .i.h6 (or 1 3 . . . li)f6 14 i.. d3 3
2
.id7 [also possible is 1 4 . . . .i.h6 1 5 l:lc 1 .i.xd2 2
'h- 'h, D.Rajkovic- Vukic, Cetinje 1 992] 1 5 1
3 l:lb8 1 6 a4 b6 1 7 'ii'c 2 a5 1 8 g2 h6 1 9
a
b
c
f g h
d c
li)xf6+ 'ii'x f6 20 .i.e4 g5 2 1 h3 .U.f7 22 1i'd3
.U.bf8 with a strong position, Eingom-Smirin, Odessa 1 989) 14 li)fgS li)f6
1 S li)xf6+ 'ii'xf6 1 6 tt:le4 'ii' g7 17 .i.c3 'We7 18 .U.e1 .td7 19 i.. d 3 Aae8 with
chances for both sides, Ftacnik-Kr.Georgiev, Palma de Mallorca 1 989.
...
1 2 'ii'x e2 h6 13 .i.d2
8
7
1 3 ... f5
'ifxf8 20 lLlb5 (An improvement on 20 l:lfl Ji.h3 ! ! [20 . . . 'ii'd 8?! 21 hxg3 ? !
i.h3 2 2 l:lf6 ii'd7 23 e 5 ii.xe5 2 4 l:lxh6 ii' f5 25 ii'e4 ii.xc3 0- 1 Stanec
Nataf, Fuerstenfeld 200 1 ; 2 1 lLlb5 ! ] 2 1 l:lxf8+ l:lxf8 22 lLlxg3 l:lt2 23 'ii'xt2
i.xt2 24 lLld 1 Ji.d4 with a slightly preferable position for Black, Prakash
Konguvel, Nagpur 1 999) 20 . . . ii't2 2 1 ii'xt2 ii.xt2 22 lLlxc7 ! (Better than 22
hxg3 Ji.b6 23 c3 a6 24 lLld4 r/;f] 25 r/;g2 r/;g6, Prakash-Konguvel, Chen
nai 2000) 22 . . . Ji.h3 23 hxg3 l:lc8 24 r/;h2 ! l:lxc7 25 r/;xh3 l:lxc4 26 Ji.c3
i.d4 with an equal game, Lukacs-Rajlich, Budapest 200 1 .
18 g3
3
2
a
1 8 ... fxg3 ?
Correct was 1 8 . . . lLlg6 ! 1 9 gxf4 (if 1 9 'li'h5 'iie 8 20 lLld5 'ifxe6 2 1 l:.ae 1
Black) 20 . . . gxf4 2 1 r/;h 1 (2 1 l:lt2) 2 l . . .'iif6 and Black was better, BleesKlarenbeek, Irakl io 1 993 .
2 1 'ifh5
After this move Black faces an unpleasant choice.
2 l . cxd5
..
for Black.
3
2
a
27 :ae t ! dxc4
So that on 28 %tf2 he could try 28 . . . c3, but. . .
One of Black' s first reactions was to exploit the fact that the e2-h5 diag
onal is presently covered and to carry out the active manoeuvre 9 .!Li h5 .
White players considered it necessary to prevent . . . .!Lih5-f4, so in the first
decades of this practice they played 10 g3 (for 1 0 .!Lid2 see the game
Petrosian-Giigoric, given below) 10 ... f5 1 1 ll:)g5 .!Lif6 1 2 f3 f4, with the idea
of exploiting the weakening of the white kingside and the far advanced
k n ight on g5 .
..
2
a
13
...
l J SSR
fxg3 14 hxg3 h6 (14 ... .!Lih5 15 'ii'e 1 .!Lif4 16 gxf4 exf4 Stein-Gufeld,
1 967) 1 5 ltle6 i.xe6 1 6 dxe6 d5 Taimanov-Bilek, Leningrad 1 957.
2
1
a
I n this position, Black players searched for their best chances in several
different continuations, as for instance 10 a5, which was also tried a move
earlier, or to accept the challenge with 10 .!i:Jf4 1 1 .i.fl , or to undermine
t he white centre at once by 10 f5 when after 1 1 .!i:Jg5 there were two different conceptions: 1 1 .!i:Jf4 or 1 1 .!i:Jf6.
.
...
...
...
...
There follow examples from characteristic games that illustrate all these
options for both sides.
Game 3 2
Petrosian - Gligoric
Rovinj/Zagreb 1 9 70
8 e7 9 b4 h5
...
2
1
a
1 0 d2 !
The exclamation mark represents the surprise with which I was
confronted. The ex-world champion once admitted to me that he had a
notebook full of his new ideas in various openings-which he started using
only after he had lost his world championship title. Up to the present
encounter he had kept this idea secret.
With the text move White saves the time necessary to increase the
pressure on the queenside, whereby his king ' s knight can have an important
role, and Black is no longer able to block that side as after 9 .!i:Jd2 c5.
..
1 0 lDf4
.
The knight is strong here, but it can 't stay on this square for ever, and
Petrosian's idea is based on the assessment that Black has spent two tempi
on this manoeuvre and that the knight stands in the way of the black pawn
mass on the kingside. He applied this idea very successfully in the Siimisch
variation against Hort in Palma de Mallorca in 1 969 as well.
1 1 a4
The bishop can't run away to f3 at once: 1 1 i.f3 lt:ld3 12 i.. a 3 a5 so the
other white bishop doesn't have a good place to hide.
l l ... f5
At this moment I felt I was in grave danger of being outplayed via the
queenside, so all my moves were motivated by my hurry to carry out a
counter-reaction that would neutralize White 's initiative. My first idea was
to reduce White's menacing pressure after 1 2 c5 by 1 2 . . . fxe4 and 1 3 . . . lt:lf5.
However, perhaps the simple 1 1 . . .lt:\xe2+ was also playable, clearing the
way for the black pawns.
12 i.t3
Up to here Black has been fighting in the dark, not knowing exactly the
essence of White' s plan, and this move made by his opponent was a small
psychological shock for him that lasted some five minutes. Should he have
taken this bishop earlier? Because now it is too late for 12 . . . lbd3 13 i.a3
and the other bishop is active and safely hidden behind his a-pawn, which
was the idea of White's 1 1 'h move.
1 2 g5 !
.
After the initial surprise, Black spent twenty minutes searching for the
best solution in this critical moment of the battle. The move played is
probably the only sound solution. Black weakens the light squares, but
speeds up his action at the kingside, which, in this dramatic position, is
important for maintaining the balance.
Here 1 2 . . . fxe4, intending t o reduce the pressure o f the white pawn mass,
was much slower. After 1 3 lDdxe4 lDf5 14 g3 the other black knight is in a
very unpleasant position. After the text move, because of the threat l 3 . . . g4,
Black reaches a similar position two tempi earlier.
13 exf5 lbxf5
Not 1 3 . . . il.xf5 because the primary task is to introduce the passive knight
into the game.
14 g3
White doesn't have time for 1 4 lDde4 because of the threat 14 . . . lDh4.
a
1
a
14 ... lt:Jd4 !
At the time this seemed to me to be the only good reaction. This is no
wild piece sacrifice but rather a positionally active continuation which,
together with the material investment, should solve the problem of
maintaining the balance. This is so because after 14 . . . lDg6 the black pieces
would be pushed back and White would not only have a spatial advantage
but would also be superior on the light squares.
Only later did Soviet grandmasters find the
continuation 14 lDh3+ 1 5 g2 1i'd7 ! . (see
diagram) This not so obvious move solves the
problem without sacrificing a piece, because it
prevents 1 6 il.g4. For instance: 1 6 lDb3 (it is
not good to play either 1 6 Jl.g4 lDxt2 ! , or 1 6
lDde4? lDd4 1 7 i.. h 5 lDf4+ 1 8 gxf4 1i'h3+ 1 9
'it>h 1 g4 with a decisive attack) 1 6 ... lDd4 17
lt:Jxd4 exd4 1 8 lDb5 c6 1 9 lDa3 :xf3 20 1i'xf3
g4 21 'Wb3 1i'e7 22 :a2 i.. f5 23 f3 d3 24 fxg4
'ife4+ 25 :n lDg1 (stronger than 25 . . . i.xg4
..
6
5
3
2
8
7
7
6
26 'ii'xd3) 26 'ii'x d3 'ii'x d3 27 .!:txd3 i.xd3 28 xg l cxd5 29 cxd5 l:te8 h1h, Keene-Kavalek, Teesside 1 97 5 .
1 5 gxf4
Accepting the challenge because Black would feel pretty good after 1 5
i.g4 i.xg4 1 6 'ii'xg4 h5 1 7 'ii'd l .!Llh3+ 1 8 'itt g2 g4 1 9 f3 'ii'd 7.
1 5 xf3+
...
Black could also have played 1 5 . . . exf4, but, considering that he chose to
take ultimate measures by sacrificing material, he prefers to speedily
remove one of the pieces that is protecting the white king.
16 'i'xf3
A more cautious option was 16 lt'lxf3 exf4 1 7 i.b2 g4 1 8 h I ! , intending
to give back the piece in order to balance the position on the endangered
kings ide.
1 6 ... g4 !
2
a
1 7 'i'h 1
I neither anticipated nor expected such a passive move. However, the
manoeuvre is typical of Petrosian 's original style. In our encounter in
Zagreb in 1 965, that is five years earlier, having the king at h i and three
pawns on the second rank in front of it, Petrosian had surprised me by
moving his queen as far as g l , in order to protect the sensitive h2 square
and gain a free hand at the queenside, where the outcome of the game was
decided in his favour.
But, here it was probably better to play 1 7 'ii'd 3 i.f5 1 8 .!Llde4 exf4 1 9
i.. xf4, giving back the piece in order to have counterplay after 1 9 . . . i.. xe4 20
'ikxe4 i.xc3 2 1 .!:ta3 with unclear chances.
With the text move White wants to protect his king from mating threats,
but the queen remains quite out of the game.
1 9 llfel f3
There was a threat of activating the queen by 20 'i'g2, so this was the last
chance to firmly tighten the position around the white king and queen.
20 tl:Jde4
20 h4 was also playable, but even then Black's attack remains strong.
20 ... 'it'h4 2 1 h3
White can 't allow 2 1 . . .'i'h3 which would definitely sentence the white
queen to languishing impotently in the comer at h l .
2 l . .. e5!
2
a
22 lle3
Hopeless is 22 hxg4 'ifxg4+ 23 fl i.xc3 etc.
24 'ifh l h2+ 25 g2
On 25 fl . 25 . . . :f3 ! is decisive.
25 .. .'i'h5!
It took Black some time to find this fine manoeuvre which is the most
efficient way of carrying on the attack, with the battle revolving around the
light f3 and h3 squares close to the king. White ' s response is forced because
he has to protect the f3 square.
26 tbd2 i.d4 !
Black attacks the main defensive piece-the rook that protects the third
rank.
27 'i'e l
This response is again forced because White has to free the h 1 square for
his king. If 27 :ae 1 .th3+ 28 :xh3 'ii' g4+ with mate after 29 . . . :xf2+.
a
1
a
27 ...l:lae8 !
Introducing the last reserve into the battle and this is what decides the
game. White would now rather return the piece than face the continuation
28 h 1 :xe3 ! (saving the dark-squared bishop for the final attack) 29 fxe3
.tf3+ 30 lDxf3 'fi'xf3+ 3 1 xh2 .te5+ with imminent mate.
28 tbce4
The other knight has to protect the f3 square.
28 i.xb2 29 J:.g3
.
White does his best to achieve counterplay. This is the very reason why
Black postpones the capture of the exchange.
Game 3 3
Gligoric - Donner
Berlin 1 9 71
1 d4 tbf6 2 c4 g6 3 tbc3 .tg7 4 e4 0-0 5 .te2 d6 6 tbf3 e5 7 0-0
tbc6 8 d5 tbe7 9 b4 a5
On the alternative 9 . . . lL!hS 1 0 l:Ie 1 aS White is clearly better after 1 1 bxaS
l:IxaS 12 lL!d2 lL!f4 1 3 J.. fl cS 14 a4 l:Ia6 1 S l:Ia3 when Black's retreat
I S . . . lL!hS 1 6 lL!bS lL!f6 1 7 g3 lL!d7 1 8 i.b2 f5 1 9 f4 resulted in an initiative
for White in Xu Jun-Magai, Olympiad, Istanbul 2000. The reader will see
more on this line (9 . . . lL!hS 1 0 l:Ie 1 aS) later.
a
1
a
10 bxa5
This is simpler than 1 0 i.a3 b6 ! 1 1 bxaS lL!hS ! 1 2 l:Ie l f5, which is the
most recent solution to obtaining successful counterplay for Black-for
more about this see the next example.
1 0 .. Jba5
Black's queenside defence would be tougher after the blockading 10 . . . cS.
19 tbc4 tbe8
White is happy with the early opening of the position on the queen 's flank
because now he develops an initiative there even more quickly and, as
il lustrated by this forced backward step, he also hampers Black's kingside
counterattack.
2
a
Without queens, Black can't attack the white king-and White has a
superior endgame.
Gligoric - Simeonidis
Zonal, Panormo 1 998
1 d4 CiJf6 2 c4 g6 3 CiJc3 it.g7 4 e4 d6 5 CiJf3 0-0 6 i.. e 2 e5 7 0-0
CiJc6 8 d5 CiJe7 9 b4 a5 10 bxa5
a
7
6
3
2
a
"
8
7
5
4
...
d e r 8 11
lBd2 l:tf6!-by covering the d6 square Black
8
8
prepares
the key counterstroke . . . c7-c5- 1 7
7
7
i.. f3 ! ? c5 ! 1 8 axb6?!-through necessity
6
6
White opts for an unclear piece sacrifice5
s
1 8 . . . cxb4 1 9 l:txa6 i.. x a6 20 lBc7 i..b 7 2 1
4
4
'ii'b 3 lBc8 22 exf5 gxf5 23 c5 dxc5 24 d6+
3
Cit>h8 25 i.. xb7 lBxb6 26 lBb5 'ii'd 7 27 i.. a 6
2
2
l:tg6 28 g3 'iWc6 29 'iWf3 lBbd5 3 0 h 1 e4 3 1
lBxe4 fxe4 32 'ii'xe4 l:te6 3 3 'ii'f3 'Wxa6 34
a
h
c
d e f g h
d7 i.. f6 3 5 l:td 1 'ii'x b5 36 gxf4 l:td6 and
White resigned, Ivanisevic-Antic, Novi Sad
2000) 14 ... lBf4 15 a4 (it is also possible to play 1 5 l:te 1 lBxe2+ 1 6 'ii'x e2 f5
1 7 a4 fxe4 1 8 lBxe4 lBf5 1 9 'ikd 1 llf7 20 a5 bxa5 2 1 l:txa5 h6 22 'iWa 1 l:txa5
23 'ii'x a5 i.. f8 24 'ii'a 8 lBd4 25 lBa7 with the
a
b
c
d
e
f g h
initiative, Van Wely-Topalov, Rapidplay,
8
8
Monaco 2000) 15 ... f5 1 6 l:te1 !i:Jxe2+ 1 7
7
7
'Wxe2 fxe4 18 lBxe4 lBf5 1 9 'ii'd3 lBd4 20
6
6
i.. c3 lBb3 2 1 l:ta3 lBa5 22 i.. d 2 i.. f5 23 'ii'e 2
5
5
!i:Jb7 24 i.. g 5 'Wd7 25 f3 with a slight
4
4
advantage for White, Van Wely-Smirin,
3
Olympiad, Elista 1 998)
2
typical plan with c4-c5. This i s stronger than I 3 lLld2 lLlf6 I 4 .i.fl .l:txa5 I 5
.i.b2 .i.d7 I 6 a4 .i.h6 I 7 f3 fxe4 I 8 tLldxe4 lLlf5 I 9 tLlxf6+ 'i'xf6 2 0 tLle4
'ii' h4 2 I .i.c3 .l:taa8 22 a5 bxa5 23 .l:txa5 .i.e3+ 24 h i .i.f4 25 h3 tLlg3+ 26
l.t>g i tLlxe4 27 .l:txe4 'ii'g 3 28 'iVa i .l:txa5 29 'ii'x a5 .i.f5 30 .l:te2 'i'li2+ 3 I
Wf2 .i.g3+ 3 2 e3 'ii'g l + White resigned, Tukmakov-M.Pavlovic, Crans
Montana I 999) 13 bxa5 14 ..ta3 lll f6 ( I 4 . . . tLlf4 is a quite playable
fashionable alternative. For example I 5 .i.fl fxe4 I6 tLld2 [ I 6 tLlxe4 .i.g4
1 7 lLled2 e4 I 8 .l:txe4 .i.xf3 I 9 ll:lxf3 .i.xa I 20 'i'xa I lLlh3+ 2 I gxh3 .l:txf3
22 1Lb2 with compensation, Danailov-Nijboer, Wijk aan Zee 2000]
1 6 . . . tLld3 I 7 .i.xd3 exd3 I 8 tLlde4 lLlf5 I 9 'ifxd3 ..th6 20 .i.e I .i.xc I 2 I
.l:taxc I a4 22 .l:tb I tLld4 2 3 .l:tb4 .i.f5 with chances for both sides, Delchev
Smirin, Pula 200 I ) 15 .i. d3 fxe4 ! ? 16 tLlxe4 ..tg4 17 lLlxf6+ .l:txf6 18 .i.e4
lll f5 19 'i'd3 .l:tti 20 .l:tab1 ..th6 21 c5 'iff6 22 cxd6 cxd6 23 .l:tb6 .l:td8 24
l:tfl 'ii'e7 25 h3 ..txf3 26 ..txf3 ..tf8 27 g3 tLld4 28 ..te4 'ifc7 29 .l:ta6 .l:tc8
30 g2 'il'd8 31 .i.b2 .l:tc5 32 ..txd4 exd4 33 'il'xd4 with an almost decisive
advantage although Black somehow managed to draw a pawn down in a
long endgame, Kramnik- Smirin, Belgrade I 999.
...
10 c5!
...
2
1
1 1 a4
2
a
...
i.xe4 fxe4 24 lbxe5 i.xe5 25 'i'xe4 i.f5 26 'fi'e2 'fi'h4 27 g3 'i'g4 28 i.xh6
'Wxe2 29 l:.xe2 l:.fa8 30 i.f4 l:.xa4 3 1 l:.xa4 l:.xa4 32 i.xe5 dxe5 33 l:.xe5
i.h3 34 lbd6 b6 35 ..ti>t2 Black resigned, Yermolinsky-Fedorowicz, New
York 2000) 13 lbb5 l:.a6 14 l:.a3 f5 15 i.b2 i.h6 16 exf5 lbxf5 ! ? with
original manoeuvres: 1 7 lbe4 lbf6 18 lbxf6+ l:.xf6 19 i.g4 ! lbd4 20 i.xc8
'i'xc8 21 g3 ! {The point of the attack is the f4 square ! ) 2 1 . .. l:.ti 22 '1fi>g2
'ii'f5 23 i.xd4 ! cxd4 24 'ii'e 2 'ili'c8 25 f4 ! i.g7 26 l:taf3 exf4 27 l:.xf4 l:.xf4
28 gxf4 ! (not giving the e5 square to Black) 28 ... l:.xa4 29 lbxd6 'ii'd 7 30
f5 ! gxf5 31 lbxf5 l:.a8 32 'i'g4 ..ti>h8 33 'ili'xg7+ 'ili'xg7 34 lbxg7 xg7 35
l:r.f4 ! d3 36 f3 ! l:r.c8 37 '1fi>e3 b5 38 cxb5 l:r.c5 39 ..ti>xd3 l:r.xb5 40 '1fi>c4 l:4b2
41 d6! l:r.c2 42 ..ti>d5 Black resigned.
8
"
6
5
6
s
ll .. .lba5 1 2 .tla3 h8
Or 12 . . . lbe8 13 lbb5 l:r.a6 14 g3 ! f5 1 5 exf5 gxf5 1 6 lbh4 lbg6 1 7 lbg2
lbf6 1 8 f4 lbe4 1 9 i.d3 exf4 20 ltJxf4 'ili'e7 2 1 'We2 i.d7 22 lbe6 i.xe6 23
dxe6 l:r.aa8 24 g4 lbe5 25 gxf5 lbxd3 26 l:r.xd3 'ii'h4 27 i.f4 lbg5 28 i.xd6
lbh3+ 29 '1ti>g2 'ii'xc4 30 i.xf8 l:r.xa4 3 1 f6 i.h6 Black resigned, Antic
Radovanovic, Leskovac 2002 .
13 tt'lel !
h
8
8
7
6
2
a
This time the main point of White ' s attack is the e5 square.
Game 3 5
Bacrot - Motylev
Linares 2001
1 d4 f6 2 c4 g6 3 c3 .tg7 4 e4 d6 S f3 0-0 6 .i.e2 eS 7 0-0
CDc6 8 dS e7 9 b4 hS 1 0 liel fS l l lLlgS f6 1 2 f3
a
1
a
1 2 . . . c6
8
4
3
-!!!i!!-..a
f 8 h
8
7
6
5
6
5
3
2
.._f_
g !...h...a
f
6
s
13 hl
13 i.e3 is an alternative, then:
l3 ... h6 (see diagram) 1 4 lbe6 i.xe6 1 5
dxe6 'fkc7 1 6 .l::r.b 1 .l::r. fd8 1 7 b5 'fkc8 1 8 bxc6
bxc6 1 9 'fka4 h7 20 .l::r. ed 1 'ii'x e6 2 1 .l::r.b 7 .l::r.d7
22 'fka6 lbe8 with equal chances,
Komljenovic-A.Kuzmin, Villa de Benasque
1 999,
l3 . lbe8 (see diagram) 1 4 c5 cxd5 1 5 exd5
e4 1 6 cxd6 lbxd6 1 7 i.d4 h6 1 8 i.xg7 'ili>xg7
1 9 'fkd4+ 'it>g8 20 lbe6 i.xe6 2 1 dxe6 'iib 6 22
.l::r. ad 1 .l::r. fd8 23 'fkxb6 axb6 24 fxe4 lbxe4 25
lbxe4 fxe4 26 .l::r.d 7 .l::r. x d7 27 exd7 .l::r.d 8 28
i.. c4+ <l;g7 29 .i.e6 c.tf6 3 0 l:lxe4 with the
superior endgame, Timoshchenko-Motylev,
Ohrid 200 1 .
..
8
7
3
1
a
a
c
c
d
d
h
h
8
7
3
1
.-..i-.._,..
7
6
6
5
3
14 h4 cxd5 1 5 cxd5 (see diagram) 1 5 f4
2 ( 1 5 . . . h5 1 6 %te l g3 1 7 'ii'd2 .id7 1 8 .ib5
.......:o-...a 1 .ixb5 1 9 xb5 fxe4 20 fxe4 'ii'd 7 2 1 c7
a
b c
d c r g h
.l:tac8 22 ce6 was in White' s favour,
Savchenko-Volokitin, Cappelle Ia Grande
200 1 ; or 1 5 . . . .id7 1 6 .l:tc 1 a6 1 7 b5 f4 1 8 .if2
.ixg5 1 9 hxg5 h5 20 'ifb3 a5 2 1 b6 c8 22 .ib5 .l:tf7 23 .ixd7 .l:txd7 24
b5 'ii'x g5 25 c7 llb8 26 e6 i/Je7 27 .l:tc7 g3 28 .ixg3 fxg3 29 .l:tec 1
_.h4 3 0 .l:txc8+ %txc8 3 1 .l:txc8+ f7 32 i/Je3 Black resigned, Ivanov
Rudolf, Panormo 200 1 ) 16 .if2 .ixg5 17 hxg5 h5 18 l:c1 (if 1 8 .id3
.id7 19 'Wd2 c8 20 b5 a6 2 1 a3 'ifxg5 with sufficient counter
chances, Orso-Berczes, Hungary 2000) 18 g3 19 'ifd2 .id7 20 .ib5
.ic8 21 e2 .id7 22 c3 .ic8 23 .l:tc2 a6 24 .id3 .id7 2 5 e2 xe2+ 26
.ixe2 .ia4 27 l:r.c3 'Wd7 28 l:r.ecl .l:tac8 29 .in
8 l:r.xc3 30 l:r.xc3 .l:tc8 3 1 g3 fxg3 32 .ixg3 with
8
the better chances, Radjabov-Nijboer, Wijk an
7
7
Zee 200 1 .
6
6
3
..
..
5
4
3
---...a
7
6
3
2
l 3 . . . h8
The alternative is 13 ... h6 14 li:Je6 i.xe6 15 dxe6 ll:Je8 (see diagram) 1 6
(or 1 6 'iVb3 ll:Jc7 1 7 i.e3 'it?h7 1 8 c 5 d 5 1 9
"
b
c
d
e
f g h
xd5 cxd5 20 l:tad 1 d4 2 1 li:Jb5 li:Jcd5 22 i.g 1 8
8
aS 23 a3 axb4 24 axb4 li:Jc6 25 i.c4 li:Jdxb4 26
7
7
l/)xd4 exd4 27 e7 ll:Jxe7 28 "iixb4 "iic 7 29 l:tb 1 6
6
I1ab8 30 'iVb6 l:tfc8 and Black maintained an 5
5
a pproximate balance, Degraeve-Howell, Gron
4
i n gen 1 995) 16 ... c5 17 li:Jd5 ll:Jc7 18 ll:Jxc7 4
3
jkxc7 1 9 exf5 gxf5 20 g4 'it?h8 2 1 l:tg1 l:.f6 22 3
2
2
gxf5 li:Jxf5 23 i.d3 "iie 7 24 "iie 2 e4 25 i.xe4
J:lxe6 26 l:tb1 l:tf8 27 "iid 3 li:Jd4 28 l:tb2 with 1
b
c
d
e
f g h
better chances for White, Lima-Romero, Leon
1 996.
hS
After 2 5 . . . ll:Jxe6 2 6 "iixe4 the pawn at e 5 will fall. Now Black places vain
hopes in the future of the pawn at f2.
26 'ii e4 ltf2 27 l:r.xf2 exf2 28 "ikxeS J:te8 29 lDc7 ! .:Us 30 l:.O lDe7
.H xeS lDg6 32 "ie3 "fie7 33 ltJdS 'ii' x e6 34 l:.xf2 "ikxe3 3S l:xf8+
0:'lxf8 36 lDxe3 lDge6 37 g2 lDd7 38 i.f3 lDecS 39 ltJf5 Black
resigned.
Game 3 6
Gligoric - Nataf
Cannes 1 998
9 ti)h5
..
1 2 .to c 6
8
7
3
2
a
3
2
a
Discovered by Kramnik.
Unclear is l3 'iib3 ! ? (see diagram) l 3 ... h6
t 4 l2Je6 i.xe6 1 5 dxe6 fxe4 1 6 tll x e4 l2Jf5 1 7
1t.e3 lbh4 1 8 l2Jxf6+ 'iixf6 1 9 i.e4 'iix e6 20
bS 'ii g4 2 1 i.d2 l:tac8 22 l:tab1 l2Jf5 23 'ii d 3
Wh8 24 bxc6 bxc6 25 h3 'ii' h4 26 l:tb7 d5!
27 cxd5 lbd6 28 l:tb2 l2Jxe4 29 l:txe4 'iix f2+
30 'ifi>h2 cxd5 and in mutual time pressure
Black succeeded in mating his opponent after
3 1 'iix d5 l:tcd8 32 'ii' a 5 Ad3 33 'ii'b 4 'iig 3+
34 'it>g1 l:tf2, D.Antic-P.Popovic, Novi Sad
2000.
g
8
h
a
3
2
3
2
1
8
6
6
s
1 3 .te3 !
8
7
6
s
3
2
1 9 .ta3 .l:tc3 20 xc3 ti:)fxd5 and Black wins] 1 9 . . . xd6 20 .ta3 'fic7 with
better chances for Black) 18 ... ti:)fxd5 19 exd5 e4 20 d6 .l:tc3 ! 2 1 .ta3 ! .l:txb3
22 dxe7 ! .!:txa3 23 exd8= .!:txd8 24 l:tadl .tf6 25 .te2 r,t>rs 26 .!:txd8+
.txd8 27 .!:tdl .tg5 28 .tc4 .l:tc3 29 .tb3 .!:tel 30 .!:txcl .txcl 31 r,t>n th-th,
Kir.Georgiev-Ponomariov, Olympiad, Istanbul 2000.
The bishop on e3 is most active and
provokes
. . . f5 -f4, which increases White ' s
7
control of the light squares. The previously
6
6
attempted 13 .tb2 (see diagram) is a less
5
Eromising alternative-1 3 . . h6 (or 1 3 . . . a6 1 4
4
tt:Je6 i.xe6 1 5 dxe6 fxe4 1 6 ti:)xe4 ti:)xe4 1 7
.1
3
i.xe4 d5 1 8 cxd5 cxd5 1 9 .tc2 th- th,
2
Kasimdzhanov-Nataf, Linares 1 998) 14 ti:)e6
.txe6 1 5 dxe6 fxe4 16 ..ixe4 (if 16 ti:)xe4
"
b
c
f g h
d
c
ti:)xe4 1 7 .!:txe4 ti:)f5 1 8 b5 .l:tc8 19 .l:te2 .l:te8
20 bxc6 bxc6 2 1 c5 d5 22 i.xe5 i.xe5 23
"
b
c
d
e
f g h
8
8
.!:txe5 'ii' f6 24 .!:te l
t/2- th,
Dautov7
7
Kindermann, Nussloch 1 996) (see diagram)
1 6 ... ti:)xe4 (interesting is 1 6 . . . 'ii'b 6 [if 1 6 . . . d5
6
6
1 7 cxd5 cxd5 1 8 i.c2 'itd6 1 9 i.b3 'ifxe6 20
5
ti:)a4 e4 2 1 ti:)c5 'iff5 22 f3 ! r,t>h7 23 fxe4
4
4
ti:)xe4 24 i.xg7 'if2+ 25 r,t>h l r,t>xg7 26 ti:)e6+
3
3
favours White, Tukmakov-Hulak, Tucepi
2
1 996] 1 7 3 d5 1 8 cxd5 cxd5 1 9 ..ixd5
ti:)g4 20 ti:)e4 [20 .l:te2 leads to a draw after
"
b
c
d
c
f g h
20 . . . ti:)xf2 2 1 ti:)a4 ti:)h3+ 22 r,t>h l ti:)2+]
20 . . . :ad8 2 1 W/g3 ti:)x2 22 ti:)x2 .!:txd5 23
.l:te2 'ifxe6 24 .!:tae l ti:)c6 with advantage to Black, Ivanov-Avrukh, Beershe
ba 1 998.) 17 ti:)xe4 ti:)f5 18 iVg4 r,t>h7 (or 1 8 . . . 'ife8 1 9 e7 ! 'ifxe7 20 'ii'x g6
.!:tad8 2 1 .!:tad l with unclear chances, Tukmakov-Delchev, Kastel Stari
1 997) 19 .!:tadl 'fie7 20 .l:td3 .!:tae8 21 1Ih3 d5 22 cxd5 cxd5 23 ti:)c5 e4
with mutual chances, Averkin-Zakharov, Krasnodar 1 99 8 .
8
7
8
7
3
2
1 3 ... h6
8
7
6
5
20 'ii g4
Stronger than 20 .ia4 'ir"d6 2 1 .id7 'ifxb4 22 l:.b l 'ii"h4 23 f3 l:.f5 24 .i.f2
'iff6 25 .ig3 h5 26 h3 l:.f8 27 'it>h I 'ir"g5 28 .ih2 e4 29 fxe4 l:.f2 30 l:.g 1
dxe4 Malakhatko-Jenni, Olympiad, Istanbul 2000.
20 . . ..U.f6 ! ?
8
1>
7
6
..a
8
7
2 4 'ii'a 4!
I played this move 'a tempo', telling myself "This is how Kramnik would
play too" (he is the one who has made the greatest improvements to White ' s
play i n this variation), not knowing of the recently played game Kramnik
Shirov, Linares 1 998, where Kramnik continued with the more cautious 24
'ir'xe6+ .l:.xe6 25 .te l rf;f7 26 .i.a3 and had to be satisfied with a draw in a
somewhat more favourable endgame after returning the material.
24 ...l:.d7 25 f4 !
A critical position.
25 ... e4 ! ?
Huzman recommends a s a better chance 2 5 . . . exf4 2 6 .td4 'ii'd6 2 7 .i.e5
'ir'c5+ 28 .i.d4 1i'd6, and if 27 .i.xf6 .i.xf6 28 'ike4 g7, although after 29
'ir'f3 ! the position remains unclear.
26 i.c1 'iff5
Strange, but Black has no way of defending his pawn because of the
multiple pins along the diagonals and files !
Kramnik - Shirov
Tilburg 1 99 7
1 tb f3 tbf6 2 c 4 g 6 3 tbc3 i.g7 4 e 4 d6 5 d 4 0-0 6 i.e2 e 5 7 0-0
l2Jc6 8 d5 l2Je7 9 b4 tbh5 10 :le1
a
1 0 . . . f5
8
6
5
4
3
3
2
..
..
1
a
Black has opened the diagonal for his dark-squared bishop, but the draw
back of this line is that White will create a pawn on e6, which will not be
easy to eliminate.
1 3 ... .tf6
"
b
c
d c f g h
An attempt to improve B lack' s play. Another
8
idea is 13 fxe4 ! ? 14 lbcxe4 lbf5 (see 8
diagram) 1 5 g4 ! (it would be a loss of time 7
7
t o play 1 5 c5 lbd4 1 6 cxd6 cxd6 1 7 g4 xg4 6
6
1 8 1Wxg4 l:tf5 ! 1 9 lbe6 lbxe6 20 dxe6 1We7 2 1 s
h4 d5 22 lbg5 'i!i'xb4 23 e7 l:te8 24 l:tb 1 'ii' c 3 4
4
25 l:te6 f6 26 l:txb7 xg5 27 l:lxa7 fl 28 3
3
l:taa6 l:lxe7 White resigned, Dumitrache- 2
2
Kupreichik, Olympiad, Erevan 1 996) 1 5 ... lbd4 1
1 6 e6+! lbxe6 17 lbxe6 xe6 18 dxe6 l:tf5
a
b
c
d e f g h
(Timoshenko recommends as better 1 8 . . . e5
1 9 'ii'd 5 'ii' e 7 20 c5 with somewhat better chances for White) 1 9 c5 d5 20
lbd6 ! ! cxd6 2 1 e7 with a decisive initiative, Pelletier-Gormally, Hastings
1 996/97 .
.
6
4
3
2
2
1
.l:.xf5 26 b6! l:e5 ! 27 .l:.c7 'iff6 28 'iti>h2 ! .l:.xe6 29 'ifxf6 .l:.xf6 3 0 .l:.xb7 d5
3 1 'iti>g3 with the better endgame, Veingold-Cevcenko, Olympiad, Elista
1 998) 18 .l:.xe4 e5 1 9 g3 g5 20 h4 h8 2 1 'ii'd 4 fxg3 22 .l:.xg3 gxh4 23
.l:.g2 'iff6 24 'ii' e3 .l:.ae8 25 f4 xg4 26 .l:.xg4 .l:.g8 with equal chances.
3
2
2
1
a
1 9 .U.cxe3 c6
Kramnik mentioned 19 ... a5 20 b5 c6 21 bxc6 bxc6 as an unclear
possibility.
Game 3 8
Kramnik - Kasparov
Novgorod 1 99 7
1 t'i) f3 t'i)f6 2 c4 g 6 3 t'i)c3 j_g7 4 e4 d 6 5 d 4 0-0 6 j_e2 e 5 7 0-0
t'i)c6 8 d5 t'i)e7 9 b4 t'i)hS 1 0 .:tel t'i)f4 1 1 j_fl
a
The black knight is actively placed on f4, but it doesn't have a concrete
target and obstructs the black pawns in their quest to grab space around the
white king-and besides, this outpost will not be secure in the future. From
here on, Black players have chosen several quite different ideas.
l l . . . a5
Black uses this moment to break up the compact white queenside pawn
mass. There are numerous alternatives :
The developing move l l . . i.. g4, after 12 h3 i.. x f3? ! 13 'ii'x f3 f5 14 'ii'd 1 !
fxe4 1 5 ltlxe4 h 6 1 6 l:tb1 ! 'i!Vd7 1 7 'i!Vg4 ! 1i'e8 18 l:tb3 ..t;>b8 1 9 g3 ltlh5 20
'We6 ! , leaves White master of the light squares over the whole board,
Tukmakov-Cvitan, Zurich 1 99 5 .
.
II
,,
3
2
"
'
j,
1 2 bxa5
12 .ia3 i.g4 (or 12 . . . axb4 13 i.xb4 i.. g4 1 4 l:te3 ! b6 1 5 h3 i.d7 ! 16 a4
f5 17 a5 with a slight advantage, Granda Zuniga-Van Wely, Amsterdam
1 996) 1 3 h3 .ix3 14 'ili'x3 axb4 1 5 .ixb4 c5 1 6 dxc6 ltJxc6 1 7 a3 ltJe6 1 8
ltJb5 ltJed4 1 9 'i!i'd3 ltJxb4 20 axb4 6 2 1 ltJxd4 exd4 22 'ii'd2 l:la7 23
l:tec 1 l:tfa8 24 l:lxa7 'h- 1h Kramnik-Gelfand, Dortmund 1 996.
1 2 .l:Ixa5
...
13 lbd2
1 5 ... g5 ! ?
Black wants to save a tempo, but the pawn
sacrifice is not quite correct.
Unclear is 1 5 h6 (see diagram) I 6 li)b5
i.. d 7 (if I 6 . . . g5 I 7 g3 li)fg6 1 8 i.. e 2 f5 I 9 exf5
li)xf5 20 i.. h 5 li)ge7 2 I i.. g4 with a strategically better game, Krasenkov-Fedorov, Vilnius
1 997) 1 7 g3 li)h5 1 8 i.. b2 i.. e 8 1 9 i.. e 2 li)f6
20 f4 li)d7 2 1 'ii'a 1 f5 25 exf5 li)xf5 Topalov-Nijboer, Wijk aan Zee 1 99 8 .
...
8
7
1
3
2
3
2
a
II
"
6
5
6
5
2
a
16 g3 ll'1h3+
If 16 . . . lL!fg6, then after 1 7 'ifh5 White is
clearly better.
3
2
26 ... i.xf3?
Played because of the threat 27 lL!g5+ but 26 . . . 'iii' h 8 ! 27 i.b2 was better.
Index of Variations
1
0-0
9 ll:le I
d5 ltle7
One
ltld7
10 ll:ld3 f5
..............................
I I f3 . . . . . . . . . . . . . . . . . . . . . . . . .
1 1 exf5 ll:lxf5 ( l l . ..gxf5) . . . . . . . .
12 f3
. .
1 2 ltle4 . .
l l .i.d2 . . . . . . . . . . . . . . . . . . . . . . .
.....................
l l . . . ll:lf6
.....................
l l . . . fxe4
10 .i.e3 f5 1 1 f3 . . . . . . . . . . . . . . . . . . . . . . . . .
l l . . . f4 1 2 .i. f2 g5
1 3 ltld3 ltlf6 1 4 c5 ll:lg6
............
............
............
............
............
............
............
............
............
13
13
25
25
29
31
31
34
17
.................
1 6 "ifb3 . . . . . . . . .
1 6 l:tc2 . . . . . . . . . .
13 b4 ll:lf6 14 c5 ll:lg6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
15 cxd6 . . . . . . . . . . . . . . . . . . . . . . .
1 5 a4 ! . . . . . . . . . . . . . . . . . . . . . . . .
13 a4 ll:lg6
1 4 ltld3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
14 a5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
13 l:tc I ll:lg6 14 c5 ltlxc5 I 5 b4 ll:la6 16 ll:ld3 . . . . . . . . . .
1 6. . .h5 . . . . . . . . . .
1 6 . . .l:tf7 . . . . . . . . .
1 1 . . .ltlf6 1 2 cS f4 13 .i.f2 gS 1 4 a4 ll:lg6 IS aS hS 16 cxd6 cxd6
1 7 ll:lb5 g4 1 8 .i.xa7 . .
17
17
20
35
35
37
Part
9 ll:le l (9 h i ) ltlc8
1 0 f4 . . . .
1 0 lt:\ d 3 f5
...
....
..........
.
..........
. . . . . . . . . .
..........
..........
..........
..........
. . . . . . . . . .
. . . . . . . . .
.......
.......
.......
.
. . . . .
.......
.......
.......
.......
......
.
15 l:tc l l:tf7
. . . . .
. . . . . . . .
.
.. ... .. ..
l l . . .ltJ f6 . . . . . .
( l l ...'i'h8) . . . . .
l l . . .fxe4 . . . . . .
.. ... .. . ...
l l . . .c5
......
i l . .'h 8 . . . . . .
l l . . . f4 12 i.f2 h5
. .
. .
.
. . .
I O .i.e3 f5 I I f3
. . .
42
Two
.
. .
. . .
. . .
..
. .
. . .
. . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I I f4 .
i l il.d2
39
40
48
48
50
. . . . . . . . . . . . . . .
. .
. . . . . . . . .
...... ...... ....... .. .
............................
.. ........................ ..
. . .
. . . . . . . . . . . . . . . . . . .
. .
. . . .
. . . . . . . . . . . . . . . . .
. . .
.
.
.. ......................
........ ....................
1 3 cS gS 14 a4 . . . . . . . . . . .
14 ... ltlg6 . . . . . . . . . .
1 4. . .'i>h8 . . . . . .
. .
1 4 . . . ll:lf6 . . . . . . . .
1 4 . . . dxc5 . . . . . . . . .
1 4 . . . l:tf6 . . . . . . . . .
.
55
55
. . . 58
. . . . . . 62
. . . . . . . 62
. . . . . . . 63
. . . . . . .
67
. . . . . . 71
. . . . .
.
71
. . . . . . . 74
. . .
. . .
79
.
. . . . . . 81
. ..
81
.
82
. . . . . . 85
. . . . . .
87
1 60 Index of Variations
Part Three
9 ll'le l ll'ld7
I 0 f3 f5 I
9
9
9
g4 'it>h8
.. ....... .. .........
1 2 ll'lg2 . . . . . . . . . . . . .
1 2 .te3 . . . . . . . . . . . . . .
1 0 ll'ld3 f5 I I .td2 ll'lf6 1 2 f3 f4 1 3 g4 . . . .
ll'ld2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9 . . . .th6 . . . . . . . . . . . . . . . . . . . . . . . . . .
9 . . . ll'ld7 . . . . . . . . . . . . . . . . . . . . . . . . . .
9 ... c5 . . . . . . . . . . . . . . . . . . . . . . . . . . : .
9 . . . a5 . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 0 a3 . . . . . . . . . . . . . . . . . . .
I O . . . ll'ld7 ( I O . . . ll'le8) . . . . . .
I O . . . .td7 . . . . . . . . . . . . . .
.td2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.tgS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
b4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9 . . . ll'lh5 . . . . . . . . . . . . . . . . . . . . . . . . . .
I 0 g3 f5 I I ll'lg5 ll'lf6 1 2 f3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
..
..
..
..
..
..
..
..
..
..
.....
.....
.....
.....
.....
.....
.....
.....
.....
.....
.
.
.
.
.
.
.
.
.
.
...................................
...................................
. . . . . . . . . . . . . . . . . . . . . . . . . .
...................................
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
...................................
f4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 3 b5 . . . . . . . . . . . . . . . . . . . . . . . .
l 3 c5
....................
1 3 'it>g2 . . . . . . . . . . . . . . . . . . . . .
1 0 ll'ld2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
( 1 0 c5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 0 J:l.e l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
( I O . . . a5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I O . . . f5 l l ll'lg5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
..............................
l i . . .ll'lf6
1 2 f3 . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 2 .t f3 . . . . . . . . . . . . . . . . . . . . . . . .
1 1 . . .ll'lf4
..............................
1 0 . . . ll'lf4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
91
91
93
95
97
97
99
1 04
1 06
1 06
I 06
1 10
1 15
121
1 27
1 27
1 27
1 27
1 27
1 28
1 29
1 30
141
152
141
141
141
1 47
153
1 55
9 . . . a5
1 0 bxa5
IO
IO
J:I.xa5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 5
cs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 8
1 0 .ta3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 3 7
...
...