Pawn Chains
•
SCHACHVERLAG
u.
VERTRIEB
OLBRICH GmbH
WURZBURG
1994
CJP-Titclaufnahmc dcr Deutschen Bibliothek
C1"0uch, Colin:
Pawn Chains: The present
book
looks at one J1ar1icular
tyrc
'lf pawn tormntion.
the central pawn
c
hain
,
and \\ith various examples the various ways in which the
positions resulting from such a fonnation may
be
handled.
(Schachvcrlag u. Vcrtricb Olbrich
Gmbll)
ISBN 3-929J24-10-S
ISBN 3-929324-10-5
0 1994
by
S
c
hachverlag
u.
Vertrieb
Olbrich GmbH
All rights reserved. No part of this book
may be
reproduced, stored in
a
retrieval system
or transmitted in any form or by any
means:
electronic, magnetic tape. mechanicaL
photocopying. recording or otherwise, without J1rior permission in writing from the
publisher.
Designed by Raddatz, Gennany
Technical Editor: Alexander Kulagin,
Jerzy
Dobosz
Printed in Poland
£.ontentH
FOREWORD by Colin Crouch ................................................................ 6
1.Pawns chains: the basics .
.
.
.....
.
...
.
..
.
.
.
......
..
....
.
.
..
. .............
.
...
.
.
. . ........ ..
.
9
Game
1
Korchnoi- Kasparov,
Amsterdam
1991
.
..
.
..
.
.
.....
..
....
.
.............
14
2.The
French is not the King's lndian
. . . ..........
.
. . ..
...
..
.
. .
.
..
.
.
.
.
.
.
.
..
..
..
. . ..
.
. . .
.
19
Game
2
Abramovic- Kovacevic, Yugoslav Ch.
1989 .......................... 22
Game 3
Gallagher- Crouch, Krumbach
1991
.
.
.
. . ............
: ................. 25
3.Pawns
or squares?
........................................................................... 30
Game 4
Larsen - Bareyev. Hastings
1990/91
...................................... 32
Game
5
Nimzowitsch- Salwe, Car1sbad
1911
.
.
.
.
....
.
.
..
.......
.
...............
.
.
38
Game 6 Korchnoi- Geller, Moscow 197Q (1 st
iiiei!\:,�.
�c.ii-•c)
. . . ...
.
.
.
...
.
...
43
4.The
Tarrasch French
..........
.
....
.
.
....
.
. . . .
.
............................... ..............
48
Game
7
Ljubojevic-
M.Gurevich, Linares
1991
...
..
..............
.
....
.
.
.......
.
4
9
Game
8
Pyda - Likavsky. Czechoslovakia
1991 .................................. 53
Game
9
Smagin - Vaiser. Bamaul1984
..
.
.
.
.
.
..
.
. ......................
.
...
...
. . . ...
60
Game
10
Ernst-
Crouch, London
1991
.
.
.
....
......
.
................
.
..
..
.........
..
.
64
S.The
unt>roken c:llain
. . ...........
.
..
..
.
..
..
.
.
.
.....
.
........
.
. .
.
.. . .
.
..........................
.
69
Game
11
Piket - Kasparov, Tilt>urg
1989
.
.
.
.
.
.....
.
. ...
.
... . ...
.
.... ......... . ....
69
Game 12 Cet>alo- Cvitan, Yugoslav Championship
1986 ................. 75
•.
6.The
paradoxical push
...
.
...
.
.....
.
......
..
. .
.
........ . . .
.
....
.
.
..
.
......
.
..
.
.......... . ....
80
Game 13 Khalifman- Gelfand, Reggio Emilia
1991/92 .................... 80
Game 14
Lobron - Renet, Novi Sad
1990
.
.
...... . . . . .
.
. ..
.
................ ....
.
.
82
Game
15
Mircovic- Todorcevic. Yugoslav Championship
1991
.
.......
85
7.Biock and t>reak
............ ................... ...................... .. . . ............. . . . ......
94
Game 16 Olafsson- Petrosian, Bled
1961 .......................................... 95
Analytical
exercises
......................................................................... 1 02
5
The computer age has brought about a remarkable informa
tion explosion in chess, and this has meant that a certain type of
writing, pioneered by
lnformator,
has become prevalent.
.. Languageless" notes, using sequences of moves and a range of
symbols, can be read by anyone, and can be stored in comput
ers. The convenience of such notes can often hide the fact that
while it is possible to examine tactics {precise sequences of moves}
in great depth, it is made very difficult to write about strategy, xd6
for example says very little about
why
a weakness on d6 should
be a dominant feature of the position; to explain, one needs to
write notes in natural language.
Pawn structure lies at the heart of strategy, and thus to begin to
write about strategy it is necessary to write about particular types
of pawn structures. We can leave such generalities as ucontrol
the center", "develop your pieces", .. attack where your opponent
is weakest" for the relatively elementary texts; more specific ques
tions need to be addressed. In the current book I examine the
types of position which result from when a pawn chain (d4, e5 v
d5, e6, or e4, d5 v e5, d6} is estab:ished in the center.
This was a natural starting point for me since when I started writ
ing the book the French Defence was my front-line reply to
1
e4,
and since in the age of Kasparov many of the most interesting
theoretical debates and games are in the King's Indian.
I hope that other titles, covering other types of position will follow.
There
is
of course no single way of playing a pawn chain position,
and indeed games played in the French and the King's Indian
often appear as though they have nothing in common ... It is pre
cisely this variety of different approaches which I wished to cover
in the illus&:rative games I have presented. Even sub-themes {e.g.
the battle for control of the d4 square in Fiench Defence) could
have books devoted to them, but in practice the strong player will
gradually acquire a stock of understanding of such themes.
All I can hope is that I have given the reader something to think
about.
6
As part of the basis for my own notes, I have of course made
much use of already existing notes, in
lnfonnator
or elsewhere.
Often even the best written notes leave questions unanswered,
and often published not�s can be dubious or quite simply wrong.
I have deliberately avoided entering analytical debates in the main
text, not through laziness. but because
I
felt it would be useful for
the reader himself or nerself to face the same analytical problems
that
I
faced when annotating the game. Thus at the end of the
book there is a series of about twenty analytical exercises for the
reader to try out.
London.Janua�. 1994
7
1. Pa"·n Chains :
The Rasles
At
the heart of the opening
struggle in chess is the battle
for the center. Naturally, with
players of similar strength, the
opening battle is unlikely to be
resolved decisively; what in
stead happens is that particu
lar types of opening battles
lead to particular types of
pawn structure, and that the
pawn structures which result
set the agenda for middle
game battle. In this book we
examine one typical pawn
structure, the central pawn
chain, we see how this struc
ture arises, and we consider
t�.e typical middlegames which
result. Most of the examples
are taken either from the
French Defence or the King's
Indian Defence, but the basic
central formation can arise
from other openings as well,
for example the Ruy Lopez,
the Sicilian (particularly the
2 ... d6 3 Bb5+1ines), the Caro
Kann (3 e5 lines), the Nimzo
lndian (where Black plays ... d6
and ... e5), the Hungarian, the
Philidor, and many others be
sides.
Firstly, we consider in detail
9
how and why the basic struc
ture arises.
1-
French Defence
Stage
one - the
small center
The French Defence is the
best illustration of the small
center.
1
e4 e6
2
d4
dS
Familiar enough, but what has
been happening?
With his first move,
1
e4, White
has laid claim to some space
in the center. Black has vari
ous ways in which to respond.
The .. classical" method would
be for Black to stake an equi
valent claim with
1
. . .
e5.
White· e sole advantage then
would be his extra move, and
he would use this advantage
by trying to lay siege to the e
pawn, starting perhaps with 2
Nf3 (attacking) 2 ... Nc6 (de
fending) 3 Bb5 (attacking the
defender).
Black
has his center, but this
center
can be attacked.
The small center approach
!s
based
instead on the idea of
creating a central strong
point so secure that it can
not possibly be overrun.
In
the diagram position for ex
ample, thera is simply no point
in trying to attack the d5
square; White must seek his
advantage in ether ways. The
most important point in his
favour is that he has an advan
tage in space
;
White has con
trol and use of a central square
( e5) in the Black half of the
board, whereas Black has no
corresponding outpost in the
White half of the board. Note
however that Black is attack
ing the White pawn on e4; this
creates some tension in the
central position.
In such position, with one
pawn being attacked by an
other, there are three choices:
(I) Simplify by exchanging
pawns (3 exd5)
(II) Advance the pawn (3 e5)
(Ill) Maintain the tension by
protecting the attacked
pawn (3 Nc3; 3 Nd2)
10
With the second option, 3 e5
White can create a pawn chain
immediately; we shall consider
a little later why he generally
prefers not to. The third option
is the most common, and
takes the central battle to a
new stage.
Stage two - central tension.
3Nc3
White protects his attacked
pawn. Now Black must decide
whether to simplify by ex
changing pawns (3 ... dxe4), or
whether to add to the central
tension by a further attack on
e4 (3 ... Bb4; 3 ... Nf6).
After 3 ... dxe4 4 Nxe4 Nd7 5
Nf3 Ngf6 Black's position is
perhaps sound enough, but
White has unqu
e
sti
o
na
bl
y
got
an advantage in space and
freer piece mobility. Black must
play carefully to equalize.
3
...
Nf6
Again White has a threefold
choice: simplify (4 exd5), ad
vance (4 e5) or maintain the
tension (4 Bg5).
Simplification does not ach
ieve very much here, but 4 e5
is a perfectly valid altemative
to the text move.
4
BgS Be
7
Black in his turn maintains the
tension, which he could also
have done with 4 ... Bb4, meet-
ing 5 e5 with 5 ... h6 6 Bh4 (6
Bd2 is better) 6 ... g5.
He could also have relieved
the tension with 4 ... dxe4 5
Nxe4 Be 7, more popular here
than on the previous move
since it is easier with the White
bishop already on g5 for Black
to offer exchanges to make his
position less cramped. After 4
. .. Be 7 we reach a critical posi
tion.
Stage
three-
formation of the
pawn chain.
White still has his three-way
choice, but his alternatives are
slowly diminishing; 5 exd5 of
fers little, while the pressure on
the e-pawn is now strong en-
11
ough to force White into a de
CISIOn.
Further protection of the e
pawn gives White less than
nothing after 5 Bd3? Nxe4 6
Bxe7 Qxe7 (6 .. . Nxc3 7 Qg4
is unclear)
7
Nxe4 dxe4
8
Bxe4 Qb4+. White could keep
some tension by removing an
attacker of the e-pawn with 5
Bxf6 Bxf6 6 Nf3, but then d
pawn comes under attack with
6 ... c5. Maybe White can try
to claim a tiny edge after 7
exd5 exd5
8
Bb5+ Nc6
9
dxc5
QaS
10
Na4, but after
10
. . .
0-
0
or
10
.
.. Bd7 Black can cer
tainly dispute this claim.
Really, 5 eS is the only way to
try for an edge.
5e5
Now the pawn chain has been
formed; White's pawns on d4
and e5 encounter Black pawns
on
dS
and e6. The next few
moves indicate how the strug
gle might develop.
Stage
four
- s�ruggle over the
pawn chain (early stages)
5
...
Nfd7
6
Bxe7 Qxe7
7 f4
0-0
If immediately
7
..
.
c5? then
8
NbS is awkward.
8
Nf3 c
S
9
Qd2
Nc6
The form of the struggle is
gradually becoming clearer.
White has more space and his
pawns look imposing, but all
this gain of space is at the ex
pense of compactness of stru
cture. Black can attack the
central White pawnsl
The priority for Black is to at
tack the base of the pawn
chain, the White pawn on d4.
Should this pawn be dis
lodged, the White pawn on eS
will be open to challenge with
a timely . .. f6. Quite often
White's d- and e-pawns will be
exchanged for Black's c- and
f-pawns, leaving Black with an
isolated pawn couple on e6
and dS, and leaving the dark
squares on d4 and eS open for
occupation by pieces. Should
White be able to establish a
12
secure blockade of these dark
squares, than he will have the
advantage;
Black
for his part
can aim for counterplay along
both c- and f-files. This is one
way in which events can de
velop from such a position;
other possibilities will be out
lined in later chapters.
The small pawn center strat
egy may also be employed in
queenside openings, although
it is not quite as easy for Black
to establish a stable d6-e5
center as to establish a d5-e6
center. The main line of the
King's Indian Defence leads
however to a pawn chain after
a phase of tension.
1 d4 Nf6
2
c4 g6 3 Nc3
Bg7
4
e4
d6
5
Nf3
0-0
6
Be.2 eS
i!
�.-i
'iV !:
l!t� :
1Ail.
i.Q.il
'
I. 1.1·, i
I.
r, t\
;�\
:
t."'""\
,:-,
.
f
.
.
(L_j
1-;1
I
.
I
� i.\
;:
'-�
;�
..
lS i\ 1\i
i-£ _A· �Y r/_!j
gl
l_
-
·--
--
.
------"
The small center established:
White cannot win a pawn here
because of
7
dxe5 dxe5 8
Qxd8 RxdB 9 Nxe5? Nxe4 and
Black is already better.
As in the French example,
White has three options here:
(I) Exchange with 7 dxe5
(7 ... dxe5 8 Qxd8 Rxd8
9
BgS!)
(II) Advance the pawn (7 d5)
(Ill) Maintain the tension
(7
0-0).
Again, as in the French De
fence, White usually prefers
not to release the tension with
a pawn advance until abso
lutely necessary.
7 0-0 Nc&
More pressure o� the d4
square; note how the fianche
ttoed bishop on g7 adds to the
pressure. White must soon
make an important decision as
to how to handle the central
position.
Black can also play 7 ... Na6
(probably best met by 8 Be3)
or 7 ... Nbd7. The position af
ter
7
•..
Nbd7
(diagram) merits
closer attention.
White can form a pawn chain
here with 8 d5?1,
but it is not
advisable.
Black replies with
8
.
.. Nc5
9
Qc2 aS with a se
cure queenside position.
13
(after 7 ... Nbd7)
Later Black will be able to set
in motion his thematic attack
on the kingside by preparing
.. .f5. White's d5 move sells
control of the c5 square far too
cheaply: similarly, if White
were to play e5 in the French
before Black has played ... Nf6,
Black would have extra defen
sive options based on ... Ng
e
7
or ... Nh6. keeping an eye on
the correspondingly important
f5 square.
With d5 in the King's Indian,
or e5 in the French, White
guarantees himself a spatial
advantage on his leadingflank.
This spatial advantage ne
eds to be converted into
an
attack. The most appropri
ate form of attack is the
pawn storm.
In the King's Indian, White pre
pares to open lines with c4-c5;
in the French Defence the cor-
responding break is f4-f5.
If
the b
r
e
a
k
thr
ou
gh
square
is
well blockaded by Black,
then
it is generally prema
ture
for White to form
the
pawn chain.
If 8 d5 is a poor response to
7
.
. .
Nbd7, then what should
White play instead? Remem
ber that even without the move
d5 White still has the advan
tage in space (pawns on e4
and d4 versus pawns on d6
and e5), and as a result has
more freedom of action to
maneuver pieces behind his
pawn center. White has scope
for a greater number of con
structive waiting moves than
Black. He should therefore
play q uietly and see what
Black does. One possibility is
8 Qc2 Qe7
9
Rd1 c6, and only
now
10
d5.
10
.
.. Nc5 is inef-
fective, and 10 ... aS is weak-
ening, so Black generally
chooses to close the queen
side with
10
... c5, after which
White prepares to renew the
queenside attack with a3 and
b4.
After 7
.
. . N
c6
(instead of 7
. .. Nbd7) the d-pawn is under
great pressure, and White has
no really effective way to play
quietly. 8 Be3 has been tried,
but Black seems OK after 8
. .. Ng4
9
Bg5 f6
10
Bh4 g5 11
Bg3 Nh6.
The time is ripe however to
form the pawn chain; Black
cannot successfully blockade
c5.
8
d51
Ne7
Both players are losing! Black
is positionally lost on the
queen side, but once. he has
··
played .. .f5 and ... f4, \1\/hite is
p
ositionally lost on the king
side. The result on the game
depends on which comes first;
Black's collapse on the queen
side or White's collapse on the
kings ide.
14
Ga•e
I
Korelanol
•
K•sparo"
_
J\tDs&erd•• 1991
From the diagram we follow
the illustrative game Korchnoi-
Kasparov, Amsterdam
199 1.
One could fill a textbook on
King's Indian play from Kas
parov's games alone.
9 Ne1 Nd7 1 0 Be3
We shall not consider ques
tions of opening theory too
deeply here; it is the general
strategic shape of the game
that is important. With the text
move White announces that
he is happy to allow Black to
get on with his pawn storm on
the kingside, so long as White
can have the queenside to
work on.
A
generation ago,
VVhite would usu;�lly have pre
ferred to sidestep Black's
pawn storm with for examp�e
10
Nd3 f5
1 1
exf5 Nxf5 (11
.. gxf51?)
12
f3 Nf6 13 Nf2
Nd4, Gligoric - Fischer, Bled
1961. Black is comfortable
though.
1 0
.. .
f
5
1 1
f3 f4
(diagram)
Extendind the pawn chain. It
would have been pointless to
play
1 1
.. .fxe4? since after
12
fxe4 all Black's attacking
chances on the king side would
have vanished, while White,
with the d5 pawn still firmly an-
15
chored, could continue to play
for a queens ide attack.
White's e4 is so firmly held as
to be virtually immune to at
tack. Therefor Black gain s
space on the king side, and
prepares to prise open lines
against the White king with
... gS and ... g4. Fischer said of
this variation that ��Black's
king side attack has practically
been worked out to a forced
mate", yet if White responds
vigorously enough on the
queenside, anything could
happen - including a White vic
tory:
12 Bf2
Here the bishop helps both in
defence and attack.
12
...
g5
1 3
a4
One cf
Korchnoi's attempts to
enliven White's play. Previ
ously the standard plan was to
roll away with b4 and c5, con
tinuing perhaps with Rc1,
cxd6, a4, NbS, etc. But what
is White actually doing with his
b-pawn? Perhaps not very
much. Korchnoi reasons that
if he can force through
c5 with
out
having to play b4 first, he
might be able to save an im
portant tempo .
In su ch a
sharp variation, every tempo
c ounts. Both sides must
strive to deploy their pieces
with the utmost efficiency.
13 ... Ng6
Black gets on with his plan. If
13 ... aS 14 Nd3 (preparing cS)
14 ... b6 15 Be1! with advan-
tage to White, Kozul- Popovic,
Yugoslavia
1991.
The idea is
to clear the f2 square for
knight, making it difficult for
Black to play ... g4 ( ... hS is met
by h3), while also preparing to
play NbS and b4, opening up
lines on the queenside. De
fence combined with attack.
14 Nd3 Nf6 15 c5 h5
(diagram)
16 h3
16
White understands that he
cannot stop ... g4 forever, but
at least he can make it more
difficult to play. Black must now
rearrange his kingside pieces
to prepare for his natural
break. This gives White a little
more time to prepare his own
attack.
In an earlier game, Larsen -
Torre, Bauang, 1973, this po
sition was reached by transpo-
··sition {13 Nd3 Nf6 14
c5
Ng6
1S a4 hS) and White pro
ceeded with rather less sub
tlety: 16 cxd6? cxd6 17 aS g4
18
NbS. (diagram)
White has ·won" on the queen
side, but now
18
.
.
. g3! for
once, the effect of the exten
sion of the pawn chain is to
open the position up. If now
19
hxg3- fxg3 20 Bxg3 h4 21 Bf2
Nh5 followed perhap s by
... Qg5, and Black's kingside
attack must surely be decisive.
In the game White tried 19
Bxa 7, but Black still had his at
tack: 19 . .. Nh7 20 h3 Qh4 21
Bb6 Bxh3 22 gxh3 Qxh3 23
Rf2 Nh4 24 Qf1 gxf2+ 25 Nxf2
Qg3+ 26 Kh1 and now simply
26 ... Nxf3 should win. After 26
... Rf6??, as played, White's
material loss suddenly looked
like a formidable sacrifice: 27
Nh3! Ng5 28 NxgS QxgS 29
Nc7 Ng6 30 Qh3 and White
later won the game.
Back to the Korchnoi - Kaspa
rov game, and 16 h3.
1 6
.. .
Rf7
17
c6
(diagram)
An incredibly sharp position,
which Kasparov tends to skate
over in his analysis in
Infer
mater.
It is a good exercise for
a player of any strength just to
get out the pieces, set up this
position and analyze it for an
hour or ai1 evening. to try to
l
--
P;.w., Chams
17
find out what is happening
here.
17 ... g4? is obviously incorrect;
White has 18 fxg4 hxg4 19
cxb7 Bxb7 20 Bxg4.
Kasparov also questions 17
... Bf8 because of 18 NbS a6
19 Na7, an unusual maneuver
designed to exchange Black's
light squared bisnop and make
it more difficult for him to play
... g4. Black can vary however
with 18 ... b6, intending to meet
19 aS with 19 ... g4 20 axb6
cxb6. What do you the reader
make of this position? Which
is more important - that White
has his protected passed
pawn? or that Black has al
ready playe� ... g4? You will
soon come to appreciate that
White will have a few regrets
about the placement of his
knight on bS; if he should later
play fxg4 then ... Nxe4 could be
dangerous.
17
...
a5!?
Kasparov chooses a different
plan of defence. He is quite
prepared to let White have a
passed a-pawn, provided he
can keep all entry squares on
the b-file under control.
Thus if White should try 18 b4,
Black is quite happy to play 18
... b6 19 bxa5 bxaS and if nec
essary just let his a-pawn drop.
18 cxb7 Bxb7 19 b4 Bc81 20
bxa5 Bh61 21 Nb4?
Too greedy. Kasparov sug
gests that White had to try 21
a6 Bxa6 22 Nb4 Bc8 with an
unclear position. One passed
a-pawn is enough
I
Maybe this
is another position for the re
ader to examine more carefully
(Exercise 2). I suspect that
most strong players would
rather be Black here.
21 ... g4 22 Nc6?1 Qf8
(diagram)
White bangs in his horse on
c6, just to show how much
ground he has taken on the
queen side.
Yet does it really matter? White
is not going to queen. He will
18
not be able to control the back
ranks and thereby press Black
i.'lto one small comer. And his
proud knight?
Merely a statue; it does noth
ing to protect the White king.
23 fxg4 hxg4 24 hxg4
Or 24 Bxg4 Nxg4 25 hxg4 f3
with a winning attack.
24 .
..
Bg5 25 Bf3 Qh6 26 Re1
White's only hope is to escape
with the king via e2. He is just
a little bit too slow.
26
.
.
.
Nh4!
(diagram)
To meet 27 Kf1 with 27 ... Nxf3
28 gxf3 and a piece sacrifice
on g4. Black's pawn roller has
done its job; now it is the turn
of the piece mass to weigh in.
Should White play quietly,
... Rh7 will follow.
27
Bxh4 Bxh4
28
g5
A vain attempt to divert Black's
attention from the h-file. 28 Kf1
offers more resistance, but
Kasparov points out that Black
is winning after 28 ... Nxg4! 29
Bxg4 (29 Ke2 Ne3 etc.) 29
... Bxe1 30 Bxc8 Qh1+ 31 Ke2
Qxg2 + 32 Kxe 1 (32 Kd3
Qg3+) 32 .. .f3.
28
...
Qxg5
29 Re2
Ng4 30
Rb1 Bg3 31 Qd3 Qh4
White
resigns.
2.
The French Is
�ot
T•e
King's Indian
In the Korchnoi - Kasparov
game, Black was attacking on
a very narrow front, with only
one possible pawn break, and
yet his attack succeeded.
White was operating on a bra-
19
ader front, which in theory
should give him more scope
for maneuver and the opening
up of lines, and yet his attack
was only partially successful;
he took
comniAte control of the
queenside, and yet there was
little he could do with this con
trol. So why the difference?
The answer lies in the place
ment of the kings; Black's
kingside attack is potentially a
mating attack: whereas White
is playing for lesser stakes, the
Black king being far away from
the firing line. Black's plan of
playing ... f4 and aiming for a
break with ... g4 (even though
it sacrifices a pawn) would not
look so impressive if the White
king were nowhere in sight.
What of the French Defence?
Outside the Winawer Variation
(1 e4 e6 2 d4 dS 3 Nc3 Bb4,
often with later ... Bxc3+),
where the doubled White c
pawn changes the strategic
complexion of the position, it
is rare for Black to play ..
.
c4.
Either the move is so strong
that White will not allow it, or
the move is irrelevant to the
needs of the position. There
are few intermediate cases.
Consider for example the fol
lowing sequence:
1
e4
e6 2 d4 d5
3 Nc3 Nf6
4
e5
Nfd7 5 f4
Already we have almost a mir
ror image of the King's Indian
pawn structure. White is aim
ing to gain space on the
kingside; Black is aiming for
queenside counterplay. There
are however two important dif
ferences from the Kina's In
dian, one of which w
�
rks in
White's favour, and one of
which works in Black's favour.
Black's problem is that to try
to castle queenside and tuck
his king behind a fianchettoed
bishop would be more time
consuming than the corre
sponqing plan, involving king
side
castling, in the King's In
dian. By the time that Black
has played ..
.
b6,
.
.. Bb7, ... c5,
. .. N
c6,
.
. .
a6 (to circumvent
Nb5), ... Qc7, ..
.
0-0-0,
.
..
Kb8,
etc., White will be away on the
kings
ide.
20
In
compensation
however,
Black will
be ab
le to create
pressure
on the
base of the
pawn chain
with
...
Qb6, a plan
which is simply not available
in the King's Indian. The battle
for the d4
in the French
is likely to be more intense
•than
the .. b.atti
e
"
for e�i
in
. .
- .. _.._.
.
.__,.. __
�
King's Indian.
5
...
c5 6
Nf3 Nc6 7 Be3
aS
7 ... cxd4 is also possible,
meeting 8 Nxd4 with 8 ... Bc5.
The immediate 7 ... Qb6 has a
slightly suspect reputation;
Black has to sacrifice a piece
for thr&a pawns after 8 N a4
Qa5+ 9 c3 cxd4 1 0 b4 Nxb4
11 cxb4 8xb4+ 12 Bd2 Bxd2+
13 Nxd2 g5, but might have
sufficient play.
·
8
Qd2
b5
Here no strong player
'.vould
c
o
n
t
e
m
p
la
te
9
0-0-0?? Tnis
m
ov
e would
be
p
os
i
t
i
o
n
all y di
sastrous; Black plays
9
.. .
c
4
a
n
d
prepares to give
check
mate.
White must therefore try
to find
other ways to make
progress.
In
A
n
a
n
d - Dreyev, 4th
match
game, Madras
1991, W
h
i t
e
tried
9
h4 Bb7 10 h5?!,
but with
the central position
still
unre
solved, this proved
extrava
gant:
1
0 ...
b4 11
Na4
and now
11
.. . cxd4 was good,
but 11
.
.
. Nxd4!
12
Nxd4 cx
d
4 13
Bxd4 Bc6
would have
been
even
better. Not however
11
.
..
c4?
12
b3 when the stabiii
zation of the center (d4
is
n
o
longer under pressure)
will
greatly assist in the de
v
e
lo
p
ment
of White's kingside ini
tiative. Black has
n
o
compen
sating initiative
on the queen
side; his pawns are advanced,
but they do not really
attack.
A
more common respo
n
s
e
1s
9 dxc5 Bxc5 10 Bxc5
Nxc5
when the position looks as
much like a Sicilian as a
French.
11 Qf2 Qb6 12 Bd3
21
In this
p
o
s
iti
o
n
Black should be
OK.
but
he
needs to be a bit
careful. Note how the ex
c
ha
n
g
e
of
pawns
in
the cen
ter has shifted the emphasis
away from
pawn play and on
to p·iece play. For example, if
Black
castles
here. V\'hite is
itch:�g to
play 13 Bxh7+ (ana
lytical
exercise: is
12
...
0-0
playable here? what happens
with
best
p
l
ay?). Even if Black
delays castings unti
l
.a
safer
m
o
m
e
nt
.
White can
play for
a
kingside attack
with pieces.
A
ga
m
e Mokry
- Eruk, Haifa
1989, continued 12
... b4?! 13
Ne2 aS
14 0-0
Ba6
15
Kh1
B
x
d
3
16 c
xd
3
Rb8
17
Rad1!
0-0
18
Qh4 f6 19 Ng3 and
White had some irritating, but
not yet decisive.
kingside pres
sure.
Per
h
a
p
s
Black too would have
done
better to play with pieces
r
a
t
h
er
than pawns: 12
... Rb8!
13 0-0
Nb4
14
Rad1 Na4! 15
Nxa4 bxa4 and because of the
pressure on the b-pawn, White
has nothing better than
16
b3
with equality, Bareyev - King,
Hastings 1990/91.
12
..
. Nxd3+? is a positional
blunder; after 13 cxd3 the
Black bishop is by the worst
minor piece on the board; it will
be able to attack nothing.
All the above is far removed
from the violent attacks on the
opposite flanks characteristic
of the King's Indian or the Si
cilian, yet it only takes a slight
difference in move order to
make a big difference in the
character of the game.
Ganae
2
Abramovlc
•
Ko"'acevlc
Yugoslav Ch.
1989
1 e4 e6 2 d4 d5 3 Nc3 Nf6 4
e5 Nfd7 5 f4 c5 6 Nf3 Nc6 7
Be3 cxd4 8 Nxd4 a6 9 Qd2
Bc5
(diagram)
In such systems it is inevitable
that Black's c-pawn will at
some stage be exchanged for
White's d-pawn, the base of
his pawn chain.
It makes a difference though,
whether Black or White ini-
tiates the exchange.
If White exchanges with dxc5,
d4 becomes an
---
-
a lot of
��L�
..
���-
If as here Black starts the ex-
.....
�
hange wit
�
... cxd4, .. then the
d4 square becomes occupied.
that
It is not so easy for White to
move the knight from d4 -
White must for examJ,Jie watch
out for
...
f6-
but for as long as
-
.
the knight can be safely main-
\
tained on d4 it has a genuinely
cramping effect on Black. The
center is in effect sti�l blocked
by a pawn chain, even though
part of White's chain happens
22
o be
a
knight! With the center
blocked, pawn play on the
flanks becomes the indicated
strategy.
1 0
0-0-0 0-0 11
Kb1
1 1
Nb3
is possible, but decen
tralizes the k n i g ht: .B lack
avoids the exchange of
bish
ops wi
th 11
. . .
Bb4 or 1 1 .. . Be7
11
. . .
Nxd4 1 2 Bxd4 b5 1 3
g
4
Still paying the blocked center
strategy, although may be the
re was a strong case for emp
tying
the d4 square by 1 3 Bxc5
followed by 1 4 Qe3 or Qf2.
White's most effective strategy
in the French is often to play
solidly, gently pressing, rather
than relying on all-out attack.
To
be firmly holding· the da(k
squares and to be making it
difficult for Black's queen's
bishop to get into play - that is
enough to cause defensiva
headaches.
1 3 . .. b4 14
Ne2
If
14
Na4 Bxd4 1 5 Qxd4 f6!
and suddenly Black is the one
attacking on the kingside. The
destruction of the White half
of the pawn chain! 1 3 h4 (in
stead of 1 3 g4?!) would have
avoided this particular pro
b
lem.
23
14
..
. a5 15 Bg2 Ba6
The first sign of real progress:
Black is preparing to exchange
his bad bishop for White's
much more useful knight.
16
Bxc5 Bxe21
All
Black's efforts would have
been wasted after 1 6 . . . Nxc5
1 7 Nd4! Qc7 1 8 Qe3. White's
d4 square would then be com
pletely secure, and occupied
b y a powerfu l kn i g ht. H i s
spearhead pawn on e5 would
also be complete ly secure
(with . . .f6 being ineffective) ,
and this gives him the oppor
tunity to work on a pawn storm
on the king side. Such a storm
would be as forceful as any
queenside pawn storm. Such
a storm would be as forceful
as any queenside pawn storm
in the King's Indian, and would
be doubly effective in that the
Black king is fixed in the
storm's path. Black's attack on
the queenside would be puny
in comparison.
17 B xf8?!
White misjudges the pace of
the position. Kovacevic gives
.....
17 Qxe2 Nxcs; 8 Qe3 Qc7
19
Rhe1 Rfc8 20 f5 Nd7
21
Rd2
Qd8 as .. unclear". It may look
as though Black is still on the
defensive, out once he has
played ... N b6 that impression
will change.
17
...
B xd118 B xg7 B xc2+ 19
Q xc2 ReS 20 Qd2 Kxg7
A
lot has changed in the few
moves. Would it be too cat
egorical to suggest that it is be
cause White no longer con
trols d4 in any meaningful sort
of way that Black has the ad
vantage? There are other fac-
24
tors as well. Although White's
kingside pawns look danger
ous, Black's queenside pawns
are in fact farther advanced
and he already has control of
an open file leading directly to
wards the White king.
Should White oppose rooks on
this file, as is probably correct,
Black will still have the better
endgame: 21 Rc1 h6 followed
by
. .
. Rxc1 + and ... Qb6. Black
would have three main advan
tages in the endgame: his
passed d-pawn, the fact that
his knight would have more
scope than the White bishop,
and the weakness of White's
kingside pawns, especiaily the
f-pawn.
Not liking this, White plays for
an attack.-
21 h4?1 Qc7 22 h5 h6 23 Bf3
a4
24 gS
The endgame is lost: 24 Rc1
Qxc1+ 25 Qxc1 Rxc1+ 26
Kxc1.
(diagram)
And now Black destroys the
last remnants of the White
pawn chain with 26 .. .f6!
(a move you will see often in
later chapters). 27 exf6+ Kxf6
28 Kd2 Nc5
29
Ke3 (had
White gone into the endin� a
n\ovu
uarlier, with 23
Rc1 in
P4tnnd of 23 Bf3,
.
.
.
Na4 would
11nw
hnve been playable)
29
,e:il
�iO b
x
a
3
hxa3 31 Kd4 (31
r,d:lloses
to
31
.
. . d4 and . . . e5)
:\ 1
N b3+!
32 Kc3
Nc1
33
Kb4
Nxa2+
34 Kxa3
Nc1 fol
lowed by . . . Nd3 (variation by
Kovacevic).
24
...
hxg5 25
Rg1
b 3
26
Rxg5+
Kh8 27
Rg2
White
also loses after
27
axb3
axb3 28 Bd1 Qc4
29
Rg3 Nc5
followed by . . . Qe4+.
27
...
Qc1
+I
White resigns.
25
28
Qxc1 Rxc1 + 29 Kxc1 bxa2
and queens.
At one level, a typical game
with
players castling on oppo
site sides and attacking furi
ously. At a deeper level, no
tice how the fortunes of the
game fluctuated according to
the degree of control White
kept on the d4 square. We
shall cover this subject in more
detail in the next chapter. Be
fore that we consider one of
the rare examples i n wh ich
Black can treat the French
Defence in the manner of the
King's Indian.
Game
3
Gallagher
•
Cro•eh
Krunabaeh
1891
1 e4
e
6
2 d4 d5
3
Nc3 Nf6 4
Bg5 Be7 5 e5 Nfd7
6
Bxe7
Qxe7
7
f4
The game pawn structure as
in the previous game, but with
the dark squared bishops al
ready exchanged. White's ba
sic aim with this early simplifi
cation is to take some of the
s�ing out of the battle for the
d4 square; there is going to be
no tension, for example, be-
tween
a
Black bishop on c5
and a White bishop on e3. Fur
ther more, the Black queen on
e7 is not particularly well pla
ced if queenside pressure is
the aim; b6 is the natural de
veloping square. The immedi
ate attempt to rectify the situ
ation by 7
.
. . Qb4? is easily re
pulsed:
8
a3! and if 8
.
.
.
Qx
b2??
9
Na4 traps the queen.
7
...
0-0
8
Nf3 c5
9
Qd2
9 dxc5 is perhaps more stan
dard . With the text White tries
to keep the Black queen away
from c5.
9
...
Nc6
1 0
0-0-0?1
c41
A
strong move unj ustly ne
glected by theory as a result
of a twenty year old game lost
by Black through a positional
blunder! My own analysis had
convinced me Black·s attack
26
wa s s l ightly q u icke r th an
White's.
For 1 0 .
. .
cxd4? see the Larsen
- Bareyev game in the next
chapter.
11 NbS RbB!
My attention was drawn to this
variation by
a
passing mention
of this position by Dolmatov,
who noted that White was bet
ter after 1 1 . .
.
a6 �2 Nd6 . Why·
though should Black waste a
tempo with . . . a6 when White
is planning to play Nd6 any
way? It is more important for
Black to unravel his queens ide
pieces, and to be prepared for
immediate attack as soon as
the White knight leaves b5.
1 2 g4-Nb6 1 3 Nd6 Bd7
(diagram)
Black's plan begins to take
s hape; he i11ten d s to play
... NcB, and if White exchanges
or
c8, he will recapture with
the king's rook, leaving rooks
on
both b-
and
c-files to back
up
the pawn storm.
White's kingside pawns are in
fact farther advanced than the
Black queen side pawns, but
are less well supported by
pieces. Thus 1 4 f5 is met by
1 4 . . .f6! and Black is the one
to benefit from the opening up
of lines on the kingside, for ex
ample 1 5 Qf4 Nc8 ! and all
lines look good for Black (in
lnformator I g ave 16 fxe6
Nxd6 here; a slip of the pen!
1 6 . . .fxe5 wins of course).
A
radical demonstration of the
power of the ... f6 break, made
even more effective by the un
prot�cted position of the knight
on f3.
1 4 h4 NcB 1 5 f51eads to a po
sition from Kapengut- Weste-
27
rinen, Vilnius 1 969; the actual
move
order
after 1 0
. . .
c4 was
1 1 h4
Rb8
1 2 Nb5 Nb6 1 3 g4
Bd7 14 Nd6 NcB 1 5 f5.
Here 1 5 . . .f6! is the thematic
move, when it is Black, not
White, who will be able to take
advantage of the opened lines
on the kingside.
Black is better. In the Kapen
gut - Westerinen game how
ever, Black snatched the pawn
with 1 5 . . . Nxd6?, the conse
quences being 1 6 f6! gxf6 1 7
exd6 Qxd6 18 Qh6 Qe7 1 9 gS!
f5 20 h5
f6
21 Rg1 KhB 22 Nh4
Be8 23 g6 (diagram)
23 . .. Rg8 24 Bh3 Qg7 25 Qe3
hxg6 26 Nxg6+ Bxg6 27 R
x
g6
Qf7 28 Rh6+ Kg7 29 Rg1 + Kf8
30 Rxg8+ Qxg8 (30 ... Kxg8 31
Qg3� Kf8 32 Rh
S
+ Ke7 33
Qc7 mate) 31 Rxf6+ Ke7 32
Rg6 Qf7 33 Qg5+ Kd6 34 Rg7
QeB 35 Qf4+ e5 36 Qh6+ Qe6
37 Rg6 Black resigns.
A punchy attack by White, but
with just a little care by Black it
could have been repulsed be
fore it started. In
the French
Defence Black must, if he
can, blunt the spearhead of
the White pawn chain on eS,
otherwise he is going to get
overrun on the kingside.
Back to the 1 991 game.
14
c3 Nc8 1 5 Nxc8 Rfxc8 1 6
Qc2
1 6 .
..
f6!
Yet again, Black plays a well-
28
timed defensive move on the
king side, and succeeds in slo
wing the opponent's attack to
a crawl.
Ng5 is prevented, and there is
no obvious way for White to
advance his kingside pawns to
batter the Black defences.
When both sides are attack
ing on opposite sides of the
board, as quite often happens
when pawn chains are length
ened,
the principle of mini
mum defence
becomes im
portant. Most of your moves in
such situations should be at
tacking moves, but you should
play just enough defensive
moves to s low your opponent
down on the flank where he is
attacking. Play too many de
fensive moves and you will
never get your own attack go
ing (and furthermore it is un
likely that you will be able to
hold off your opponent's attack
in the long run). Play too few
.defensive moves and you will
lose - very quickly!
Let us suppose that it takes
three moves to set up a really
big threat; c3, Qc2 and Ng5 to
threaten Qxh7+. The kingside
is not to be defended in the
long run, so it is not worthwhile
to play a series of fussy little
defensive moves. It is better
just to play
one
defens ive
move - the one preventing the
m ove wh ich creates a big
threat. In this case, it is better
to
prevent
Ng5 rather than to
find a c
u
r
e
for Ng5 once it has
been played. By limiting your
own defensive moves to a
minimum. you are giving your
self more time to develop your
own attack.
17
h4
b5
18 Kb1?
A surprise. The king decides
not to desert his burning ca
stle; a misplaced sense of ho
nor! Hiding in
the
forest with
Kd2 and Ke1 offered better
defensive chances; the gate
way on b2 could have been
protected by Rh2. Too many
defensive moves? Maybe, but
Black's attack is getting
too
dangerous to ignore.
18 ... b4 19
cxb4 c3
Overelaborate.
The
straight
forward
19
. . .
Nxb4 20 Qc3
Nxa2 21 Kxa2 Rb3 followed by
. . . Rxf3 i s an easy technical
.
w1n.
20
exf6 gxf6
21 Qxc3!
29
On
21
bxc3?
Nxb4 is crushing .
Now Black must be a bit care
ful, otherwise the win will slip.
21
...
Nxb4
The endgame after 21 ... Ne5?!
22 Qa3 Nxf3 23 Qxf3 Qxb4
2A
Qb3! is not particularly clear.
Had
Whit�
played 20 Qxc3 im
mediately, without inserting the
pawn exchange 20 exf6 gxf6,
then
20
. . .
Nxe5 would be a
capture . and Black would have
had a clear extra pawn for the
endgame.
22 Q a3 Q d6!
The or:-ly clear way. Now the
f4 square must fall. White the
refore seeks counterplay, but
Black! having control of more
squares! is quicker.
23 Bd3 Q xf4 24 Rc1
KhS!
So that Bxh 7 is not check. Now
. . . Qxf3 a n d . . . Qe3 are rea l
threats.
2 5
Rxc8+ Rxc8 26 Rf1
26 Qxb4 Qxf3 wins for Black.
But now for something a bit
prettier.
26
...
Bb51 27 BxbS Qe4+ 28
5d3 Nxd3 29 Nd2 Qe2 30
QaS Qxf1 +
White resigns.
3.
Pawns
Or
Sqaares?
By advancing a pawn in the
center to form an interlocking
pawn chain. White i ndisput
ably lays claim to more terri
tory in the center. The extra
space for White is perhaps not
so i m porta nt; what re ally
counts is Black's lack of space.
If for example Black's pawn
are pinned down to e6 and
d5
by the White pawns on e5 and
30
d4, it is difficult for Black to
develop his bishop on cB, or
to move pieces across from
the Gueenside. One way for
Black to deal with this problem
is, as we have already seen,
to bypass the White pawn
chain with ... f5-f4 in the King's
Indian, or, more occasionally,
by . . . c5-c4 in the French. The
skeletc:>n position in the dia
gram may then be reached.
Each s ide spearheads h i s
pawn chain with a pawn on the
fifth. giving perhaps an initia.l
impression of symmetry and of
equal ity of space. White's
spearhead however is in the
center, whereas Black's is on
the flank. Laterally therefore
White has more space; he can
develop his attack on a wide
front (a, b, c, d files) whereas
Black must work in restricted
territory (f, g, h files). For Blac�
this is quite a handicap, and
for the most part the bypass
ing maneuver is worth trying
only if the White king presents
a tempting target (as in the
Korchnoi - Kasparov and Ga
llagher - Crouch games). In the
French Defence, this target is
usually not there.
A more common plan i n the
French is a direct att�ck on the
White pawn chain. The ad
vanced pawn on e5 is of co
urse the real long-term target,
as that is the pawn which is
doing the most to cramp Black,
but no real progress is made
if B lack attacks
solely
the
pawn on e5, as it is well pro
tected by the d4 pawn. Fir�tly
Black must undermine the
base of the pawn chain wit:-.
. . . c5.
To this attack White has two
basic responses. One is to
play . . . c3, preserving the pawn
chain with pawns.
After . . . cxd4, cxd4 White's d
pawn is effectively isolated.
(diagram)
Black has various strategic op
tion �open with this type of
pawn formation, usually involv
ing some combination of direct
pressure against the d-pawn,
play along the open c-file, and
. . .f6 break, opening up the f-
31
file.
A
more detailed discussion will
follow in the next chapter.
There is no reason however
for Black to be terrified by this
formation.
White's other option is to omit
c3 and to prepare to recapture
on d4 with a piece, as in the
Abramovic- Kovacevic game.
This does not mean that White
is abandoning the pawn chain;
rather, he i s constr�cting a
piece-and-pawn chain. What
is restricting Black is not so
much White's
occupation
on
d4 and e5, but rather his
con
trol of d4 and e5. If White, af
ter an exchange of c-pawn for
d-pawn , ca n keep the d4
square firmly under his control,
and can a lso provid e ad
equate support for his e5
pawn, then Black has
a
diffi
cult position to play, as we
shall shortly see. If however
White loses his grip on d4, as
in the Abramovic - Kovacevic
game, Black usually has chan
ces for counterplay.
Ganae 4
Larsen
•
Bareye''
Bastings
1990191
1 e4 eS 2 d4 d5 3 Nc3 NfS 4
Bg5 Be7 5 e5 Nfd7 6 Bxe7
Qxe7 7 f4 0-0 8 Nf3 c5
9
Qd2
NcS
10
0-0-0?! cxd4?!
10 ... c41
1 1 Nxd4
The basic central structure is
now established. Black here
has to decide whether to base
his hopes on an attack on the
q ue e n si d e , o r whether he
s hould attack the White e
pawn with 1 1
.
.
.
f6
.
Here 1 1
. .
.f6
just creates weaknesses; af
ter 1 2 exf6 Qxf6 1 3 g3! Nb6
32
1 4 Bb5 Bd7 1 5 Rhe1 Rac8 1 6
Kb1 Black's position is not par
ticularly comfortable, in this
line, White still has the squares
d4 and e5 well covered, while
Black's backward pawn
on
e6
is under siege.
Black decides instead to get
things moving on the queen
side. And White, what does he
do? He must make every ef
fort to consolidate his center,
to keep d4 and e5 under con
trol. If he can do this, his cen
tralized pieces will help defend
the queenside and at the same
time attack the kingside.
11
...
N b6 12
Qe3
Bd7
12
.
. . Qc5?! 1 3 h4! Bd7 1 4 Rh3
Rac8 1 5 - Rg3! Nxd4 1 6 Rxd4
Qe7 1 7 h5 f5 1 8 exf6 Qxf6 1 9
Rg5! and Black's position is
very unpleasant, Short - Gul
ko, H astings 1 989/90.
The maneuvering of the \/Vhite
rook is attractive. From h3 it
protects the White
queen
on
e3, thus unpinning the knight
on d4. From g3 it pressurizes
the Black king, the move 17 h5
forcing Black to move his t
pawn at an inconvenient time.
After an exchange on
f6,
the
e5 square is open to the White
pieces, and White's Rg5 move
consolidates White's hold on
the critical square.
Fi rm control of
d4
a
n
d
e5
with pieces is least a
s
effec
tive
as o c cu pati o n with
pawns.
1 3 Kb1 Qc5
14
h4 Rac8
1 4 . . . NaSI?, as suggested by
Larsen, is a n a lternative. If
Black chooses this plan , he
must do so while the White
queen is still unprotected.
1 5 RhJ
(diagram)
As in the Short game. How
ever, White has taken time out
to play Kb1 , and so Black has
an extra half-tempo to play
with.
15
...
Na5?
3- Pawn Chains
33
A miserable little move of the
type which leaves the French
Defence player so often wear
ing a gloomy counte nance.
Black makes a
gesture
on the
queenside, but White can im
mediately force an exchange
of queens with 1 6 Nb3. This,
as we shall see, accentuates
White's central advantage;
without the queens Black has
little active play on
the
queen
side and he 3till has to decide
I
what to do with his bishop on
d7.
So oftere i n the Cl ass ical
French it seems that Black is
condemned if he exchanges
queens (a bad endgame) and
condemned if he leaves the
queens on (a kingside attack);
and indeed if Black remains
passive, then he is h eading
quietly for a loss. To defend,
he must play actively, take ev
ery possible tactical opportu-
nity that arises, before White
can lay the big squeeze on
him. A nifty ... f6 at the right
time can work wonders, and
here is one such moment -
maybe. Ou r fourth readers
�,..�.-cl�c i.:. &.u
ct11alyze the po
sition which results after 1 5
.
. .
f6.
Larsen passes over this pos
sibility without comment!
1 6 N b3 1 Qxe3 1 7 Rxe3
What a contrast to the position
which could h ave been re
ached after
15
..
.f6 1 6 exf6 e5!
White has a stable positional
advantage. There is no single
feature of his position that one
could point to as causing over
whelming problems to Black,
but there
are
lots of little things
which irritate Black.
- Black's bishop is by a long
way the worst minor piece
34
on the board. It has little
scope for movement at pre
sent, and even if the posi
tion opens up, it will have
little to attack.
- White has firm control of the
dark squares in the center,
with the empty d4 square be
ing a useful outpost, and the
pawn on e5 cramping Black.
- White has extra control spa
ce on the kingside, again be
cause of his pawn on e5.
Black must always beware
an attack on this part of the
board.
- It is difficult for Black to cen
tralize his king, because of
th� potential weakness of
the kingside.
What connects all these ad
vantages? They are features
of a position where White has
successfully maintained his
d4/e5 strongpoints i nto an
endgame. Furthermore it is all
gain; White has no real weak
nesses for Black to attack.
17
...
Nac4
1 7
. .
. Nxb3 1 8 axb3 makes it
even more difficult for Black to
gain counterplay; how does he
bring either of his minor pieces
into the game?
1 8
Rf3?1
i...at
::,�n
�uyy�SL�
u·aat 18 Ree1 ,
keeping pressure on the e-file.
would have bee n more
harmo
nious. Maybe during the game
he wanted to avoid having to
spend a tempo
p
rotecting the
f-pawn with g3 (after 1 8 Ree1
f6 a� �xi6 Rxf6
20 g3), but this
is not so
m
ething which should
worry White; he
can
besiege
the e-pawn with Nd4 and Bh3.
If after 1 8 Ree1 Black plays
passively, th9n White could
quietly build
up
with g3, Nd4,
Bh3, etc., and leave Black to
worry about the possibility of
f5. Or maybe he could keep
something else up his sleeve
. .
There is no hurry; Black has
no play.
1 8
...
f6 1 9 exf6 Rxf6
If 1 9 ... gxf6, White has a stan
dard reply - 21 f51 wrecking
Black's pawn formation.
20
Nd
4
RefS
(diagram)
21 Bxc4
35
In such posi
ti
ons White should
never be rel u ctant to ex
change bishop for knight;
af
ter all, the White king's bishop
is not involved in th.e struggle
for the dark squares; the Blac�
knight is, though.
Even so, White would have
liked to have played 2 1 g3,
but
here the drawback of his 1 8th
move
shows itself; Black
has
21 .. . e51
(22
Nxd5 exd4 )
.
21
...
Nxc4
21 . . . dxc4
22
Rdf1 keeps
uo
the pressure for White.
22 b3
Nd6
Again if
22
.
. .
Nb6 23 Rdf1 ,
or.
even better, 23 Nce2 followed
by Re3. Not though 23 Nde2.
when 23 . . . e51 is again trouble
some.
23
Re3
Yet again 23 g3 is met by 23
. . . e5! White just needs to play
one little pawn move, g3, be
fore he is in complete posi
tional control again.
23 ... bS?
Black ought not to give White
time to consolidate. He can
probably scrape a d raw by liq
uidating weaknesses with 23
... Rxf4 24 Nxe6 Bxe6 25 Rxe6
Ne4 26 Nxd5 Rxh4, as 27 Re7
can be met by 27 . . . Rf7.
24 a3 a5?
Still 23 . . . Rxf4.
25 g3
At last!
25 ... b4 26 axb4 axb4 27 Na2
36
Rb8
28 Kb2 Ne4
Presumably the position that
Bareyev been aiming for. He
has blocked off the e-file, and
has reached e4 just in time to
keep the knight on a2 out of
the game (29 Nc1 Nc3 etc.).
While Black's b-pawn is obvi
ously week, it is surprisingly
difficult for White to get a rock
to d4, since 29 Ne2 is met by
29 . . . Bb5. The White rooks
have no inroads either. So
Black is safe?
29 Nf31
No. White takes advantage of
a tactical trick (29
. .
. Nxg3? 30
N e5 Nf5 31 Nxd7 Nxe3 32
Rd3!) to gain time to shift the
knight to the other critical dark
square in the center, e5, or to
challenge Black's own strong
knight with Ng5.
29 ... Be8
30
Ng5 Bh5 31 Nxe4
Bxd1
32
Nxf6+ gxf6
33 Rxe6
Kf7 34 Re3
Bh5
3
5
Rd3 Ke6
36
Rd4
The b:-pawn goes as well, and
the Black bishop, although no
longer stuck behind pawns, re
mains targetless.
36 ... Rg8 37
Nxb4
Bf3 38 Rd3
Be4 39 Re3
Kf5
40
Nc6
Black
resigns.
The b-pawn is ready to run.
Not quite the smooth perfor
mance as it appears to be at
first sight, and for that very rea
son all the more instructive.
The blockade of d4 and e5
"hangs by a hair", to borrow a
p h rase from N imzowitsch .
If
the blockade holds, and
White can permanently re
strain the Black pawns on e6
(especially) and d5, without
allowing significant counter
play elsewhere, then White
has every chance of being
able to congratulate himself on
a smooth positional victory. Yet
the blockade is not always so
easy to maintain, particularly
if Black is seeking active co
unterplay.
Two lapses in concentration in
37
the Larsen - Bareyev g ame
come particu l arly to min d .
Firstly there was Bareyev:s
fail
ure to take advantag e of
a
fleeting opportunity to play . . . f6
before White was properly co
ordinated. He could have bro
ken the blockade and created
wild complications; instead he
subsided into a clearly inferior
position. Then Larsen retumed
the compliment with his care
less
18
Rf3?! leaving several
possibilities open for tactical
breaks by Black with ... e5; had
he secured the center by re
treating the rook along the e
file, and secured the f-pawn
with an early g3, then Black's
chances of creating active play
would have been minimal.
Our next game shows the d4
and e5 squares being cleared
at a very early stage, and
White successfully mai ntain
ing the blockade. It is
of
his
torical interest in that N im
zowitsch describes it as lithe
first in which
my
new philoso:.
ph
y
of the center was exhib
ited". Chess historians, or his
torians of chess thought, may
di scu ss wh eth er the re a re
clear antecedents; the game
itself looks modern enough not
to be out
of
place here.
Ganae
5
Nlm.w»wltseh
•
Salwe
Carlsbad 19 1 1
1
e4
e6
2
d4 d5 3
e5
c5
4
c3
Nc6 5 Nf3 Qb6 6 Bd3
At the moment the fashion
able, and controversial move
•
I
IS
6 a3. White's intention is to
pre-empt Black's queenside
play with an early b4 (6 . . . Bd7
7 b4), but the weakness of the
b3 square allows Black to ex
tend the pawn chain with
6
·
.
.
c4, bypassing the d4 square.
6
...
Bd7?1
More accurate is the immedi-
ate 6 . . . cxd4 7 cxd4, and only
then 7 . .. Bd7. The theory of the
day suggested that White then
had to play
8
Be2 in order to
protect the base of his pawn
chain, a possible continuation
being
8
. . .
Nge7 9 b3 Nf5 10
Bb2 Bb4+ 1 1 Kf1 , and now
Nimzowitsch recommends 1 1
.
. .
0-0! 1 2 g4 Nh6 13.Rg 1 f6
1 4 exf6 Rxf6 1 5 g5 Rxf3 16
Bxf3 Nf5. (diagram)
This position is worthy of close
study, as it is almost an ideal
French position for Black de-
•
I
sp1te the missing exchange.
38
White's d-pawn is evidently
very weak, and should it drop,
Black's d- and e-pawns will
beco me a powerful mobile
force in the center. Black also
has the potential for consider
able pressure along the f-file,
where White's g4-g5 thrust
has left many weak squares.
And White does not even have
control of the eS square, gen
erally the m ost vul nerable
point fQr Black after the ex
change of f-pawn fore-paw
n
.
White has no central control·
I
only central weaknesses.
Nimzowitsch gives as a likely
continuation 1 7 Rg4 Be8 (17
. . . Rf8!?) 18 Qe2 Ncxd4 1 9
_Rxd4 Nxd4 20 QeS BbS+ 2 1
Kg2 Nf5 2 2 Bxd5 exd5 23
Qxf5 Rf8 24 Qxd5+ Rf7 and
Black wins; White has defi
nitely not solved his problems
of poor development and king
exposure.
Should the tempo loss with
8
Be2 not appeal,the Milner
Barry Gambit with 8 0-0
Nxd4
9
Nxd4 Qxd4 1 0 Nc3 has to
be considered. Black can try
snatching the second pawn
with 10
. . .
Qxe5 1 1 Re1
QbB;
White has obvious compensa
tion but no clear advantage.
7 dxc51 1
Nimzowitsch'sll
7
...
Bxc5 8 0-0
White is more concerned to
maintain his strongpoint on e5
than to maintain the integrity
of his pawn chain. Indeed, cir
cumstances being right, he is
quite happy to exchange his
d-pawn for the Black c-pawn,
lea
v
i
n
g the
d4
square open,
in order to remove from him
self the obligation of protect
ing the d-pawn. For this plan
39
to be successful, White must
be in a position to provide the
eS pawn with very secure pro
tection. Indeed, Nimzowitsch
uses the term l&overprotection",
·
and advises that White should
pile up all his pieces behind
the strongpoint, defending it so
securely that any attack by the
opponent is fruitless.
The two basic points about
overprotection
are
firstly that
the overprotected strongpoint
shall be maintained, and sec
ondly that the overprotective
pieces themselves have flex
ibility. If a strongpoint is mini
mally defended,
it
can be held,
but the pieces protecting the
strongpoint are tied down to
defence; none can move with
out abandoning the strong
point to liquidation or wo.-se. If
however the strongpoi nt i s
overprotected,
any of the
overprotective pieces may be
come involved in opportunis
tic forays elsewhere without
having to worry about the sa
fety of the strongpoint.
8
..
.f6
"Black swells in triumph and
throws himself hungrily on the
last remaining member of the
once so proud chain-family, to
destroy him. H is war cry is
·Room for the e-pawnr but
it
h ap pe n s q uite otherwise.··
(N imzowitsch)
9 b4
Maybe Black could have tried
8 . . . aS to prevent this move . .
9 ... Be7 1 0 Bf4
White's plan is beginning to
take dear shape. If he can find
time for Qe2, Nbd2, Nb3, then
m aybe a3 and c4, Black
is
g oing to suffocate . I n the
meantime Black's kingside
pieces get in each other's way.
Black must act very quickly. In
the game he tries an immedi
ate exchange on e5, but this
does not help him much. Our
fifth reader's exercise is to de
termine whether Black could
h ave got a satisfactory game
40
by breaking up White·s queen
side with 1 0 . .. a5.
1 0
.. .fxe5 11 Nxe5
Nxe5
12
Bxe5
So the pawns may have gone,
but who coul d d o u bt that
White has full control of the d4
and e5
squares?
12
. . .
Bf6
fails to 1 3 Qh5+ and if 1 3 ... g6
14 Bxg6+
1 2 ... Nf6 1 3 Nd2
To tighten his grip on the cen
tral dark squares. White a
voids dabbling with shallow
tactical threats (1 3 Qc2?) and
concentrates on strategy. Af
ter 1 3 Qc2? 0-0 1 4 Bxf6?
Rxf6 1 5 Bxh7+ Kh8 Black is
happy; White's grip on the
··
center has gone, his queen
side is weak and undeveloped,
and Black is ready to advance
in the center ( .. . e5) and attack
on the kingside. it's
not
worth
chasing the pawn_!
1 3 . .. 0-0 14 Nf3 Bd6 1 5
Qe2
Con_scientious and necessary
overprotection of e5.
White ensures that his outpost
is both occupied and defen-
ded. Nimzowitsch points out
that 1 5 Bd4? Qc7 1 6 Qe2
would be inaccurate; Black
frees himself with 16 . . . Ng�! 1 7
h3 e5! with total freedom for
Black. We leave it as a rea
der's exercise (number 6) to
determine whether White can
successfully play 1 7 Bxh7+
instead of 1 7 h3; some typical
French Defence tactical sequ
ences are involved.
1 5 . . . Rac8 16 Bd4 Qc7 1 7
Ne5
The piece chain is complete!
The pawns have gone; long
live the squares!
Of course the piece chain will
not last long as Black will feel
obliged to exchange on e5 at
some stage. In such an ex
change however on e of
Black's "dark square" minor
pieces will be exchanged for
a White 11dark square" minor
41
piece, with the end result that
White
will continue to domi-·
nate the dark squares. On the
light
squares, White's bishop
on
d3,
patrolling an important
open
diagonal leading to the
Blac'<
king, is vastly more ef
fective
than the hemmed-in
bishop
on d7.
Blaci<
may already be position
ally lost.
17
...
Be8 1 8 Rae1 Bxe5 1 9
BxeS
Qc6 20 Bd4 Bd7
Black
behind the cathedral
doors.
21 Qc2 Rf7 2 2 Re3 b6
23
Rg3?1
A slip; Nimzowitsch for once
forgets about the iron logic of
overp rotectio n . Now Black
could have fought his way
bac'<
into the game with 23
. .. Ne4! , meeting 24 Rg4? or 24
Rh3? with 24 . . . es. And if 24
Bxe4 dxe4 Black is preparing
to play
.
. .
Qd5
and . . . e5 (al
though perhaps 25 Qd2 keeps
an edge). Finally, if 24 Re3
Black can of course repeat
with 24 . . . Nf6, while the pawn
sacrifice 24 . . . Qd6!? (25 f3 e5!)
looks promising; if 25 Bxe4
dxe4 26 Qxe4 RfS! and Black's
position is difficult to break.
The correct move is of course
23 Rfe 1 !
23
...
Kh8? 24 Bxh7
Since if 24 . . . Nxh 7 25 Qg6 win
ning. The rest :s simple mop
ping u p; Black is a pawn down
and his king is exposed.
24
.
.
.
e5 25 Bg6 Re7 26 Re1
Qd6 27 Be3 d4 28 Bg5
Black's attempts at counter
play have come to nothing; af
thoug h h e has gained space
in the center, his pieces are
poorly coord inated and his
king is exposed.
28 ... Rxc3 29 Rxc3 dxc3 30
Qxc3 Kg8 31 a3 Kf8 32 Bh4
Be8 3 3 Bf5 Qd4 34 Qxd4
exd4 35 Rxe7 Kxe7 36 Bd3
42
Kd6 37
Bxf6 gxf6
38
h4
Black
resigns.
We can see this ga;-ne that in
the fight against the White
pawn chain, it is not enough
simply to exchange a couple
of pairs of pawns, if White can
still keep control of the critical
squares
he keeps his posi
tional advantages. Naturally a
similar situation can arise in
the King's Indian; "every Rus
sian schoolboy knows that
Black must not take on e4 in
the King's Indian", but what are
the consequences? Again our
illustrative game is taken from
the history books, for the sim
ple reason that in modem tour
nament practice White is ge
nerally q�ite happy in the main
lines to play f3 at some stage,
allowing the bypassing move
. . . f4, and seeing his play on
the queenside. It is now more
fully appreciated that, fero
cious though Black's king-side
initiative may be, it is still pos
sible for White to win on the
queenside first.
Twenty o r thirty years ago
Black's standard kingside at
tacki-ng plan was much more
feared, and White often tried
to avoid playing f3, so as to
give Black no kingside weak
nesses to bite on. I n such
cases the e4 square often be
came the focus of st�uggle.
Twenty years a p a rt, both
Fischer and Kasparov have
ut:en great King's Indian aficio
nados, yet their games in this
opening look completely differ
ent. It is White who has chan
ged plan though, not Black.
Our next game is a gentle re
minder that Fischer was not
the only strong player in the
early 1 970s.
Ganae
6
Kerehnol
•
Geller
Itloseow
1970
(1st mat:eh game)
1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4
e4 d6 5 Be2 0-0 6 Nf3
eS
7
0-0 Nc6 8 d 5 Ne7
9 Nd2
Two games later Korchnoi was
to
try
9 Bd2, getting somewhat
the better of the draw. Tai�
manov, whe n being white
washed by Fischer, also tried
the bishop move, and twice
lost from better positions. It
might be interesting to follow
one of these games for a bit,
to see how opening was be
ing handled in 1 970.
43
Taimanov - Fischer {Vancou
ver, 3rd match game) contin
ued
9 Bd2 NeB 10 Rc1
(10
b4!? f5 1 1 Qb3 Nf6 1 2 exf5
gxf5 1 3 c5, Korchnoi - Geller,
3 rd game. Wh ite has h i s
q ueenside play, and Black's
kingside attack has had its
teeth drawn, but White has no
control of e4)
10 ... f5 11 Qb3
b6 12 exf5 gxf5 13 NgS Nf6
14 f4
N ow White is attacking the
Black pawn center! Black has
recaptured on f5 with his pawn
rather than a piece, so as
avoid giving White long term
control of the e4 square; this
h owever makes his pawn
structure a bit vulnerable. It
h as long been appreciated
that in such positions as that
i n the diagram, Black gains
very little in creating a pro
tected passed pawn with ... e4;
aft9r Nd 1-e3 the passed pawn
i s nicely blockaded, Black's
pawn structure lacks mobility,
and
White can gradually pre
pare for a break with g4. Even
so this might have been pref
erable to the game: 14 ... h6 1 5
fxe5! dxe5 1 6 c5l Nfxd5 (as 1 6
.. . hxg5 1 7 d6+ Kh8 1 8 dxe7
Qxe7 1 9 cxb6 followed by 20
Bxg5 is unpleasant) 1 7 Nxd5
Nxd5 1 8 cxb6 axb6 1 9 Rc6!
(the threat of Bc4 gives Black
no time to capture the knight)
1 9 . .. Kh8 and now White ner
vously retreated with 20 Nf3?
Bb7 21 Rg6 Nf4! with a big ad
vantage to Black. Instead 20
Qh3l N f6 21 Bc3 , making
maximum use of open lines,
would have been strong.
Among other strategic possi
bilities, it must not be forgot
ten that if enough central
points are keenly fought over
by both sides, a pawn chain
position can quickly become
an open position.
Back now to our main game.
9
...
aS!?
(diagram)
Geller introduces an interest
ing idea. Naturally Black must
aim to play . . .f5, and of course
the knight on f6 must move
first - but where?
H is intention is to place a
clamp on White's queenside
ac
ti\litu ""'" "',.,,..inn hie!
knl"ght on
1.1 •
''"� - J ,..,
,
.... �.,��I
llv
c5 and pawn on aS. This ef
fectively eliminates any possi
bility that White has of playing
for an early c5 break; the prob
lem is however one of time.
Black is using his king's knight
as though it were a q ueen's
knight, but what is he to do with
his quean's knight, currently
··on e7? Turn it into a ki ng's
knight by maneuvering
it
to f6?
Perhaps - the idea is less far
fetched than it seems, a s we
shall see later.
44
It would be much easier for
Black if he could have placed
a quean's knight on c5 (Na6-
c5 or Nbd7-c5, rather than
Nf6-d7-c5). And this explains
why- White usually waits for
Black to play ... Nc6 before he
plays d5. The sequence 1 d4
Nf6 2 c4 g6 3 Nc3 Bg7 4 e4
d6 5 Be2 0-0 6 Nf3 e5 7 d5
(the Petrosian System) is cer
tainly playable. but White will
have hard work to prove an
.
ad
vantage after 7
.
. . a5.
1 0 b3
Taking things quietly. A more
obviously thematic sequence
is 1 0 a3 Nd7 1 1 Rb1 f5 1 2 b4
Kh8 1 3 Qc2, and now a recent
try is 1 3 . . .fxe4 1 4 N cxe4 Nf5
1 5 Nb3 axb4 1 6 axb4 Nf6 1 7
Bd3 Nxe4 1 8 Bxe4 Q h4
"equal", Stohl - Marin, Stara
Zagora 1 990.
If this position is only equal,
does this mean that we can
consign to the dust bin of his
tory all Russian schoolboy les
sons? N ot quite ; the point
about this particular position is
that Black's grip on the d4
square counterbalances Whi
te's grip on the rather poorly
45
supported e4 square. White's
knight on b3 will never be able
to sit proudly on its favo rite
empty square; indeed it may
soon be exchanged after 1 9
Bd2 Bd7 20 Bc3 Nd4 2 1 Nxd4
exd4 22 Bd2 Rae8
.
It should
be added that in recent games
Whit� .. has preferred to play 1 3
f3 so as to recapture on e4
with a pawn.
In The French Defence, if the
d4 square is open , an impor
tant method of counterplay is
... Nc5-e4; correspondingly in
the King 's I ndian , if B la ck
should find himself forced to
leave the e4 square open, it is
often essential for him to pre
pare counterpiay with . . . Nf5-
d4. It is hoped that the reader
will appreciate that in many
such positions the King's In
dian is almost a mirror image
of the French.
1 0
...
Nd7 1 1 Ba3 NcS 1 2
b4
In the King's Indian it does n ot
matter so .. :nuch which lines
White opens up on the queen
side, so long as he opens up
some li nes . After the e x
change of pawns on b4 White
will concentrate his attention
on attacking the c5 squa re,
and
if
Black plays . . . b6, then
play for an a5 bre ak. 1 2 Bxc5
dxc5 v;ould ue meaningless;
White would h ave doubled
Black's pawns, but he would
also have cut out his chances
of making space for his pieces.
1 2 ... c:xb4 1 3 Bxb4 Na6 1 4
Ba3
b6 1 5
Nb3 fS
1
6 Bb2
1 6 .. .fxe4?!
The classic mistake in this type
of pos
·
ition. In his younger days
Geller was one of the great
pioneers of the King's Indian,
so
why
does he appear to ig
nore elementary principles?
When a strong player plays an
obviously anti-positional move,
it often means that he is trying
to squeeze something out of
the position that simply is not
there. In this case Geller was
presumably eyeing up the d4
square, but it turns out that
46
control of this square gives him
nothing
.
To play 1 6 . . . f4 would also be
a serious error with Bg4 still a
possibility for White (even
though not necessarily immP.
diately strong). After 1 7 a4 Nc5
1 8 Nxc5 bxc5 1 9 N b5 White
has a clear plus on the queen
side, while lack of influence on
the light squares makes it dif
ficult for Black to storm tha
kings ide.
T he correct plan is to attack
the base �f the pawn chain,
and this is done by 1 6 .
.
. KhB!
followed by . .. Ng8 and . .. Nf6.
Should White find himself de
fending with f3, then it is time
for the bypassing maneuver
.. .f4 followed by ... g5 etc.
··1 7 Nxe4 NfS 1 8 a4 NcS
19
NbxcS bxc5 20 Bd3
White's strongpoint o n e4 is
now secure, and given time
White will quite sit on Black,s
position. Black shou ld now
probably play an immediate 20
. . . Nd4 (analytical question 7:
what is then happening after
21 Bxd4 ?). Instead he plays
for counterplay on the king
side, a doomed effort given the
power radiated by the White
knight.
20
...
Bh6?1 21 a S Qe7?1 22
Ra3! Ba6 23 Re1 Qf7 24 Bf1
The resemblance to a French
Defence is b e coming ever
more apparent. Just as in the
French the ebb and flow of
play depends on how much
control White can keep on the
d4 square (Abramovic - Kova
cevic; Larsen - Bareyev), here
a lot depends on how much
grip White has on e4.
As you play through the rest
of the game, observe how
Black can make no progress
47
whatsoever on the kingside,
despite his massing of pieces
there . Yet if both players
knights could be removed from
the board, Black �auld have
the makings of quite a reason
able attack against the White
king (as in Abramovic - Kova
cevic).
Black's most immediate prob
lem is that White's quean's
rook is starting to look good;
therefore he attacks along the
b-file to force an exchange.
24
...
Rab8 25 Rb3 Rxb3 26
Q�b3 Nd4 27 Bxd4 exd4
Or
27
. . .
cxd4
28
c5 Bxf1
29
Rxf1 and there is not much to
stop the White a-pawn. I have
not stressed the importance of
the outside passed pawn in my
notes, since it can almost be
regarded as an incidental fea
ture of the position; if Black
can have an extra pawn in the
center, White is "entitled" to an
extra pawn elsewhere. Still,
the pawn must be respected.
28 Qb2
Avoiding the cheap threat of
. .. Bd2.
Ndf3 Qb6
The assault st�rts. An interest
ing alternative is 7 . . . cxd4
8
cxd4 f5! ? (blocking White's
activity on his stronger flank)
with the idea of . .. Nb6 followed
by . .. Kd7-c7-b8. White would
then have little to attack on the
kingside, whereas Black has
chances to create pressure on
the queenside.
8 h4
The pawn chain needs a lot of
protecting! 8 g3 is the usual
move, but 8 . . . cxd4 9 cxd4
Bb4+ 1 0 Kf2 g51 stops White
stabilizing the pawn chain; if
1 1 Be3 f6. We shall consider
all this later in the chapter.
8
...
cxd4 9 cxd4 Bb4+
10
Kf2
50
This rather awkward looking
move is forced because of the
weakness of the d4 pawn.
But White cannot have every
thing; the slight displacement
of the king is merely the price
that White has to pay for all
the time he has spent creat
ing his formidable pawn cen
ter. The important question is
what happens next. Quiet play
holds absolutely no future for
Black; he is cramped on the
kingside, his queenside is hard
to develop, and he has less
space in the center. White's d
pawn is under a bit of pres
sure, but this can soon be re
lieved. Black has only one re
alistic possibility - to explode
the position. Maybe then his
lead in development and the
exposure of the White king will
count for more than White's
space advantage.
1 0
. . .
f6
Now a capture on e5 is threat
ened, while 1 1 exf6? Nxf6
leaves White seriously weak
on the light squares. He must
break the pin on the g1-a7 di
agonal.
11
Kg3
Possibly 1 1 Be3 is safer, alt
hough. it gives Black a chance
to organize counterplay
against the pawn on b2, for
example 1 1 Be3 Be7 12 Qd2
0-0 1 3 Rd 1 a5 1 4 Kg3 a4 1 5
Bd3 Qd8 1 6 Ne2 f5 1 7 h5 Nb6
with a fully satisfactory position
for Black, Belyavsky - Kinder
mann, Munich 1 991 .
Plan
8
in action! White has de
veloped his pieces, but has not
improved his pawn structure.
Black meanv�hile has played
the blocking move ... f5 (com-
pare also the move ... c5 in the
King's Indian), which prevents
White's stand ard kingside
pawn storm. White would like
to open up the position with g4
but his king on g3 is in the way;
Kindermann in fact criticizes
White's 1 4 Kg3 in his notes,
preferri n g 1 4 Bd3 . On the
queenside Black has the ini
tiative thanks to his regroup-
51
ing with . . . Qd8 and ... NbS; now
. . . a3, . . . Nb4 and ... Nc4 are all
possibilities. After 1 8 Kf2 Nb4
1 9 Bb1 Nc4 20 Qc3 Bd7 2 1
a 3 Nc6 22 Rdg 1 b� 23 g4 b4
it was dear that Black's queen
side play was developi n g
much faster than White's king
side play.
A typical example of "full pawn
chain" play, with each player
in a blocked position pressing
hard on his stronger wing; the
King's Indian provides many
mirror images. But the Ljubo
j evic - Gurevich. game, to
which we now return, provides
a completely different resolu
tion to the central pawn ten
sion.
1 1
...
0-0 12 Bd3?
Definitely a mistake. Gurevich
gives as the only move 1 2
Nh3, and now 1 2 .. .fxe5 1 3
dxe5 (1 3 fxe5? Rxf3+ is a
strong exchange sacrifice, de
stroying the White center) and
now the position is assessed
as .. unclear". Exercise 9
-
cla
rify!
In the game something really
unpleasant now happens.
1 2 ... Nxd4! !
In the French Defence espe
cially,
it is always worth con
sidering sacrificing material
in order to destroy an oppos
ing pawn ch ain.
Often the
compensation lies in Black's
having a mobile central pawn
mass (particularly aft�r an
.
.
.fxe5); here the payoff is in
the form of open lines against
the White king.
1 3 Nxd4
Pe rh a p s Wh ite h a d ove r
looked 1 3 exf6 g6! 14 h5 Nf5+
1 5 Bxf5 gxf5. The sacrifice on
h7 h e re is com p l etely un
sound: 1 3 Bxh7+? Kxh7 1 4
Ng5+ fxg5 1 5 hxg5+ Kg8 1 6
Qh5 Nf5+.
1 3 ... fxe5 1 4 fxe5
52
On 14 Nde2 e4 followed by
. . .
Nf6
Black has his mobi le
central
pawn muss, while 1 4
Nc2
exf4+ 1 5 Bxf4 Bd6 1 6
Bxd6 Qxd6+ 17 Kh3 eS is not
too happy for White either. fi
nally,
14 Qh5? e4 is hopeless.
14 . . . Nxe� 1 5 Bc2
The
bad positioning of the
White king crops u p in tactic
after tactic, for exampl e 1 5
Bxh7+? Kxh7 16 Qh5+ Kg817
Qxe5 Bd6 winning the queen.
1 5 ... Ng6!
A useful move which blocks off
any counterattack along the
b1 -h7 diagonal, prevents Bf4,
clears the b8-h2 diagonal, and
··
leaves the Black e-pawn free
to advance, thereby opening
up a diagonal for the other
bishop. 16 ... Bd6+ 1 7 Kh3 e5+
is now a big threat.
1 6 Bxg6 hxg6 1 7 Nde2
Horrible, but how else is White
to save his extra piece? If 1 7
Ndf3-Bd6+ 18 Kh3
Q
f
2
1 9 Qe1
e5+
20
g4 Rxf3+ wins.
1 7
.
.
.
Qf2+ 1 8 Kh3 Bd6
_:....
humiliating position for any
grandmaster to reach! Rarely
has a pawn chain bee�
d�
stroyed with such complete
brutality. Black now threatens
a
mating attack with 1 9 . . . e5+
20
g4 Bxg4+ 21 Kxg4 Qf5+ 22
Kg3 e4+ 23 Kg2 Qg4+.
19
Qb3 e5+ 20 Kh2 Qxh4+ 21
Nh3 Bxh3
White resigns.
The endgame after 22 Qxh3
e4+ 23 g3 Rf2+ 24 Kg·1 Qxh3
25 Rxh3 Rxe2 is not worth
playing.
Quite often Black is forced to
react even more violently on
the kingside in order to break
up the d4-e5-f4 pawn wedge.
Not infrequently this early ac
tivity backfires completely. My
own advice to players with
31ack would be to forget about
destruction of the White's spa
�e advantage on the kingside,
53
and then to concentrate on
queenside play. The more vio
lent lines are interesting, but
so many games are unneces
sarily lost because· one of the
p!,�'�rs seeks to do something
.. interesting". Here is one, per
haps.
Game
8
Pyda
•
Llkavsk,-
•
£zeehoslcn'akla
199 1
1 e4 e6 2 d4 d5 3 Nd2 Nf6 4
e5 Nfd7 5 f4 c5 6 c3 Nc6 7
Ndf3 cxd4 8 cxd4 Qb6 9 g 3
Bb4+ 1 0 Kf2 g5!?
Boldly pl ayed , indeed o ne
might possibly suggest over
boldly played? What is Black
trying to do, given that . . . g xf4
will not even break the White
pawn center? Black can o pen
up the g-file, certainly, but if
White develops his bishop to
h3, the one important e ntry
square, g4, will be safely cov
ered. The answer is that Black
wants to play . . . g5 and . . .f6,
pre pari n g with two sets of
pawn exchanges to rip open
lines against the White king.
But where does the Black king
go in all this? One cannot ex
pect to be able to open up the
position so violently and then
be allowed to play a few quiet
m oves to tuck his majesty
away on the queenside. Nei
ther is the king safe in the cen
ter. Surely he cannot be con
templating kingside castling?
Just wait and see!
1
0
.
. .f6 does not look too he
althy after 1 1 Kg2 0-0 12 Bd3,
while 1 0 . . .f5, the blocking
move. leads to loss of tempo
after the inevitable regrouping
maneuvers with ... QdB, ... Nb6,
. . . Be7, etc.
1 1 Be3
1 1 fxg5 is also perfectly play
able; Black removes the e5
pawn
,
but White gains the e5
square
in standard French
style. O'Donnell - Foisor, Saint
J o h n 1 9 88 , conti nued 1 1
. . . Ndxe5 1 2 Nxe5 Nxe5 1 3
Kg2 Nc6 1 4 Nf3 Bf8 1 5 b3 !
Bg7 1 6 Bb2 Bd7 1 7 Qd2 wit
advantage to White.
White has no real intention
here of occupying e5; the key
to the position is that Black's
bishop on d7, stuck behind the
Black pawn chain, is extremely
passive, and Black is gasping
for the chance to free it with
. . . e5. So long as White has
good dark square control, this
is not possible, and Black must
seek counterplay .. else�here.
·· In the game cited, Black tri�d
to open up the queenside, but
as so often when one's posi
tional vulnerability rather than
one's security forces one to
open up lines, it was the op
ponent who was the ben efi
ciary: 1 7 . . . a5 1 8 Be2 a4 .1 9
Bc3 h6 20 bxa4 Rxa4 21 Rab1
with pressure; note the tactic
21 . . . Qa7 22 Rb2 Nxd4? 23
gxh6 Bxh6 24 Bxd4.
54
All this tempting , but Black
may well be able to improve
(1 3 ... Nc4 can also be consid
ered). With 1 3 Be3 however,
White gives Black the chance
to destroy his own king�ide.
1 �
.
.
.
f6
Professional players will be
well familiar with the concept
of the "ghost variation": a move
is tried once, commended in
pri nt, and then never seen
again. Neither examples nor
refutation are published. What
usually happens in such cases
is that a refutation is found,
perhaps by every master who
looks closely at the position,
but is not published. Indeed,
what possible incentive can
there be to publish? Much bet
ter to win an easy game armed
with pre-match analysis, and
only later to publish.
One such ghost variation is 1 1
.
.
. g4, played i n Jakovich -
Machulsky, USSR 1 985. Our
tenth reader's exercise is to
ask you to find out why this
move is no longer played.
Jakovich gives 12 Nd2 f6 and
now:
(I) 1 3 Qxg4 Bxd2 1 4 Bxd2
Qxd4+ winning for Black.
(II) 1 3 a3 fxe5 1 4 fxe5
Rf8+
55
1 5 Kg2 Be7 16 Qxg4
Ndxe5 and Black, having
successfully broken the
White center, stands
better.
(Ill) 1 3 Nb3 fxe5 1 4 dxe5 Bc5
1 5 NxcS Nxc5 1 6 Bg2?
(16 Bxc5 QxcS+ gives
chances for both sides)
16 ... d4! 1 7 Bxd4 (1 7 Bc1
hS with advantage to
Black was the game) 1 7
. .. Nxd4 1 8 Qxd4 Nd3+ 1 9
Ke3 Qxd4+ 20 Kxd4 Nf2
winning material.
1 2 Bh3
An explosive position! Some
th ing drastic must happen
soon. In some of the earliest
games in this line, Black tried
to keep the lid on with 1 2 . . . hS,
but the piece sacrifice 13 Qd3!
g4 14 Qg6+ KdB 1 5 exf6 looks
da;,gerous.
Black's only real option is to
castle and attack down the f
file; the question is whether he
wants to preface this with an
exchange of pawns. The im
mediate 1 2
.
. . 0-0 used to be
quite popular, and there are
many tactical tricks; a game
Dobrovolsky - Tibensky, Cze
choslovakia 1 988, showed
however that with simple de
velopment ( !) White can con
solidate his center and main
tain an edge: 1 3 Bxe6+ Khtj
1 4 Ne2 ! fxe5 1 5 Nxg5! exf4 1 6
Nxf4 Nf6- 1 7 Kg2. N o quiver
ing tension, just a safe edge
to White. So:
1 2 . .. fxe5 1 3 fxe5 0-0
in such positions, Bxe6+ is
rarely especially dangerous for
Black, who usually has freeing
combinations in reserve with
. . . N cxe5 or . . . Ndxe5 or . . . Nf6.
14
Rc1
Designed to sidestep a sur
prising tactical resource found
by Botterill: 1 4 Bg4 BcSI I 1 5
Bxe6+ Kh8 1 6 dxcS Qxb2+ 1 7
Bd2 g4! 1 8 Bxd5 (After 1 8
Bxg4 Qd4+ Black regains his
pi ece with a slightly better
56
game) 1 8 . . . Ndxe5 1 9 Bxc6
bxc6 20 Rb1 Qd4+ 2 1 Be3
Nd3+ 22 Ke2 gxf3+ 23 Nxf3
Qe4 24 Qxd3 Ba6! 25 Qxa6
Qxf3+ 2o Kd2 Rfd8+ 27 Kc2
Qe4+ 28 Kb2 Rab8+ 29 Ka3
uxej+
�u
Rb3 Qxc5+ 31 Kb2
Rd2+ 32 Kb 1 Rxb3+ 0-1
Emms - Kosten, British Cham
pionship 1 985. The impression
is however given that Black is
using a string of tactics to hold
a fundamentally dubious posi
tion together. If White can
avoid such tricks then presum
ably he is better.
1 4
. . .
Kh8 1 5 Bg4
Better timing ! White did not
have this move order option in
Emms - Kosten which went 12
. .
. 0-0 13 Bg4 and only now a
pawn exchange on e5.
1 5
.
.
.
Be7
It is now time
to
take stock of
the position . White's pawn
cente r re mains u n b roke n ,
which must
of
course please
him.
There are however va rious
tactical blows he must watch
o u t fo r, mostly b a s e d on
. . . N dxe5 and j u sti fi e d by
White's weaknesses o n the
g 1 -a7 diagonal and the f-file.
T h e b2 pawn also n eeds
watching . Black is therefore
not without counterplay, and
therefore his decision to open
up the king side cannot be dis
missed too badly, except for
that woeful pawn on g5, the
misbegotten child of Black's
king side lust. It is now
White's
turn to open up the kingside.
1 6 h41 gxh4 1 7 N h31
The point. Black is to be en
couraged to open up the h-file
against himself.
1 7
. . .
hxg3+?!
But this seems unnecessarily
cooperative. Pyda gives the
variation 1 7 . . . Qxb2+ 1 8 Rc2
Qa3 1 9 Qe2 hxg3+ 20 Kxg3
wi th a dvanta g e to Wh ite.
Again though there is no obvi-
57
ous need for Black to cooper
ate· exercise 1 1 - can Black im-
'
prove by playing
1 9
. . . Rg8
here?
1 8 Kxg3 Ndxe5
1 9 Nhg51
Black's little freeing combina
tion does 'lot work. The king
side, Wh1te's favoured zone of
operations, is fully open, whe
reas Black has succeeded in
only partially opening the cen
ter, and his queenside pieces
are still locked away.
Such a position casts doubt on
the wisdom of hurling pawns
forward on the kingside, your
weaker wing, just before cas-
..
tling there.
If now
1 9
.
.
.
Nxf3 20 Rxh7+
Kg8 and White llmust" be win
ning; this is so obvious that
Pyda does not even give a
variation! Not everything is so
simple though; 21 Rh8+ is met
by 2 1 . . . Kg7! , while there are
possibi liti es of an irritating
Black counterattdck with . . .
Bd6+ or . . . Qc7+.
S o h ow does White wi n?
There is a clear method, but it
is not easy to find; readers
exercise number 12!
1 9
. . .
Rxf3+ 20 Bxf3 Bxg5
21
B xgS Ng6 22 Bf6+ Kg8
The first wave of the attack has
been beaten back, and indeed
it would appear that Black
d oe s not stand too badly.
Given the time for . . . Qc7+ fol
l owed perh a p s by . . . Qf7 ,
. . . Bd7 and . . . Rf8, he might
eve n stand rather well. But
now comes the second wave
of the White attack, which re
moves all hope.
23 Rxh71 Kxh7 24 Be411 Qc7+
25 Kg2
58
The more obvious point to
White's bishtlp sacrifice is to
clear the way for Qh5+; the
less obvious point is that the
knight on g6 is pinned, so that
Black has no real counterat
tack, . . . Nf4+ being
ill.s�z.:. -;-�.:
rest is straightforward, as if 25
. . . Qf7 26 Qh5+ KgB 27 Bxg6
Qxf6 28 Rf1 Qg7 29 Rf7.
25 . . . Kg8 26 Qg4 Qf7 2 7
Qxg6+
Liquidating into an easily won
endgame.
27 ... Qxg6 28 Bxg6 Kf8 29
R'l1 Ne7 30 Rh8+ Ng8 31 Be5
Bd7 32 Rh7 Be8 33 Bd6+ Ne7
34 Bxe7+ Kg8 35 Bb� ReS 3 6
BgS Rc
�
37 Be3 e5 3 8 dxe5
d4 39 Bd2 d3 40 Re7 Bc6+
41 Kf2 Rh4 42 Bxd3
Black
resigns.
It is still an open question
whether White can maintain a
meaningful space advantage
with the f4 syste m in th e
Tarrasch. What he is doing is
to sacrifice time to gain space,
and this loss of time g ives
Black the opportunity to ar
range _counterplay.
Unfortunately, Black's search
for counterplay can itself im
pede the proper development
of his pieces. To attack the
pawn on d4 Black must .play
his queen to b6; to develop his
queenside he must play . . . Nb6
and . . . Bd7. Alas, double occu
pation of b6 is not allowed, and
so it becomes difficult for Black
to develop his queenside.
White will probably not have
enough time
to
keep his pawn
wedge intact. but he is quite
likely to be able to attack on
the kingside with. pieces. Play
is sharp.
There is another way for White
to handle the ... Nf6 Tarrasch;
h e can say " Sq u a res n ot
pawns" and can concentrate
o n d eve lopin g his p iece s ,
rather than rolling his pawns
forward. It still remains impor
tant for Black to break up the
White pawn center, and so
. . .f6 will tend to be played. Af
ter the exchange of pawns on
f6 however, White can play for
co ntro l of th e e5 s q u are .
A common opening sequence
is
1 e4 e6 2 d4 d5 3 Nd2 Nf6 4
e5 Nfd7 5 c3 c5 6 Bd3 Nc6 7
Ne2 cxd4 8 cxd4 f6
9
exf6
Nxf6,
leading to the diagram
position.
59
White's knights may look at
first strangely placed. but Nf3
soon comes.
The pawn structure is interest
ing; both sides have their prob
lems and their opportunities.
White'·s d-pawn is dearly weak
and will need to be defended
by pieces. At the same time
the pawn is very useful as it
controls an important central
square (e5), and also restrains
two Black pawns (dS,
e6).
The
pawn on e6 is a nuisance for
Black in that it prevents the
free development of the light
squared bishop; on the other
hand the absence of the
�
I
pawn gives Black chances of
real counterplay down the f
file.
U ntil fa irly recently, White
aimed to keep pieces on the
board in this variation. to keep
things complicated, but Black
seems to have perfectly satis-
factory pl ay. This approach
has more or less been aban
doned ; White now prefers to
play an early Bf4, exchanging
bishops and attempting to
keep p re s s u re o n the e5
square {"squares not pawns").
We consider a game with each
approach.
Game
9
SIDagln
•
Valser
Bo rnaul
1984
1 e4 e6 2 d4 dS 3 Nd2 Nf6 4
eS Nfd7 5 Bd3 cS 6 c3 Bc6 7
Ne2 cxd4 8 cxd4 f6 9 exf6
Nxf6 1
0
Nf3 Bd6 1 1 0-0
1 1 Bf4 is now preferred.
1 1
. . .
Qc7 1 2 Nc3 a6 1 3 BgS
0-0 1 4 Bh4 Nh51
Korchnoi's move, which takes
away much of the fun from
White
.
f4 is a pleasant square
60
for Black to aim for with his
knight, while the open f-file
adds a little bit of pressure.
Black also has in mind the
maneuver . . . g6 and . . . Qg7, pil
ing more pressure
011_
the d
pawn.
It is i nteresting to note that
most of Black's play is c;>n the
dark squares, indicating that
perhaps White has not been
very successful in carrying
through his pawn chain strat
egy (pawns then squares) ,
since the theoretically weak
squares, notably e5, are well
covered. One cannot go much
further than to say "perhaps",
since there are stiil some un
solved problem with the Black
position, notably the role of the
queen's bishop; any sudden
simplificatio n of the po.sition
will generally favour White.
1 5 Bg3?1
The very move that Black had
supposedly prevented ! The
exchange of bishop for bishop
would have made good posi
tional sense for Whi te, as
Black starts to lose control of
the vital e5 square. Exchang
ing his bishop for the Black
knight makes less sense how-
ever; the knight on h5 is a use
ful attacking pieces, but not a
linchpin of the Black position;
the White bishop however is
unquestionably strategically
important.
In any position where strong
and weak squares are impor
tant (and this is particularly
likely where there are decayed
p awn ch a i n s) ,
g re at ca re
must be taken to distinguish
between pieces which are
strategically important and
those which are merely ac
tive .
If
o ne c a n remove a
strategically important piece
of the opponent's in return
for a merely active piece of
one's own, that
is
real posi
tional gain.
How can one tell
if a piece is strategically impor
tant? There is no universal
rule, but if a piece is actively
engaged, or likely in future to
be actively engaged, in the
struggle for a key square, then
its strategical role must be re
spected.
What then is the correct pro
cedure from the diagram po
sition? 1 5 Re1 is sensible, met
by 1 5 . . . g6. Black's position
looks perfectly playable, as the
following examples show:
a) 1 6 Rc1 Qg7
1 7 Bf1 Bd7
1 8
6 1
Bg5 h6 1 9 Be3, and now
1 9
. .
. Rf7 20 g3 Raf8 21 Bg2
was roughly equal in Karpov
- Mestel, London 1 984,
1 9
. . .
Nf4! 20 g3 g5 is better
for
Black though; he has made
light of his weakness on e5
and is starting to gain gro
und on the kingside.
b) 1 6 Bf1 h6 1 7 Bg3 (1 7 Qd3
Nf41) 1 7 ... Nxg3 1 8 hxg3
g5 1 9 Na4 Bd7 20 Rc1
Rf6 21 Nc5 Raf8 22 a3
Bxc5 23 Rxc5 Be8 24
Be2 Qg7 25 Rf1 h5 and
again · Black is looking
good, Ye Jiangchuang -
Belyavsky, Lucerne 1 985.
Even the uad bishop is
starting to see the light!
c) 1 6 Bg5 led to a snappy
draw in Kruppa - Dreyev,
USSR 1 985: 1 6 . . . Nxd4!
1 7 Nxd4 Bxh2+ 1 8 Kh1
Rxf2 1 9 Nf3 Ng3+ 20
Kxh2 Ne4+ 21 Kh1
Qg3 22 Bf1 Rxg2 23 Bxg2
Nf2+ �4 Kg1 Nh3+ with a
perpetual. Always Black
has active piece play to
compensate for his we-
akened pawn structure.
1 5
. . .
Nxg3 1 6 hxg3 g6 1 7
Rc1
Qg7 1 8 Bb1 g 5 1 9 Re1
Bd7
Successfully completing his
development, and avoiding
the impulsive 1 9 .. . g4? 20 Ne5
Bxe5 2 1 dxe5 Nxe5? 22 NxdS!
It is dear that Black is already
better. In Kudrin - Korchnoi,
Beersh eva 1 984, Black won
very quickly by piling down the
f-file after 20 Qd3 Rf7 21 Red 1
Raf8 22 Rd2 Kh8 23 Nd1 Bc7
24 Qe3?? (falling into a stan
dard combination, but game
was already highly unpleas
ant) 24 . . . Rxf3 25 gxf3 Bf4
White resigned. The move that
White plays is no improve
ment.
Maybe 20 Ne5 has to be tried.
20 Qd2?1
20 ... Rxf31
Thoroughly thematic.
62
21 gxf3
Nxd4
Black has a big advantage. He
has destroyed the remains of
the White pawn center, and he
has seriously weakened the
White kingside. Furthermore,
now that Black has the center,
his minor pieces become po
tentially very active. The fine
position of knight is clear to
see, but may be in a few mo
ves time the knight will want
to trot away, leaving room for
the d- p awn to advance a
square, thereby giving a won
derful diagonal to the light
squared bishop, no longer a
bad bishop, more a saint. For
all this, Black. has sacrificed no
more than the exchange for
a
pawn ; such li9.ht sacrifices
(piece for two pawns is al�o
com.mon) are important d e
vices in the French for Black
to gain central control.
22 Kg2
22 Re3 fails to 22 ... Bf41 (re
member the Kudrin - Korchnoi
game?). Vaiser gives 23 gxf4
gxf4+ 24 Kh1 1 Qh6+ 25 Kg2
Qg5+
2.6
Kh 1 (26 Kf1 e5!) 26
. . . Qh4+ 27 Kg2 fxe3 28 fxe3
Qh2+1 liquidating into a win-
ning endgame.
22
.
.
. Rf8
2 2
.
. . Bf4? g ives White a
breather after 23 Bxh7+! Kxh7
24
Qd3+.
23
Rh1 h6 24 Qd1
Looks passive, but White has
a tactical defence in mind. As
Vaiser points out, 24 Qd3 is
well met by 24 ... Rxf3
25
C�e
Qxg6 26 Bxg6 Kg7 wh en
Black has a second pawn for
the exchange, and his bishop
pair and pawn center augur
well for the endgame. 24 Rxt)6
is tri cky, but agai n Black
comes o ut on top after 24
... Nxf3 25 Bh7+ Kh81 26 Be4+
(26 Qd3 Qxh6 27 Rh1 Kg7!)
26 . . . Qxh6.
24
. . . Nxf3 25 Nxd5
63
Fully prepared to meet 2 5
. . . Bc6? with 26 Rxc6
I
How
ever ...
25
.
..
Nh4+11
C��n
lines! Open lines!
If now 26 Kg1 Black gains an
important tempo with 26 ... Qf7,
e.g. 27 Qc2 Nf3+ 28 Kg2 exd5
29 Rxh6 Nh4+.
The text is embarrassing to
White though.
26 gxh4 gxh4+ 27 Kf1 Bb5+
Any French Defence player
would enjoy this movel lf now
28 Bd3' Qg3 wins.
28 Ke1 Bb4+
As if 29 Nxb4 Qe5+ wins. But
28 . . . Qg2! would have forced
immediate resignation.
29 Rc3 exd5 30 Qxd5+ Kh8
31 Kd1 Bxc3 32 bxc3 Bc6
White resigns.
Qu ite naturally, illustrative
games tend to exaggerate the
strength and weaknesses of
particular types of play; if they
did not, there would be no il
lustration. White's kingside is
n ot obl iged to cru mble if he
avoids exchanges in the Bd3
Ta rrasch , but it ia d ifficult for
h im to keep a ny real grip on
the position . Maybe if the dark
squared bishops a re removed ,
t h e g a m e i s b ri g h te r fo r
W h ite . . .
Game 1 0
E rost
•
Cro ueb
Loodoo 1 99 1
1
e 4 e 6
2 d4 d S
3
Nd2 Nf6 4
eS
Nfd7
5 Bd
3
cS 6 c3 Nc6
7
Ne2
cx
d4 8 cxd4 f6 9 exf6
N xf6 1 0 0-0
Bd6 1 1
Nf3 0-0
1 2
Bf4
B lack cou ld h ave prevented
th is with 1 1 . . . Qc7 , but if \f\!hite
is i nsisten t he ca n play 1 2 g 3
0-0
1 3 Bf4 .
Since dark sq uare control is so
importa nt for White , g3 may be
regarded as being a strength
e ning move as much as weak
e n ing move .
A
si mpler move o rder to reach
the ga me position is
1 0
Nf3
Bd6 1 1 Bf4 Bxf4 1 2 Nxf4
0-0
1 3
0-0
1 2
. . .
Bxf4
13
Nxf4
(diag ram)
1 3
.
.
.
Ng4
64
In such positions Black must
a lways aim fo r co u n te rp lay,
nthP.rwi!";P.
his weak e5 square ,
ba ckwa rd e-pawn a nd bad
bishop will continue to bu rden
him for a long time to come.
Yet if he plays overactively, try
ing to dru m up a n attack which
isn't objectively there , he faces
a rout. To strike the ha ppy me
dium is the task. not a lways
easy.
1 3
. .
.
Qb6
- is
a
depend
able enough move, althoug h
one mig ht feel that the q ueen
ought to be doing more to in
fluence eve nts on the king
side .
1 3 . . . Ne4 is a sharper p la n ,
with an excha nge sacrifice in
mind . After
1 3
.
.
. N e4 , 1 4 g 3 is
a safe enough reply. since the
n a tu ra l 1 4 . . . Qf6 ? 1 5 h 4 !
leaves Black without a good
reply. Attentio n h a s con cen
trated however on 1 4 Ne2 ! ?
Rxf3 (again the sta ndard sac-
rifice) 1 5 gxf3 Ng5. Black has
compensation, but probably
not enough for e�uality, for ex
ample 16 Kh 1 e5 1 7 dxe5 Nxf3
1 8 Bxh7+ Kh8 1 9 Ng1 Ncd4
20 Nxf3 Bg4 21 Nxd4 Bxd 1 22
Rad1 Kxh7 23 Rd3! Qb6 24
b3 Rca 25 f4 with some advan
tage to White, Timoshenko -
Gleizerov, Chelyabinsk 89/90.
An interesting position. Black,
in order to avoid a nagging po
sitional disadvantage, has
�sed his temporary superior
piece mobility in order to pro
voke complications. These
complications seemingly tran
sform the position completely,
yet when everything has sub
sided White still has the mak
ings of typical French (or Sicil
ian) type positional advantage.
Note for example t_he roles of
the spearhead pawn on e5,
the strong blockading knight
on d4 and the open lines for
� - Pawn Chains
65
the rooks on the king side. The
Black queen had to perform
acrobatics {25 . . . Qg6! 26 Re3
Qb6 27 Rd 1 Qh6! 28 Rf3 Qg6!
29 Re1 Qb6 30 Rd3 Qb4 3 1
Rdd1 Qc3) for Black to �tay in
the game.
Please excuse the author for
choosing a rather less interest
ing game (and one of his own,
as we ll) as the i ll us trative
game, but this is the last chap
ter on the French, and I wan
ted to show a typical positional
struggle revolving around the
bad bishop.
1 4 Qd2
1 4 g3 g5 1 5 Ng2 Qf6 1 6 Be2
N h6 1 7 Qd2 Nf5 1 8 Rad 1 g4
1 9 Ne5 Ncxd4 20 Nxg4 Qg7
.with chances for both sides,
Renet
· -
Hertneck, Altensteig
1 987. A strategy we have al
ready seen in the line without
the bishop swap; Black ex
pands on the king side to cover
over his weakness on e5.
1 4
. . .
Qf6 1 5 Nh5 QhS
No choice really, but White is
happy to be moving closer to
an endgame. Still, p rovided
Black is determined not to wait
passively and let White show
his technical skills, his position
should hold together.
1 6 Qxh6 Nxh6 1 7 Na3?1
-
Overmaneuvering a bit,
1 7
Rfe 1 a6 1 8 Rac1 is slightly
better for White, one point that
. . . Nf7 can
in
many cases be
met by Nf4.
1 7
. . •
aSI
A
necessary defensive move.
If 1 7 . . . Bd7? 1 8 Bb5! followed
by Bxc6 and White is gainir.g
ground, having exchanged
a
piece that is merely active (the
bishop on d3) for one that is
strategically important (the
knight on c6). In the positions
which result, the Black bishop
is a dreadful piece which can
attack nothing and which can
do nothing to cover the weak
dark squares. Its only Lon-
6B
structive role is to defend the
pawn on e6, and this is a laug
hably small role. Almost any
textbook on the endgame will
give an example of how the
bad bishop can be made to
look very silly by a n active
knight.
Anyone who plays the French
Defence will almost inevitably
be occasionally called upon to
defend a bad bisho p end
game. The defensive tech
nique to remember is to keep
your knights, and to keep them
active. In the diagram position
Black's plan is to develop his
q u eenside, ce ntra l ize h is
knight on h6 ( ... Nf7-d6), and
then to start counterplay on the
queenside. In an otherwise
slightly depressing position,
one small detail encourages
Black; Wh ite's d - p awn i s
weaker than Black's . This fea
ture harks back to the earlier
struggle against the base of
the pawn chain (
. . . cS ; . .
. cxd4) .
1 8 Rad1 Bd7 1 9 Rfe 1 Nf7 20
Ne2 NdS 21 Nc3 Rae8 22 Ne5
White can hard l y h o pe to
make progress without this
move.
22 ... Nxe5 23 Rxe5
There is a case for 23 dxe5,
which would be rather st
r
o
ng
if White could get a knight to
d4. True, Black has a passed
pawn
,
but it is extremely well
blockaded, and has little im
mediate impact on the game.
How does Black avoid gettir:'g
into a dismal endgame? The
first step is to put the e5 pawn
under pressure, thus 23
.
.
.
Nf7;
note that this inhibits White's
Ne2 plan. The obvious reply is
24 f4, but Black can attack
White's kingside pawn chain
with 24
. .
. g5! 25 fxgS Nxg 5 26
Ne2 Re7 27 Nd4 Rg7; some
fairly standard pawn chain
strategy. White can try antici
pating Black's ... g5 with 24 h4,
but this gives Black time to
change track with 24
. . .
NdB
and
. .
. Nc6. Finally if White tries
24 Bb1 Black plays for pres
sure along the c-file with 24
. . . R
e
B
.
With care, Black should be
OK.
23
. . .
g6
(diagram)
Black's modest pawn mo
v
e is
actually a bid for the initi
a
ti
v
e!
In the next few moves Black
67
will set the pace of the game.
The tactical point is that Black
no longer has to worrv about
Bxh7, so that . . . Rf4 is going to
be a genuine threat to the d
pawn (24 h4? Rf4!). In addi
tion, a · route to the center is
given to the Black king ( ... Kg7-
f6). If one adds that the White
knight is tied to c3 because of
the danger of Bb5, it should be
clear that White has no real
edge.
24
g3
Playing for f4 .
24
. . .
b5
Playing for Nc4.
25 b3 Rf31
An unexpected weakness ap
pears in the White position
(RR - 26 Ne2? Nf7 27 Re3
Rxe3 28 fxe3 e5 is well for
Black).
26 Kg2 RefS 27 Rd2
27
. . .
Nc41
After which White is the one
who has to defend. Not how
ever 27 . . . b4?1 28
Nd1
Bb5 (28
. . . N bS 29 BxbS) 29 Bb1
t
and
Black has overplayed his at
tack a bit.
28 bxc4 dxc4
Allows a counter-tactic, but 28
. . . bxc4 29 Re3! leaves White
in control.
29 Bxc41 bxc4 30 Re3 Bc6
Black has no more bad bishop
problems, and indeed has the
more bad active pieces, plus
a passed c-pawn.
68
However this passed pawn is
itself weak, and it is this factor
which prevents Black pressing
for a win. The equilibrium is
soon reached .
31 Kg1
Rxe3
32 fxe3 Rf3 33
Re2
Kg7 34 Re1 h5 35 Nd1 !
Suddenly White's defensive
plan becomes clear. The ex
change of rooks with Rf1 is
threatened . and if this ex
change is allowed the knight
versus bishop endgame is
very good for White.
I n
such a
position as would be reached,
the bishop would be inferior to
the knight, not because it has
nothing to attack.
Rather than endure this, Black
tries to swing his rook to the
queens ide .. .
35
. . . Rf5
White parries .. .
36 Nc3
Black returns .. .
36
�
.
.
Rf3
and the draw is agreed .. .
37
Nd1
Rf5
i2
:
% .
5.
The Unbroken Chain
As we have seen, the setting
up of the pawn chain is an im
p
ortant part of White's strategy
1n the French defence, while
the demolitio n of the pawn
chain is an equally important
component of Black's strategy.
Indeed, French middlegame
�
trategy (as opposed to open
Ing strategy) is primarily con
cerned with the implications of
decayed pawn chains; some
of the pawns which originally
formed the chain remain in
place, whereas other will have
disappeared. leaving vacant
squares for White to use (d4
and/or e5), and open lines for
Black to u se (c-fqe , f-fi le ,
sometimes the h2-b8 diago
nal).
By contrast, pawn chains can
have a very long life in the
King's Indi a n , especially if
Black follows the standard by
passing strategy with .. .f4. This
strategy is usually justified only
by the uncomfortable situation
of the castle d White king,
stuck in front of an advancing
pawn roller; otherwise, as in
the French, it is best for Black
69
to maintain the tension . We
have already seen an example
(Korchnoi - Kasparov) of Ka
sparov successfully ramming
his kingside attack home after
an early . . . f4; as
a
refresher
I
here is another sample of the
Kasparov maneuver.
Gan1e
I I
P i ket
•
Kusparo''
Tllburg
1989
1 d4 Nf6
2 Nf3 g6 3 c4
B g7
4
Nc3
0-0 5
e4 d6 6 Be2
e5
7
0-0
N
c
6
8 d5 Ne7
9
N
e1 Nd7
1 0
8e3
fS
1 1 f3 f4 1 2 Bf2
gS
1 3 b4
In our Korchnoi
·_
Kasparov
gam e , Wh ite tri ed 1 3 a4.
Korchnoi has also tried the
strange-looking 13 NbS a6 14
Na7,
the idea being that if the
light squared bishop on c8 is
exchanged, Black loses con
trol of the g4 square and thus
cannot roll his king side pawns
forward, while White still has
everything open on the queen
side.
The tactical justification is that
White can survive 1 3 . . . a6 1 4
Na7 Rxa7 1 5 Bxa7 b6: thus 1 6
b4
Bb7 1 7
c5 dxc5 1 8 Rc1 !
NcB
1 9
bxc5 BaS (19
. .
. Nxc5
20 Rxc5!) 20 c6 Nf6
2 1
Bxb6
Nxb6 22 Bxa6 and White has
made ma�sive gains on the
queenside before Black has
started his kingside counterat
tack, Korchnoi - Hulak, Zagreb
1 987. In that game, Black now
completely mistimed his play,
and after 22 . . . g4? ! 23 Nd3 g3
24 h3! his resistance soon fol
ded.
A wonderfully imaginative line.
Sadly, it soon had to be aban
doned when it became clear
that after 1 3 . .. b6! 1 4 b4 a6 it
is not really worth the two
tempoes to provoke the weak
ening of the queen side. Black
just builds· up on the kingside,
as usual.
1 3 . .. Nf6 1 4 c5 Ng6 1 5 cxd6
cxd6 1 6 Rc1 Rf71
A useful defensive move.
White has made progress on
the queenside, and has
70
opened up the c-file , which
exposes various Black weak
nesses on the dark squares.
These weaknesses must be
covered, at least temporarily,
since Black is not quite ready
yet to play . . . g4. In particular,
Black needs to prevent an in
vasion on c7 after White's
NbS, while he must also bear
in mind that the passive 1 6
. . . a6? would leave a disabling
weakness on b6 (17 Na4 etc.).
Any exchange of knight for
light squared bishop would
tend to favour White, who has
his . .. g4 anxieties significantly
reduced.
The point about ... Rf7 is that it
covers the weak square on c7
efficiently.
The next stage in
Black's defensive maneuver is
to play . . . Bf8, coverin
g
the d6
·pawn , and then after . . . h5,
. . . g4, etc. , the rook is free to
move to the g- or h-files,
while
still covering c7.
Thus Black
is not defending purely pas
sively; he has eyes on attack
as well as defence.
White has also deployed his
pieces carefully. We have al
ready seen a broadly similar
formation in the game Larsen
- Torre, given in chapter 1 in
the notes to Korchnoi - Ka-
sparov. There however White
declined to play his rook to the
c-file, preferring instead to ad
vance his pawn. The result
was that when White finally
played Nb5 it was harmless;
there was no bite on c 7, and
Black could nonchalantly roll
his pawn on the kingside.
1 7 a4
Analytical exercise 1 3: what do
you think of 1 7 Nb5?
1 7
. . .
Bf811
Truly a World Champion mo
ve! Most of us would have got
on with our king side attack with
1 7
. . .
h5, etc. Kasparov shows
that
. . .
hS is
not
even neces-
sary; note that the h-pawn re
mains on h7 for the rest of the
game, and that Black saves an
important tempo by leaving it
there.
Kasparov reasons that in or
der to make progress on the
queenside, White will inevita
bly have to play Nb5 at some
,
stage. This however weakens
the e-pawn, and allows Black
to play ... g4 without hindrance;
if White plays fxg4, Black is
planning on e4, not g4!
71
1 8
aS
Bd71 1 9
Nb5?
Straight into the positional
trap. Kasparov gives 1 9 Kh1
as better, and 1 9
.
.
. .
Qe8 (giv
ing extra cover to b5) as .. un
clear"; Black is still preparing
to roll however.
1 9
. . .
g4!
. •
Observe thi s d iagram , and
think carefully on the note to
1 7 . . . Bf8 ! ! In so doing , you
should be able to leam some
thing about the art of purpose
ful maneuvering.
20 Nc7?!
White is a tempo short for all
this! He also gets throttled on
the kingside after 20 Nxa 7 g3
2 1 Bb6 Qe7 22 NbS (22 h3
Bxh3 !) 22 . .. Nh5 23 Kh1 (to
meet 23 . . . Qh4 with 24 Bg1 )
2 3 . . . gxh2 2 4 Bf2 Bxb5 2 5
Bxb5 N g3 + 26 Bxg 3 fxg 3
(Kasparov). There is no imme
diate mating attack, but the
hole on f4 and the monster
protected passed pawn on
h2
combine to make life hard for
White.
So maybe 20
fxg4
is neces
sary, but after 20 ... Nxe4 21
N c7 Ba4 22 Qxa4 Rxc7 the
long term chances are clearly
with Black, who has much the
better pawn structure.
20
. . .
g3!
The standard pawn sacrifice;
White is condemned if he ac
cepts, condemned if he re
fuses.
If he accepts, Black loses two
pawns while White loses one.
For each pawn that Black lo
ses, he gains an open file and
an entry square for a minor
piece; White however loses an
important defensive unit.
72
If he declines . . . wel l . just imag
ine how good a Black queen
on h4
wiJI
look!
21
Nxa8?!
A stray horseman ransacks an
abandoned castle i n a distant
land, while at home the gates
to the king's pala ce will no
longer hold, and \l\1hite's spiri
tual advisers can give no help.
Such is the transitory nature of
the gains made in so many fo
reign campaigns!
21 hxg3 had more relevance
to the domestic front. Nikitin,
presumably working with Ka
sparov, gives detailed analy
sis, summarized nere. Best
play for Black is 2 1 . . . fxg3 (21
. . . Nh5?! 22 g41) 22 Bxg3 Bh61
23 .. Nxa8 Nh51 (not countir:-tg
material !) 24 . . . Bf2 Ngf4 (24
. . . Bxc1 !? 25 Qxc1 Ngf4 is a
perfectly palatable alternative
for those who do n ot like sac
rificing heavily - CSC ) 25 Nd3!
(25 Rc7 Ba4! 26 Qxa4 Nxe2+
27 Kh2 Bf4+ 28 Kh3 N hg3! 29
Rxf7 Kxf7 gives B lack a win
ning attack; dark squared con
trol plus exposed king !) and
the rest we give as an analyti
cal exercise (number 1 4): can
Black win with
2 5
. .
.
Nxg2
here? and if not, how should
he play?
In such positions the strategy
is relatively straightforwa rd ,
but the tactics are com p li
cated. Black must get at the
White
K.iny,
oLr u=rwise ile
� �
worse. To get at the king he
must sacrifice, maybe only a
pawn to start with, but perhaps
more later in order to break
down the last lines of defence.
Whenever sacrifices occur,
the question oi sounciness ari
ses, and so does the need to
calculate precise variations.
There may be three different
ways to sacrifice, or to follow
up a sacrifice in a given posi
tion; one may be tempting but
unsour.d, one may be unclear,
and one may be good. H ow
can a player tell? Often CH:tly
throug h calculation, and the
calculations involved may be
really difficult. So tactical s kill
counts for a lot.
21 ... Nh51
Another reason for Kasparov
to be pleased that he avoided
. . . hS. Control of the g3 square
is crucial to Black's attack; so
long as g3 is occupied b y
pawn or knight the White king
73
has nowhere to go. There is
no escape through 22 Bxa 7
Qh4 23 h3 Bxh3 (another the
matic sacrifice) 24 gxh3 Qxh3
25 Rf2 gxf2+ 26 Kx.f2; the
White king is obviously far too
exposed to survive. The alert
reader will not easily be se
duced by such words as "ob
·viously". To such a reader we
ask the question of" how, if at
all, Black wins (exercise 1 5).
Nikitin gives 26 . . . Nh4 27 Bf1
Qh2+ 28 Ng2 Rg7 winning, but
27 Nd3! improves. Remember
the bishop on a 7 may still help
the defence; maybe White's
21 Nxa8 had some relevance
to the kingside after all!
22
Kh1
gxf2 23 Rxf2
Ng3+1
Banged into the same square
I
24 hxg3? hxg3 is hopelessly
lost.
24
Kg1 QxaS!
Cool and calm, Black collects
the knight. White has no time
to reciprocate.
25 Bc4
(diagram)
25
. . .
aS!
A delicate little touch. Now that
the mighty White bishop has
been exchanged , the Black
Queen takes control of the al
agonal immediately behind the
White pawn chain. Yet the
more orthodox route remains
available if needed: 26 hxg3
fxg3 27 Rb2 Qd8 28 Kf1 Bh6
.29 Ke2 Qg5 30 Rc3 Nf4+, etc.
26 Qd3?1
M issing his last defensive
chance 26 Nd3 Qa7 27 Nc5!?,
hoping for 27 . . . dxc5? 28 d6
with complications. Black can
sidestep this with 27 . . . Bb5! 28
BxbS axb5 29 hxg3 fxg3 fol
lowed by a capture on c5 and
. . . Nf4.
26 ... Qa7 27 b5 axb5 28 Bxb5
(diagram)
28
. . .
Nh1 1
White resigns.
74
A classic King's Indian game
from one of Kasparov's most
successful tournaments (first
oy
� Y2
points!)
White's plan in this game was
particularly uncompromising,
and therefore the play became
unusually sharp. Having a bi
shop on f2 is fine for helping
the queenside attack along,
but it does nothing to hinder
Black's .. -. g4 break. What can
White do to slow Black down
·· on the kingside? One radical
plan, which we shall look at in
more detail in the next chap
ter, is for White to play g4 him
self; the king may look ex
posed, but he is less likely to
be choked by a Black pawn on
g3. There is another plan how
ever. When White has played
f3, the f2 square is open to
a minor piece. If White plays
a
knight
there, rather than
a bishop, it becomes more dif-
ficult for B lack to play . . . g4 .
This suggests a maneuver
Nf3-e1 -d3-f2.
�ame I Z
£ebalo
•
Cvltaa Yugoslav
�•••plonsblp
1 986
1 d4
Nf6 2 c4 g6 3 Nc3
Bg7
4
e4
d6
5
Be2
It is a lso possible to pre
p
are
to build the pawn chain more
directly with
5
f3, the Samisch
System. Such an idea is allur
ing , but, White's inability to use
the f3 sq uare for his knig ht
makes kingside development
rather more difficult; often he
has to try awkward maneuvers
such as Ne2-c1 . The modern
preference is to develop the
kingside first,
before
setting
up the ful l pawn chai n : th is
g ives the minor pieces more
flexibility.
5
. . .
0-0
6 Nf3 eS 7
0-0 Nc6 8
dS
Ne7
9
Ne1 Nd7 1 0 Nd3 f5
1 1 Bd2
(diag ram)
White's minor pieces are now
in place. At some stage he will
play if requ ired f3, and then
Nf2
in response to Black's .. .f4 .
75
An immediate 1 1 .. .f4 is pre
mature however since White
has
1 2
Bg4! preparing to ex
change h is bad bisho p , for
example 1 2 .. . h5 1 3 Be6+ Kh8
1 4
f3 g5 1 5
b4
Ng6 1 6 c5 Nf6
1 7
Bxc8 Rxc8 1 8 h3 Rg 8 1 9
b5 Bf8
20
c6 Rogers - D. Pau
novic, Belgrade 1 988.
This is almost an ideal position
for White ; without the B lack
bishop on ca. Black's king side
attack lacks sting , while White
can quite happily live without
his purely defensive bishop on
e2. Looking forward a few mo
ves, Black's futur e . . . g4 will al
most certainly entail a pawn
sacri"d ce (oncP. Wh ite has
played Nf2) . But how is Black
going to justify this sacrifice if
he has no bishop to keep pres
sure on the pawn on g4? To
see how important Black's light
squared bishop can be in such
situations, consider again the
Korchnoi - Kasparov game in
the introductory chapter.
Back to the game, and a little
bit of cat-of mouse before the
full chain is finally set up.
1 1 ... Kh8 1 2 Rc1 f4 1 3 f3
Since now 1 3 Bg4?! would be
met by 1 3 . . . h5 1 4 Be6 (no
longer check) 1 4 . . .f3!
1 3 . . . Ng8 1 4 b4 Ndf6 1 5 c 5 g 5
1 6 cxd6 cxd6 1 7 NbS
As in our previous game, Whi
te attacks along the open c
file. The absence of the bishop
from the g 1 -a7 diagonal slows
down the White initiative how
ever.
1 7 ... NeB
Even so, Black must keep c7
76
covered. Exercise 16: assess
1 7
. . .
Rf7.
1 8
a4
h5
19
Nf2 Nh6
Extra support for . . . g4.
20 h3 Rg8 21 Rc31
'Nhite prepares to triple on the
c-file, obviously, but he also
prepares a third rar.!< traverse
for the (ook in the event of
Black playing a later .
�
. g4.
21 . . . 8f8
N atural enough, but Black co
uld also consider dislodging
the White knight first with 21
. . . a6. Cebalo then gives the
thematic variation 22 Na3 Bf8
23 aS Nf6 (now that
the
c7
square no longer has to be
watched) 24 Nc4 g4 25 fxg4
hxg4 26 hxg4 Bxg4 27 Nb6
Rb8 -28 Nxg4 Nfxg4 29 Bxg4
Rxg4 30 Qe2 followed by Rfc1
with, according to Cebalo, a
slight plus to White. But play
on a move, 30 . . . Qh4, an d
Black is preparing for a stan
dard King's Indian garotting
with . . . Rg3 and . . . Ng4. Exer
cise 1 7: analyze this position.
22
Qc2 Bd7
Too late for 22 . . . a6 because
of 23 Nc7.
The immediate 22 . . . g4 was
worth considering though. Af
ter the inevitable multiple ex
changes on g4, Black certainly
cannot claim any sort of win
ning kingside attack, but at
least he has some counterplay
along the g-file to offset part
of White's queenside initiative.
Cebalo gives 22 . . . g4 23 fxg4
hxg4 24 hxg4 Bxg4 (not 24
. . . Nxg4? 25 Nxa7! - a standard
tactic that is worth remember
ing) 25 Nxg4 Nxg4 26 Rh3+
N h6
27 Rc1 with perhaps a
slight edge to White.
(diagram)
Would the reader wish to dis
pute this assessment? That is
fair enough; Black has many
chances of creating play, per
haps starting with
27
.
.
. a6.
I
am not going to set this posi-
77
tion as an analytical exercise,
although
I
would certainly not
wish to discc:.Jrage the reader
from trying to work out what is
happening.
There is a general point that
needs to be made. I n the
King's Indian, and particularly
in position with extendecl pawn
chains,
the basic strategy in
terms of pawn structure is
straightforward; the detailed
i nte ra ctio n s am o n g t h e
pieces are however very in
tricate and complicated.
The
full pawn chain structure is one
of the most distinctive pawn
structures there is, yet the
fixed nature of the ce ntral
pawns mean that the empha
sis is very much on piece play
rather than on pawn play, par
ticularly when both sides have
already made their thematic
pawn breaks (c5xd6; . . . g4).
Yes, the pawn center is as the
center of everything, but it is
an inert nucleus. The action
goes on
a round
the center,
mostly in an anticlockwise di
rection.
23 Rc1 Rg7?1
A slip or a deliberate waiting
move? In either case, the im
mediate 23 . .. g4 probably de
served preference.
24 a S g4 25 fxg4 hxg4 26
hxg4 Nxg4 27 Nxg4 Bxg4 28
Bxg4 Rxg4
A typical position following a
liquidation on g4, although it
has to be added that Black has
been at fault in &loverpreparing" ..
his pawn break. Certainly the
re was no need to allow White
to double rooks on the c-file
before playing . . . g4; see note
to 2 1 . . . Bf
B
.
White is momentarily safe on
78
the kingside, but in order to
claim a secure and stable po
sitional advantage he must
completely block
off
any hint
of
Black initiative
on
this side
of the board. His
next
two cen
tralizing moves do this.
29 Be1 1 Qg5 30 Qe2!
White is now ready to p lay
ReB,
surrounding Black. Black
now tries to set up some tricks.
30 ... Bh61 31 ReS Kh71
There is nothing in 31 . . . f3? 32
Qxf3 Qe3+ 33 Qxe3 Bxe3+ 34
Bf2 Bxc1 35 Rxa8.
32 Qf31
Locking the kingside; White
now has a clear positional win.
32 Rxa8? is careless; Black
has a second kingside b reak
with 32 . . .f3! which saves
the
game. Thus 33 Qxf3 Qxc1
34
Qxg4 Qxe 1 + 35 Kh2 Bf4+ 36
g3 Qf2+ 37 Kh3 Qf1 + 38 Kh4
Nf6! !
39 Qf5+ Kg7 4 9 gxf4
Qf2+ 4 1 Kh3 Qf3+ 42 Kh4
Qf2+ with a perpetual.
Cebalo's technique is good
enough to avoid the cheap.
32
.
. .
RxcS 33 Rxc8 a6 3 4
RxeS axbS 3 5 RbS Qg7 3 6
Rxb7 Rxg2+ 37 Qxg2 Qxb7
�8 Qg4
A typical winning King's Indian
endgame for White:
(I)
White has an outside
passed pawn on the
queenside which is very
dangerous, whereas
Black's passed pawn on
the kingside is well cove
red , a nd has little real
chance of queening.
(I I)
Black's .. bad bishop" is
79
seriously hemmed in by
his own pawns and has
little mobility (a similar
problem often arises with
the light square� bishop
in the French), whereas
White's bishop is fully
mobile on both sides of
the board.
(11.1) Black is extremely weak
on the light squares
covered by the White
pawn chain.
It is no surprise that the
game is soon over.
38 .... Qc7
39 Bf21
Qc3 40 Kg21
White co nso lidate s o n the
kingsiae and allows the a
pawn to win the game for him
on the queer.side.
40 ... Qb4 41 QfS+ Kg7 42 a6
Qa
4 43 a7 b4
Now Black no longer has the
possibility of . . . Qxe4 + , but
wh at e l se co uld he h a ve
done? He was in zugzwang!
44 Qc8
Black resigns.
A highly thematic White vic
tory.
Chess is long , and books are
short, so sadly there is l ittle
scope here for a full discussion
of the various ways in which
the kin gside struggle
in
the
Ki ng's Indian may deve lop.
One popular and interesting
possibility needs to be dis
cussed though; the paradoxi
cal g4, throwi ng forward a
pawn on the flank on which
Black is attacking.
6.
The
Paradoxical Push
.. Never push pawns of the bo
ard where you are weak; you
just create further weakne
sses. " This is one of the main
precepts of classical chess
strategy, and it is on the whole
valid. I ndeed, a standard at
tacking technique against a
poorly defended sector is to
use piece pressure to attempt
to force a pawn advance, and
then to u se the newly a d
vanced pawn as a lever for the
attack; .�ither one can prepare
a pawn advance to force open
some lines (e.g. g4-g5 if Black
has been forced to play . . . h6)
or one can, especially if the
pawn has advanced in
front
of
the defending king , prepare a
sacrifice (e. g. Bxh6).
80
Yet sometimes an apparently
bizarre pawn advance may be
used to pre-empt
an
attack, to
stop it before it has started.
The defender makes a bold
push in a beleaguered sector,
and the opponent finds he is
unable to build up his attack
at leis ure. Systems where
White plays g4 have become
quite popular in the King's In
dian; it is, after all, the surest
possible way of preve nting
Black from playing . . . g4.
H ere is a simple exam ple;
even a quick draw may some
times be a useful illustrative
game.
Game I!J
Kl1allfman
•
Gelfa11d
Regg�o
Emilia
199 1192
1
c4 g6 2 Nf3 Bg7 3 e4 d6 4
d4 Nf6 5 Nc3 0-0 6 Be2 e5 7
0-0 Nc6
8
d5 Ne7 9 Ne 1 Nd7
1 0
Nd3 f5 11 Bd2 Nf6 1 2 f3
Kh8
1 3 a4 a5
Black does not want White to
gain space wi�h a5.
14
g4
The paradoxical push!
White has no immediate plans
to m a ke any
fu rther
pawn
moves o n the kingside; he
merely waits to see what Black
does. Black has three basic
choices:
(I) He can leave the pawns
as they are, but this would
be a sign of success for
White: he has halted
Black's kingside initiative.
(II) He can exchange pawn
on g4 or e4. It was poin
ted out in the introductory
chapter that with the Whi
te pawn still on g2 this
would be a very bad · ex
change for Black; he wo
uld have released all the
central tension, and he
would no points of entry
along the open f-file, while
White would continue to
have a free hand on the
6
-
Pa"' - :- "'ams
81
queenside. With the pawn
on g4, this argument app
lies with slightly less force
(White's king is a bit open),
but still the play is in
White's favour.
(Ill) He can block the king
side with . . .f4, and Whi
te's kingside remains
under pressure, but much
less so than if Black is
allowed to play . .. g4.
A difficult choice for Black,
out here made easier by
the fact that Black has
done his own pre-emptive
blocking on the queenside
(13
..
.
aS).
14
. . .
c5!
Closing the queenside. Now all
that Black has to do for com
plete equality is to close the
kingside as well.
1 5 Kg2
Tak_ing
en passant
is not to be
considered; after 1 5 dxc6?
bxc6 Black is vastly better,
White having conspi cuo us
weaknesses along the b-file,
the g1 -a7 diagonal, and on the
kings ide.
1 5 . . . f4 1
6
h4 h5 1 7 g5 Ne8
1 8 Rh1
Draw agreed.
Not much chance to play for
a
win here! Both pl�yers have
succeeded in locking the pawn
structu re o n th eir weaker
flanks.
There are chances for livel!er
play though, as we shall short
ly �ee.
Game
14
Lobron
•
aenet
Novl Sad
1990
1 Nf3 Nf6 2 c4 g6 3 Nc3 Bg7
4 d4 0-0 5 e4 d6 6 Be2 e5
7
0-0 Nc6 8 d5 Ne7 9 Ne1 Nd7
Not of course the only varia
tion of the King's Indian, but
even so a variation which ex
presses classical themes ex
tremely well.
1 0 Nd3 f5 1 1 Bd2 Nf6 1 2 f3 f4
82
1 3 g4!?
The theme of th.e chapter, al
th ough i n most
v
a ri a tivns
White plays g4 befo re Black
has played . . . f4. lhis poses
the question of whether Black
can profitably play 1 3 . . . fxg3
1 4 hxg 3 . White 's k i n gside
pawn structure is
ugly,.
and the
king's position is exposed ,
which reduces his options so
mewhat. On the p l us side,
White has the rather pleasant
c1 -h6 diagonal for his bishop,
and the h-file may well become
useful. In M. Pavlovic - Vokac,
Trnava 1 988, Black's position
became highly unpleasant af
ter 1 4 . . . c6 1 5 a4! aS 1 6 Ge3
Kh8 1 7 Qd2 cxd5?! (closing
the queen side with 1 7 . . . cS
was preferable, but White still
has kingside pre s s u re) 1 8
cxd5 Bd7 1 9 Kg2
Rf7
20 Rh 1
ReS 21 Nf2.
Without that wonderful pawn
on f4, Black's kingside sud
denly seems u ncoordinated
a n d vulnera b l e , while h is
queenside problems still· have
not been solved (in the game
cited, White
r.z.� �:-.c
i-.-.ai-acu
ver Ra3-b3 i n hand) . It is
hardly surprising that Black
usually prefers to keep his f
pawn.
In such a case, the fact that
White has safely played g4
may be regarded
as a .-.-.i.-.o.
strategical victory, although
the h4 square will need care
ful watching.
1 3 ... g5 1 4 Rc1 h5
Black must open the h-file.
1 5 h3 Rf7 1 6 Be1
White must cover h4.
1 6 ... Bf8 1 7 cS N g6 1 8 cxd6
cxd6 1 9 Qb3 Rh7 20 Kg2
(diagram)
The White fortifications are al
most complete. Rh1 and Bf2
are to follow.
N ote that Black's attack makes
no progress after 20 . . . hxg4 2 1
hxg4 Nh4+ 22 Bxh4 Rxh4 23
83
Nf2.
·20
. . .
as
21 a4 Bd7 22 Bf2 ReS
23
Rh1
Qe8?
Letting White have the impor
tant b6 square without a strug
gle. Lobron suggests instead
23 ... Rg7, and if 24 Qb6 N h4+.
Maybe Whita is still slightly
better after 25 Kf1 .
24 Qb61 Rg7
Si nce 24 . . . Bxa4 25 Nxa4
Rxc1
26 Rxc1 Qxa4 27 Nc5!
is highly favorable to White.
25 b3
Not just a simple defensive
move; White also has ideas of
Nb1 -d2-c4. Slow? Well, what
can Black do on the kingside
in the meantime? The whole
pace of the game is slower
than when the kingside is more
open and Black is aiming for
the . . . g4 break himself.
25 .. . Nh4+ 26 Bxh4 gxh4 27
Nf2
The second wave of defence
to forestall the second wave of
attack. White must cover all
potential sacrifices on g4.
27 . . . Nh7 28 Nb1
See note to White's 25th.
2 8 .. . N g 5 2 9 Nd2 hxg4 30
hxg4
·
Black is ahead on the kingside,
but he has no barnstorming at
tack. Meanwhile he has his
u sual queenside problems ,
and his bishops are both ex
tremely ineffective; one bishop
is blocked by friendly pawns,
while the other is enchained by
enemy pawns. White's knights
84
are, g iven passive play by
Black, headed towards h3 (to
exchange off the only effective
Black minor piece) and c4 (to
take control of the queen side).
To �ave such a position with
normal defensive play would
be quite an achievement for
Black. Rightly or wrongly (and
Lebron suggests that Black
should have prefe rred 30
. . . Rh 7) Black lashes out.
30 ... h3?1 31 Nxh3 Nxh3 32
Kxh3 Qh5+ 33 Kg2 Bxg4 34
fxg4
Not 34 Rxh5?? Bxh 5+ fol
lowed by 35 . . . Rxc1 .
34 ... Rxg4+ 35 Bxg4 Qxg4+
36 Kf2
Black's attack is all illusion,
thoug h only because the Whi
te queen is so well placed on
b6.
White's extra rook will soon
decide the issue.
36 ... Rxc1
37
Rxc1
Be7 38
Rg1
38 Nf3 !
would have been sim
pler.
3 8 . . . Bh4+ 3 9 Kf1 Bg3 40
Rx
g
3
40 Qf2! Qd1 +
41
Kg2 would
have been an artistic finishing
touch, but White would have
wanted to reach the time con
trol as simply and safely as
possible.
40 .. .fxg3 41
Qe3
Qh3+ 42
Ke2 Qh2+
43
Kd3
g2 44
Nf3
Qg3 45 Ng1 Qg4
46
Kd2 Kg7
47 Ke1 Kg6
48 Kf2 Kh5 49
a5
·Black resigns
.
Castling kingside
,
and then im
mediately launching the g
pawn, is a parado
xi
cal but of
ten effective stratagem. White
can also consider playing g4
before
he castle
s
,
in which
case the nature of the play be
comes rather more violent,
with Wh ite ofte n castling
queenside, and using the g
pawn to support a White pawn
85
storm on the kingside
.
Black
in return must attack on the
queenside
,
a reversal of the
normal roles.
Our example comes one step
removed from the King's In
dian.
Game
15
Itllreovle
•
Todoreevle
Yugeslav Championship
1991
1
d4
as 2 e4 Ba7
3 c4 d6 4
-
-
Nc3
Nd7
Black is cagy about his inten
tio
n
s. If White is not careful,
he will find himself in an inte
rior King's Indian where Black
can safely play . . . f5 without
having to worry about getting
the king's knight out of the way.
5
g4?1
White prepares the big clamp
on the kingside.
5 ... e5
It is tempting to adopt a Dra
gan/Benoni formation with 5
. . . c5, but the knight on d7 is
misplaced for this plan; Black
needs to p ressurize d4. White
can quite happily keep the
central tension with 6 Be3 fol
lowed by Qd2, f3, etc.
6 d5 Ngf6
Black opts for a pseudo-King's
Indian, but White can prepare
a pawn storm.
The sharpest way to play is to
play for an early . . . f5, notwith
standing White's attempt to
eliminate this move. After a
double exchange of pawns on
f5, White will have complete
control of several key light
squares, particularly along the
b 1 -h7 diagonal , and he will
also have possibilities of attack
along the g-file. It should not
be forgotten though that White
will be weak along the f-file,
a n d tha t B l a ck h a s d a rk
square control, with . . . Nd4 or
the pawn sacrifice . . . e4 being
interesting possibilities. So for
exercise
1 8
we ask what is
86
happening after 6 ... Ne 7 7 Ne2
f5.
7 h4
Most certainly not 7 g5? Nh5,
possibly followed by . . . Nf4
and/or the opening of the f-file
with . . . f6. White's pawn struc
ture woul d have lost all its
natural elasticity.
With the next move White
threatens to extend his grip on
the light squares by playing h5.
This is a good move, whether
Black allows the h-file to be
opened, or whether he tries to
keep the h-file closed. If for E)X
ample Black tries 7 . . . h6, then
8
�5!? g5 9 f3 gives White a
very favorable pawn structure.
In such a position the kingside
is completely blocked with only
White having chances of play
(Nf5), while White still has his
traditional advantage on the
queenside. Only White can
win , a nd Black's defensive
task is burdensome.
Black decides he must prevent
h5.
7
..
.
h5
8 g5
Nh7
9
Be3
0-0
1 0 Nge2 Qe7 1 1 Ng3
Advertising the possibility that
a sacrifice on h5 might later be
available. It helps that Black is
press�d for space on the king
side.
Furthermore any attempt to
break free
with
. .
. f5 or . .
.
f6
can
be
met
by
exf6,
after which the
Black pawn
on
g6 will be very
weak. Note that the bishop on
e3 effectively protects
the oth
erwise weak White pawn on
f2.
There is also a more immedi
ate tactical problem; if now 1 1
. . .f6? 1 2 N xh5! gxh5
1 3
g6 is
strong. So Black is uncomfort
able.
1 1 . . . Rd8 1 2 Be2
Now
13
BxhS followed by Nf5
is a threat.
1 2 . . . NdfB 1 3 Qd2 c5
The "block and break" theme,
87
discussed further in the next
chapter. With White standing
better on th e kingside . al
though obviously preparing to
castle queenside, . Black must
prepare counterplay on the
queenside, on his .. unnatu ral"
wing. To do this, he must firstly
block with . . . c5, then break
with . . . b5.
1 4
0-0-0 aS 15
Rdg 1
b
5 1 6
Qd1
Black's sacrifice should be ig
nored as a matter of principle.
After 16 cxb5? axb5 1 7 Bxb5
White gains a pawn, b�t so
what? Black's two open files
on the queenside p rovid e
ample compensation, and 1 7
.
. . Bg4 followed by . . . Rdb8 will
bring to a halt White's attack,
and h a n d the i n iti ative to
Black.
16
. . .
bxc4
(diagram)
1 7 Bxh5!
Thematic and very strong.
White is aiming a few pawns
at Black's congested kings ide.
Without this sacrifice, White
would have noth ing o n the
kingside. and would have to
content himself with closing
down the q ueenside, some
thing clearly not possible when
Black has already played . . .
bxc4.
Seen in this light, it is clear that
White must already have de
cided that this sacrifice was
promising when he castled on
move 1 4, otherwise he would
have tucked his king away on
the kingside, maybe on f2, and
prepared to meet . . . bS with b3
and a likely draw.
1 7
. .
.
gxh5
1 8 Qxh5 Rb8 1 9
Nf5
And this is
why
the sacrifice is
strong . Every pawn removed
from the board means that at
least one square is weakened;
here the weakening of Black's
fS is highly significant. White's
newly found control of h5 is not
so important of itself; the real
88
point is that the road is clear
for the h-pawn.
1 9 ... Qb7 20 Qe2
Ng6
21 hS?!
Sometimes it is the mistakes
that are instructive. The move
played gives Black a fleeting
opportun ity for counterp!ay
(can you see it?). The correct
.. move is 21 Nxg7 ! A. player
would be naturally reluctant to
exchange off his excellent
knight so soon after it had
reached its strongest square,
but the bishop on g7 is
crucial
to Black's kingside defence;
White has no need to feel
guilty here about exch�nging
good knight for llbad" bishop.
After 2 1 Nxg7 ! Kxg7 22
h5
Black is unable to move the
knight away from g6 because
White's g6 move would be ter
rifically strong . Black would
have nothing better than 22
. . . N hf8 23 hxg6 Nxg6 24 Rh6,
but White would then have lev
e lled material and kept his
king side initiative.
Some examples of what hap
pens if Black tries to keep his
extra piece (based on analy
sis by Mirkovic):
(a) 22 . . . Nf4 23 Bxf4 exf4 24
g6 fxg6 (otherwise 25 h6+
is too strong) 25 hxg6 Nf6
(25 . . . Nf8 26 f3! Nxg6 27
Rxg6+ !) 26 e5! dxe5 27
Rh7+! Nxh7 28 gxh7+
Kh6 29 Rh1 + and a
queen check decides.
How easily the king's
cover is stripped bare
once the bishop has
gone!
(b) 22 . . . Ne7 (22 . . . Ngf8
merely leaves an extra
target after a later g7) 23
g6 fxg6 24 hxg6 Nf6 (24
•
. . . Nxg6 25 Rxh7+) 25 Bh6+
Kg8 26 Bg5 Kg7 (26
. . . Rf8 27 Bxf6 Rxf6 28 g7)
27 Bxf6+ Kxf6 28 Rh7
with a winn1ng attack.
21 . . . Bxf5 22 exf5
(diagram)
22
. . .
Nf4?
89
Black m isses a s p l e n d i d
chance to break fre e . Any
King's Indian aficionado will
have a fond understanding of
the power of the Indian bishop
once the long diagonal has
been swept open, and so it
proves here. 22 . . . e4! is the
move, thrP.atening . . . Bxc3 ;
Black is not scared of 2 3 hxg6
Bxc3 24 gxh7+ Kh8. White
can of course throw a cloak
over the bishop with 23
f6,
but
then the knight takes over the
open e5 square: 23 . . . Ne5!
Mirkovic gives 24 fxg7 Nd3+
25 Kd 1 as ��unclear''; it is also
an interesting position to ana
lyse (exercise
1 9) .
The first
question to ask is w hether
Black stands better.
23 Bxf4 exf4 24 f6
Back to normal after some
missed opportunities. White's
ki ngside press u re is formi
dable; Black's queenside at
tack is merely irritating
.
24 . . . B h8 25 Qc2 ReS
Mirkovic shows that the co
unter-sacrifice o n f6 avails
nothing: 25
. .
. Bxf6 26 gxf6+
Kh8 27 Rg6! fxg6 28 hxg6 Rd7
29 Ne4 Qxd5 30
f7
and the
Black king is defenseless.
26 g6?!
This could have waited a bit;
26 Rh4! was strong.
26 ... Nxf6 27 g7
Painful! Of course 27 . . . Bxg7
28 h6 loses quickly, but Black
must rrepare the counter-sac
rifice somehow.
27 . . . Nh7 28 Rh4
90
28 .. .f5?
The tension reaches a peak as
White prepares Rhg4 followed
by a capture on h8 and a n at
tack down the g-file. In addi
tion, Black was getting short of
time. He makes the mistake,
common in such situations, of
seeking a premature release
of tension, when everything
subsides, White has a clear
positional edge.
Exercise 20; can you suggest
a better defensive plan for
Black? Examine this position
carefully and you will see how
knife-edged these pos!tions
with �ttacks on opposite wings
can be.
29 h6 Re7 30 Rxf4 �xg7
The release of tension.
31
Rxg7+ Rxg7 32 hxg7
Qxg7 33 Qxf5
(diagram)
Illustrating the theme
of
ata-
..
vism in chess. Looking at the
diagram, one of the most ob
vious points is that White is
taki n g co ntrol of the l i g ht
squares.
This becomes even clearer in
few moves time, when the
VVhite knight will have reached
e4 and the rook g4, while the
advanced d5 pawn will con
tinue to highlight the weakness
of the e6 square. A lot of wood
has crossed a lot of squares,
but the position is one which
is highly characteristic of pawn
chain strategy. I n particular,
the s pearhead pawn on d5
stays in placs, and while the
e4 pawn has long since dis
a p p e are d fro m th e boa rd
White still has control of the e4
square.
So does this mean that White
has been consistently follow
i n g a classical pawn chain
strategy, and is finally reaping
the benefits? Even if one ig
nores the various mistakes in
the earlier play, ·such an inter
pretation does not hold. After
all .
one can also deduce from
91
the diagram position, perfectly
correctly, that there has
re
cently been a sharp battle with
both sides trying hard to open
up lines and create attacks
against kings castled on oppo
site flanks. This is, if you like,
the ��father'' of the position. The
immediate paternity of the
po
sition is currently not so impor
tant; White may have the
oc
casional threat down the g-file,
.
but it is not going to be this
that
decides the game. The
strate
gic features which are most
im
portant now, are throwbacks
to
an earlier "generation",
atavis
tic features. The basic
tone
of
the position has been
set
by
the sequence 5 g4 e5
6 d5;
White has aimed for ambitious
control of the light
squares,
while Black has tried
to
hold
steady on the dark squares.
White has been more s uc
cessful than Black, but
that
is
because he has played
better
than Black.
It would be a mistake to
think
that White's light squared
con
trol has been an enduring
fea
ture of the position, and an
even bigger mistake
to con
sider it merely accidentaL
It
is
atavistic, a genetic trait
(this re
lating to the genesis of the po-
sition) which has not visibly
manifested itself for a few
"generations" (phases of play),
but which suddenly appears at
a much later date. If you the
reader are not convinced by
this line of reasoning , and
would wish to argue that White
was
always
better on the light
squares, I would refer you to
the position set in exercise 1 9,
where Black had just estab
lished his knight on d3, and
had pawns on c4 and e4 - not
really signs of White domina
tion ! So what is the particular
significa nce of atavism i n
chess? Perhaps the most i rr.
·
portant point is that it is worth
accumulating positional ad
vantages even in situ�tions
where it can be for seen that
the game will soon be domi
nated by tactics and by com
plicated sequences of attack
and counterattack. However
messy the complications ap
pear to be at the time, they will
eventually subside, and there
is a f�ir chance that ancient
strategic features of the posi
tion will assume new promi
nence. Believing in chess ata
vism, one can says as a mat
ter of faith,
"
I like my position,
and I want to preserve my ad-
vantages. If he wants to com
plicate, let him! Though every
thing seems confused; I be
lieve that I go into complica
tions with a good position, I will
emerge with a good position."
A long philosophical digres
sion. How does atavism work
in practice?
33 .. . KhB
33 . . . Qg 1 + is well met by 34
Nd1 .
34 Rg4 Qh6+ 35 Qf4
White is happy with an end
game, _and indeed is prepared
to meet 35 . . . Qh1 + 36 Kc2
Qa1 with 37 Qc1 ! White's d
and f-p awns are stro n g ,
.. Black's d-pawn and front c
pawn are weak.
92
35 ... Qh1 + 36 Kc2 RfB 37 Qe3
Qh2 38 Ne4!
(diagram)
The throwback to classica l
pawn chain strategy is com
plete. Black's ineffective attack
down the f-file and modest
control of dark squares do not
compensate for White's mas
sive grip on the central light
squares.
The threat of Qc3+ now forces
an endgame in which the
Black pawns are genetically
predisposed to drop off. (Curi
ously, I have never seen ripe
apples fall like pawns.)
38 ... Qe5
39 Qc3 Qd4 40 f41
Rf7 41 Nxd6 Rd7 42
Qxd4+
cxd4
43 Nxc4 Rxd5
44
Kd3
There is no real difficulty her�.;
White is a pawn up and has
the better pawn structure and
the more active pieces; the
endgame win is simply a mat-·
ter of time.
44 ... Nf6 45 Rg5 Rd8 46 Ne5
Kh7 47 Ng4 NeB
If Black allows any exchanges,
the win is even easier.
48 b3 Ng7
49 Nf6+
Kh8 50
ReS Rf8 51
Nd5 Nf5
52 Ke4
93
Nd6+ 53 Kxd4 NbS+ 54
Kd3
Rd8
55 a4 Nd6 56 Kd4 aS 57
KcS Nf7 58 Re7 Kg8 59 Ra7
Black resioned .
.....
Not the most accurate game
in this book, but of theoretical
and thematic interest.
The .. paradoxical push" can
also occur in the French De
fence, for example 1 e4 e6 2
d4 d5 3 e5 c5 4 c3 Nc6 5 Nf3
Qb6 6 a3 Bd7 7 b4 with ad
vantage to White. Black can
prevent this with 6 ... as or the
bypassing move 6 .
.
. c4.
A
more important case is that
of the Nimzo-lndian. There are
many variations in this open
ing where White forms the
pawn chain f3-e4-d5, while
having doubled pawns on c3
and c4. The shape of White's
pawn chain suggest queen
side activity, but the doubled
pawns rule this possibility out.
White's most promising plan,
unlikely and paradoxical tho
ugh it might seem at first sight,
is to keep his king in the cen
ter and use his f3 pawn as a
pivot for a general kingside
pawn advance. Maybe in a fu
ture volume we shall consider
the doubled pawn complex in
more detail; in the meantime
here is an example of White
gaining an edge with this plan.
1 d4 Nf6 2 c4 e6 3 Nc3 Bb4 4
f31 ? c5 5 d5 Bxc3+
6
bxc3 d6
7
e4 e5 8 Bd3 Nbd7 9 h4
The kingside advance starts;
White proposes h5 and h6,
while if Black prevents this with
9 . . . h5, the g5 square is weak
ened, and White can aim for
Bg5, Ne2-g3-f5.
9
. . .
Nf8
Castl i n g kingside, into the
pawn storm, is too dangerous.
1 0 h5 h6 1 1 Rb1 Qc7 1 2 Qb3
Kd8
The first sign of planlessness,
but · what improvements are
there? If 1 2 ... b6 1 3 a4 White
p repare s to attack on the
queen side.
1 3 Ne2 Ng8 1 4 Ng3 Ne7 1 5
Be3
(diagram)
A very pleasant position to
play for White, Gutman - Am.
Rodriguez, New York 1 988.
Black has been unable to take
the initiative on the queen side,
94
so the doubled c-pawns
are
not weak.
·v·v'i 1ii� i lC:t::; piay
along the b-file,
and his kingside pawns clamp
Black very effectively. Note
that whereas in the King's In
dian the key push is g4 (to re
strain . . . f5), in the Nimzo-,n
dian the leading pawn is gen
erally the h-pawn (White must
provoke weaknesses).
7.
Bloek aad 8reak
In a book this size, it would
be
impossible to give full cover
age to all major pawn chain
themes. The best that can be
hoped for is that the examples
and discussion in this book will
have improved the reader's
positional fluency when deal
ing with what is undoubtedly
a
rathe r complicated typ e of
pawn structure. There is one
theme that definitely should
not remain undiscussed and
I
that is the "block and break"
theme.
Our last chapter showed ex
amples where White pushed
h i s kn i g ht's pawn on h i s
"opponent's" flank, either re
straining . . .f5 (or . . . c5) or ac
tually attacking the opponent's
f- (or c-) pawn. Black can also
p l ay pawn moves on the
'wrongn side of the board; . . . c5
in the King's Indian or .. .f5 in
the French . The primary mo
tive is clear; Black wants to
prevent White gaining too
much s pace on his stronger
flank. With this blocking move,
Black can prevent White's c5
in the King's Indian, or his f5
in the French; on the other
hand it is made easier for
White to open up lines (b4 and
bxc5 in the King's Indian; g4
and gxf5 in the French). Yet
Black can also open up the po
sition, with . . . a6 and ... b5 in the
King's Indian (as we have al
ready seen in Mircovic - Todor
cevic), or by ... h6 and . . . g5 in
the French.
It is probably in the block-and
break type of position that
similarities between the King's
I nd ian a nd French become
most n oticeable. Here for ex-
95
ample is a game that Petrosian
won shortly before becoming
World Champion; the game
starts as a French, but the su
per-subtle strategist so com
pletely wrong-foots a world
class opponent that Olafsson
finds himself playing a grossly
inferior King's Indian, Mirror
Variation.
G•me 16
Olafsson
•
Petroslan
Bled
1961
1 e4
e6 2
d4 d5 3 Nc3 Bb4
The Winawer Defence, strate
gically the most complicated of
all the F
1
ench variations.
Black's pressure on e4 forces
an immediate decision from
White. Normally he plays 4 e5
here, even· though this invol
ves
no
gain of tempo.
The main line 4 . . . c5 5 a3
Bxc3+ 6 bxc3 Ne7. {diagram)
A pawn chain position, but with
a diff9rence; White has doub
led c-pawns! The pawn struc
ture is highly unstable; this is
shown in extreme form in lines
such as 7 Qg4!? Qc7 8 Qxg7
Rg8
9
Qxh7 cxd4 with great
complications.
In this line Black has destroyed
the White pawn center, but
has had to sacrifice his king
side to do this!
Black's kingside is undoubt
edly weaker in the Winawer
than in the Classical (3 ... Nf6),
and the absence of his dark
squared bishop might well be
come a problem (a4 followed
by 8a3 is a standard pian for
White). On the other hand,
White's center is considerably ..
more vulnerable than in the
Classical , his c3 square being
particularly weak (square, and
not just pawn; a sudden open
i n g of the c-file can leave
White weak on
c3
and c4 be
cause of the absence of the
b-pawn).
Who do these differences fa
vour? There is no simple an
swer; perhaps best to say for
the time being that both sides
have something to play for.
96
4 e5 Qd7!?
Avoiding the main lines. The
purpose of this move is simply
to facilitate queens ide castling.
4 ... b6 5 Qg4 8f8!? is another
possioiiity.
1
ne retreat of the
bishop looks strange at first,
but Black has provoked e5 in
a situation where the pawn on
d4 is poorly protected. and in
such circumstances the White
queen is indifferently placed
on g4. indeed a game Blatny
Psakhis, Amsterdam 1 99 1 ,
saw a quick return to: 6 h4!?
h5 7 Qd1 ! (the queen is a tar
get after 7 Qg3 Nh6) 7 . . . c5 8
Bg5 Qd7 9 Bb5 Nc6 1 0 Nge2
a6 1 1 Bxc6 Qxc6 1 2 Nf4 cxd4
1 3 Nce2 Qc7 1 4 Nd3 with an
unclear. position. It is perhaps
more logical for White to play
5 a3 Bf8 6 Nf3 Ne7 7 h4 h5 8
Bg5, and now Portisch gives
after 8 ... Qd7 9 Bb5 c6 1 0 Ba4,
with the idea of Ne2 and c3,
and the White pawn chain is
well secured.
5 Qg4 f5!
Th� kingside block.
6 Qg3 b6
7
h4
A reasonable move, or the first
step down a slippery path? It
all depends on how seriously
White needs to take the pos
sibility of . . . g5.
7 N�3 is definitely worth con
sidering, when
7
. . .
Ba6 led to
White's advantage in H_iartar
son - Adams, Reykjavik 1 990:
8 Bxa6 Nxa6 9
0-0
c6
1
0 Nf4
(pressure on the backward
pawn) 1 0
. . .
Nc7 1 1 Nce2! Qf7
1 2 b3 followed by c4. Yet the
King's Indian player might well
legitimately ask why it is so
necessary to exchange the
bad bishop, when Black is
solid enough after a "real"
fianchetto. 7 ... Nc6! looks emi
nently p layable; . . . Bb7 and
. . .
0-0-0
are to follow. The
trouble with playing a knight to
h3 so early is that White is ne
glecting his important d-pawn.
7 -
Pawn Chains
97
7
. . .
Bb7 8
Bd3?1
This routine developing move
is questionable; the bishop is
misplaced on d3 as White im
plicitly admits four moves later.
Clarke, in
Petrosian's B
est
Games of Chess,
suggests 8
Bd2 Nc6
9
Nf3
0-0-0
1 0
0-
0-0
followed by Ne2 and Nf4.
8 a3 also looks reasona ble.
There is a distinction to be
drawn between development
and merely getting pieces off
the back rank for the sake of
moving them.
8
. . .
Nc6 9 Nge2
Maybe 9 Nf3; who knows?
White's game is slipping, and
appears to be slippin g with
every ruove. When this hap
pens, every move looks like a
mistake at first, when in fact
there may only be one or two
real mistakes. Perhaps in the
next few moves it is White's
position that is bad , not his
moves.
9
. . .
0-0-0 1 0 Bd2 N h 6
As 1 1 Bxh6 will give Black the
g-file.
1 1 a3 Be71
The bishop retreats to its best
square, regardless of the pos
sible loss of a pawn. White is
set a challenge: snatch the
pawn, or forever play a slightly
inferior position. Exercise 21 :
the critical line is 1 2 Qxg7 Ng4
1 3 Bg5; has Black got any
thing concrete here?
1 2 Bb5
Admitting that his 8 Bd3 was
ill advised.
1 2 ... Rdg8 1 3 Qd3
The drifting continues. This
move looks ugly, but how else
does Wh ite secure his d
pawn? Black's piece move
ments are much more harmo
nious.
1 3 ... Nf7 1 4 0-0-0
(diagram)
1 4 ... Kb8 1 1
Petrosian was renowned for
the subtlety, and occasionally
the oversubtlety, of his posi
tional maneuvering. Here his
play is subtle yet comprehen
sible, altogether charming.
98
The point about this position
is that the natural side for de
veloping the Black initiative is
the queenside, not the king
side. Furthermore this initiative
is potentially a powerful one in
that Black has the right piece
development to start a King's
Indian style pawn roller on the
queens ide. Sc Petrosian turns
it into a King's Indian!
Stage I - clear the
c8
square.
1 5 Nf4 Qc8!1
Stage II - major piece to the c
file.
1 6 Nce2 Ncd81
-
Stage Ill; knight out of the way
so that the pawn may ad
vance. (diagram)
Black's last few moves may
look mysterious, when taken
one by one, but it is the posi-
tion at the end of the maneu
vering that counts. Black's po
sition is 1 00% solid - there is
not a ghost of a weakness that
White can attack - and the ac-
+i,
, ,..
""'' ..... ,.. ...... , - . •
;s J·ust about
.., , • .._.
� ""' -..,. I I ., ..., I
t"''--
J
I
to start.
Black's ... c5 '.Nill express both
French and King's Indian the
mes. In French style, Black is
th reatening to destroy the
White pawn center by a sus
tained attack on d4, which
would undermine the spear
head pawn on e5. Shou ld
White decide to complete the
pawn chain with 1 7 c3, then
Black attacks in King's Indian
style, e.g. 1 7 . . . c5 1 8 Kb1 c4
1 9 Qe3 (Qc2? loses a piece
to 1 9 .
. .
a6 20 Ba4 b5) 1 9
.
. . Bc6 ! ? 20 Bxc6 Qxc6 and
Black's maneuvering need no
longer be so subtle; . . . Nb 7 -a5-
b3, . .. Ka8, . . . Rb8, . . . a5, . . . b5-
b4, etc. , is annihilating.
99
Petrosian was never a King's
Indian advocate, but this posi
tion is perhaps most easily
grasped in King's.lndian terms.
To com pensate for Black's
queenside play White ought to
be developing a h uge advan
tage on the kingside, the flank
of his spearhead pawn, but he
has nothing: Black's .. .f5 and
Nh6-f7 have very effectively
blocked off White's play on his
stronger flank, while his . . .
RdgB announces the possibil
ity that Black can break (with
. . . g5) as well as block. Black
is so strong on the queenside
however that this possibility
remains merely hypothetical.
1 7 Qb3?
After this White is clearly lost.
1 7 Nh�_! the only chance, pre
paring to buttress the e-pawn
with f4. An important mini-tac
tical point is that 1 7 . . . c5 may
safely be met by 1 8 dxc5 be
cause 1 8 ... Nxe5? is answered
by 1 9 Bf4. To gain the maxi
mum from the position, Black
has to prepare to recapture on
c5 with a knight. Thus 1 7 . . . c6
1 8 Ba4 Ba6 1 9 Qe3 N b7 !
leaves Black i n the driving
seat; for example 20 Kb 1 c5
2 1 c3 Na5! is strong.
17
. . .
c61
Forcing the bishop to deprive
the knight of the d3 square.
1 8 Bd3 c5 1 9 dxc5 Bxc5 20
Nh3 Nxe5
This pawn may safely be sna
tched as the pin is not as dan
gerous as it looks, for example
21 Bf4 Nf7 22 Ng5 Bd6 ! 23
Nxf7 Nxf7 and Black has con
solidated his extra pawn.
21 Bf4 Ndf7 22 Bb5
Having lost the strategic battle
in the center, Wh�te aims for a
lightning attack against the
Black king. It fails, but it had to
be tried.
22
. . .
Ka8 23 Nd4 Ng6 24 Qa4
Bxd4
Before White can flick in Nc6.
25 Bd7
(diagram)
25
. . .
Qf8
White has done well in ran
domizing the position , an d
100
Black still has to think carefully,
despite being piece and pawn
up. Petrosian plays according
to the principle .. protect every
thing", but there is a suspicion
that this move is over-subtle.
Exercise 22 is two parts. firstly,
is 25 . . . Qf8 a clear win? Clarke
gives as the main line 26 Bxe6
Be5 27 Bxd5 Bxd5 28 Rxd5
QbB! and Black keeps his ex
tra piece; you might want to
look fo r· improvements fo r
White. Secondly, has Black
got any simpler alternative to
25 ... Qf8?
26
Rxd4?
Now it is easy for Black.
26
. . .
e5
To meet 27 Rxd5 with Nxf4
and 27 Bc6 with Qc8.
27
Rb4 exf4
28
Rxb6
Nfe5
29
Rx
b
7
Kxb7
30
h5
Qd6 3 1
h x g 6
Qxd7 3 2 Qxf4
Nxg6
White resigns.
A memorable game.
Here reluctantly, it is time to
bring the book to a close, oth
erwise a short book will be
come a long one. There are
several recent games that I
would have liked to include,
given more space (notable ex
amples being Shirov - Bare
yev, Hastings 1 991 /92 and
lvanchuk - Kasparov, Linares
1 992), and it would have been
interesting to include a few
games from other openings,
such as the Ruy Lopez and the
Advance Variation of the Caro
Ka.ln. I hope however that I
have done enough to acquaint
the reader with some of the ba
sics of pawn chain strategy in
the two classic pawn chain
openings, the French and the
King's Indian. I hope that the
reader will enjoy exploring the
issues further, both in playing
through printed games, and in
h i s o r her over-the-board
chess. There is still one piece
of unfinished business in this
book; all those analytical ex-
1 0 1
ercises! Here are the positions
again, with my own attempted
sclutions.
A nalytical
Exer�lses
There is only so much that one c�n learn from the pas
sive study of texts; all top trainers are agreed that indepen
dent analysis is the real key to improvement. Such analysis
does not mean simple calculation of combinations; what is
required is that one should take an interesting position (or
even a boring position) and study it in depth, to find out what
is real ly going on. Only so is it possible for the reader to gain
a
more subtle appreciation of the relative importance of par
ticular types of strategic feature in game-li ke settings. Deep
analysis improves both tactical and strategical judgment!
The exercises that fol low have arisen from the discussions in
the text. Each exercise represents a point in the notes where
the sources I have used (mostly
lnformator)
fail to clarify
a critical position. Since in my own notes I have wanted to
concentrate on strategical themes rather than analytical dis
cussion, I have not written in detail on these positions in the
main text, but have left them as analytical exercises for the
interested reader. In this chapter, I present all the positions
again, and give my own tentative solutions to the questions
I have raised .. If you the reader disagree with my answers,
then excel lent! Analyse more deeply!
1 02
1 . Korchnoi - Kasparov (note to White's 1 7th)
White to move
A typically razor-sharp Classical King's Indian
What is happening here? Who is better?
2. Korchnoi - Kasparov (note to White's 21 st)
White to move
"Unclear" accordi ng to Kasparov.- Comments?
103
3. Introduction to Chapter 2
White to move
Black has just castled. Is the standard sacrifice 1 3 Bxh7+
good, bad or indifferent? In the French Defence it often takes
very fine judgment to be able to decide whether to allow this
sacrifice or not.
4.
Larsen - Bareyev lnote to White's 1 5th)
White to move
Black has just broken with 1 5
.
.
.
f6
(an opportunity that Bareyev
neglected). What is happening?
104
5. N imzowitsch - Salwe (note to White's 1 Oth)
White to move
Can White establish an advantage after Black's 1 0 . . . aS?
6. N imzowitsch - Salwe (note to White's 1 5th)
White to move
Another Bxh7+ combination to calculate. Is it strong?
1 05
7.
Korchnoi - Gel ler (note to White's 2
0
th)
White to move
Does 2 1 Bxd4 give White a stable positional edge?
8. Korchnoi - Geller (note to White's 36th)
Black to move
Black lost on time here, yet 36 . . . Kg7 1 stil l gives White tech
nical problems. Can White win this _endgame? I s Gufeld's
suggesti'on on 37 Re6 Rf6 38 Rc6 correct?
1 06
9.
Ljubojevic - M . Gurevich (note to White's 1 2th)
Black to move
Who
stands
better?
1 0:
Pyd o -
Likavsky (note to Black's 1 1 th}
White to move
This variation was once recommended as good for Black,
but h as been avoided in master chess. Why?
1 07
1 1 .
Pyda - Likavsky (note to Black's 1 7th)
Black to move
I s this position defensible after
1 9
. . .
Rg8?
1 2. Pyda - Likavsky (note to White's
1 9th)
White to move
A traditional "find the combination" exercise. Whit•
and win. Not as easy as it looks.
1 08
1 3. Piket - Kasparov (note
to
White's 1 7th)
White to play
White
played 1 7 a4. Assess 1 7 NbS. I s 1 7 . . . h5, as suggested
in
Encyclopedia of
Chess
Openings
the correct reply?
1 4. Piket - Kasparov (note to White=s 21 st)
Black
to play
Does 25 .
. .
·Nxg2 win? If
not,
how should play continue?
1 09
15. Piket - Kasparov (note to Black's 21st)
Black to move
If the win is "obvious", you will surely find it.
16. Cebalo- Cvitan (note to Black's 17th)·
Black to move
Black played 17 ... NeB; assess 17 .
. .
Rf7.
110
17. Cebalo- Cvitan (note to Black's 21st)
Black to play
Slight advantage to White, according to Cebalo.
But \Vhat is happening after 30
.. .
Qh4?
18. Mirkovic- Todorcevic (note to Black's 6th)
White to play
What is happening?
111
1 9. Mirkovic - Todorcevic (note to Black·s 22nd)
Black to play
" Unclear, according to Mirkovic. Clarify.
20. Mirkovic - Todorcevic (note to Black•$ 28th)
Black to play
How should he defend?
112
2 1 .
Olafsson - Petrosian
(
note to Black,s 1 1 th)
Black to move
Can White be punished for snatching the pawn?
22.
Olafsson - Petrosian
(
note to Black,s 25th
)
..
B lack
to
move
How
should White have met 25
. . .
Qf8?
Can B lack i mprove?
e
-
Pawn Chams
1 1 3
Exercise
I
May be a draw is a fair result, with 21 c7 being the critical
move. After 21 . . . Qd7 22 Rxa7 (22 Nxa7 leads to similar varia
tions, but without the preparatory exchange of rooks) 22
. . . Rxa7 23 Nxa7 gxh3 24 Ne1 Rg7 25 Bxh6 hxg2 26 Nxg2
N h4 27 Rf2 Nxg2 (27 ... Qh3?1 2B Bf1 1 ) 2B Rxg2
�"�2 -r 2�
Kxg2 Qh3+ 30 Kg1 Qg3+ 31 Kh1 is perpetual check.
2 1 hxg4 hxg4 22 fxg4 Nxe4 23 Bf3 would be interesting, with
White battling for control of e4 and hoping to use his passed
pawn, except that Black has 23 . . . Ng3 with a crushing king side
attack.
Exercise
2
White has just given back his extra pawn to save a tempo,
but what can he usefully do with his extra move? Certainly
something quiet like 23 a5? g4 'Nould be total ly wrong.
A White attac!� on the queens ide wil l not be able to outspace
a B lack kingside attack. What is necessary is to undertake
prophylactic measures on the kingside._ In the actual game
White defended against Black's sacrifice on g4 by Re1 , Bf3
and attempting to run .his· .king to e2: however he had not
prepared this sufficiently early, and lost as a result.
Therefore an immediate 23 Re1 !? Rh7 24 Kf1 is worth con
sidering, and if 24 . . . g4 25 hxg4 hxg4 26 fxg4 Bg5 27 Bf3.
After 27 . . . Bh4 2B Ra2 Bg3 Black still has good prospects
though, and there also alternative methods.
Another idea for White is 23 Nc6 QfB 24 N a7 , hoping for a
draw by repetition after say 24 . . . Bd7 2S NabS ReB 26 N a7
Rb8 27 NabS ReB. Black can sacrifice the exchange how
ever: 24 . . . Rxa7 25 Bxa7 c5, when after 26 Bb6? ! N h4 White
has too l ittle covering the kingside. So 26 N bS (idea Bb8) 26
. 114
. . . g4 27 hxg4 hxg4 28 fxg4 Nh4 (But not 28 . . . Nxe4? 29 Bd3)
and Black has very good play for the exchange, for example
29 Bf3 Rg7 30 Bb8 Nxf3+ 31 gxf3 Nxg4 or 29 Bb8 f3 ! Maybe
29 Ra3 Nxe4 30 Rh3 Bg5 is best, but Black still ha� excellent
play for the exchange. Black is looking good.
Ex-ercise 3
This is most spectacular of the analytical exercises! The
sacrifice is strong. 1 3 Bxh7+1 Kxh7 1 4 Ng5+ Kg8 ( 1 4 . . . Kh6
would lose quickly to 1 5 Qg3) 1 5 Qh4.
N ow B lack must make a flight square for the king. If h e
counter -aiial,;k�
with
1 5 . . . N d3+, then simply 1 6 Kf1 Qf2+ 1 7
Qxf2 Nxf2 1 8 Kxf2 f6 1 9 exf6 Rxf6 20 g3 leaves White
a
pawn up. If Black tries instead 1 5 . . . Rd8, then �n important
flight square is blocked after 1 6 Qh7+ Kf8 1 7 Qh8+ Ke7 1 8
Qxg7.
After 1 5 . . . Re8! White must be careful; the impulsive 1 6 Qh7+
Kf8 1 7 Qh8+? Ke7 1 8 Qxg7 Nd3+ 1 9 cxd3 Qe3+ 20 Ne2 Nd4
2 1 Qxf7+ Kd8 22 Qh5 Ra7 hands over the attack to Black.
White does better
to
spend
a
move getting his king into safeiy.
1 6 0-0-0 ! ! Now Rxd5 is a possibility, for example 1 6 . . . Ne7
1 7 Qh7+ Kf8 1 8 Qh8+ Ng8 1 9 Rxd5 ! b4 20 Nh7+ Ke7 2 1
Qxg7 with a mating attack. So Black tries to dislodge the
knight immediately.
1 6 . . . b4 ( 1 6 . . . Na4!?) 1 7 Qh7+ Kf8 1 8 Qh5 ! Now if 1 8 . . . Ra7
1 9 Nh 7 + Ke 7 20 Rxd5, winning; 1 8 . . . Nd8 receives the same
punishment and 1 8 . . . Qa7 or 1 8 . . . Qb7 are not much better,
too.
1 8 . . . Qc7 ! ( Black foresees a later . . . Qxf4+ in a critical line) 1 9
Nh7+ Kg8 20 Nf6+! gxf6
21 Qg4+ Kf8 22 exf6. Now Black
has a choice of losing variations, the second one being par
ticularly attractive.
1 1 5
(a 22 . . . Rd8 23 Qg7+ Ke8 24 Qg8+ Kd7 25 Qxf?+ Ne7 (25
. . . Kd6 26 Rxd5+1 ) 26 fxe7 ReB 27 Nxd5 exd5 28 Rxd5+
Kc6 29 Qxe8+ Kxd5 30 QdB+ winning.
(b 22 . . . Re7 23 Nxd5! exd5 24 Qg7+ Ke8 25 Qg8+ Kd7 26
Rxd5+ Ke6 27 Qg5 t ! and Black, although three pieces
ahead, is helpless. If this game hasn't already been played,
then no doubt one day it will bel
Exercise
4
The position is screaming out for 1 5 . . . f6, a thoroughly
thematic move to break White a grip on the center.
After 1 6 exf6 Black might also be tempted by the wild 1 6
. . . e5? ! which works out wel l in the sharp lines (e.g. 1 7 Rg3
Nxd4 1 8 _Rxg7 + Kh8 1 9 Qxe5 Nf5! ), but fails to the mundane
1 7 Nxc6! Qxc6 1 8 fxg7 Rxf4 1 9 Qxe5 Rf7 20 Bb5.
Black's position seems perfectly adequate after 1 6 . . . Rxf6,
which amongst other things puts pressure on the pawn on f4.
If 1 7 Rf3 Na5! with real queenside play, since unlike the game
1 8 Nb3 loses
�
pawn.
An object lesson in the importance of . .
J6
in the French.
Exercise
5
This is basically a question of strategy rather than one of
tactics. White plays 1 1 b5 ! , not worrying about the surrender
of c5 square, and then prepares Nbd2 and c4, opening up
the center rather than queen side. White's advantage in space
and development should then prevail. And White does not
even have to spend a move playing a consol idatory a4; after
1 1 . . . Na7 1 2 Nbd2 ! Bxb5 1 3 Rb1 Black must lose m aterial .
While on 1 2 . . . Bc5 preparing to meet 1 3 c4 with 1 3
.
. . Ne7,
then 1 3 a4! , played as a semi-waiting move, is strong. Black, s
bishop on c5 is badly placed, and if 1 3 . . . Ne7 1 4 Nb3 gaining
1 1 6
the bishop pair i n a position which will soon be opened up
(c4, maybe exf6, etc. ).
White stands better. Th
_
e general opening strategy is i nter
esting. First of a l l
Wh
ite closes the position to gain space .
. Then when shortage of space has constricted the movement
of the Black pieces, White opens up the position again to
exploit his lead i n development!
Exercl�e
6
1 7 Bxh7+
(a 1 7 . . . Kxh7 1 8 N g5+ Kg6! 1 9 Qxg4 Rf4! ( 1 9
. . .
Bxh2+? ! 20
Kh1 , and 20 . . . e5
21
Ne6+ Kf7 22 Nxc7 Bxg4 23 Kxh2 ! or
20 . . . Rf4 2 1 Q h3 Kxg5 22 Qxh2 Rh4 23 Be3+ Kh5 24 g4+
Kxg4 25 Rg 1 + wi nning) 20 Qh3 Kxg5 2 1 Qg3+(21 Be3 Kg6
leaves B lack be
t
ter) 21 . . . Rg4!(21 . . . Kf5 !? 22 Qd3+ Kg5 2.3
Qg3+, etc. ) 22 Be3+ (22 f4+ Bxf4) 22 . . . Bxf4 . Good enough
reason already for rejecting 1 7 Bxh7 +.
(b 1 7 . . . Kh8 also seems amply playable for Black, e. g . 1 8
Ng5 Bxh2+ 1 9 Kh1 e5 20 BcS Qc6 2 1 Bxf8 Rxf8 22 f3?
Ne3 23 Qxe3 Bf4 with a big advantage to Black. Alterna
tively 1 8 Bc2 e5 and Black has taken over the center.
Therefore 1 7 Bxh7 + is not strong, and so Black stands well
in diagram position Nimzowitsch of course did not reach the
diagram position, which resulted from a hasty l ine which left
insufficient control of the cente
r
.
As a matter of general principle, it is usually best to play the
positionally l_ogical move (e. g
.
overprotection of one's own
strong points, restraint of the opponent
'
s position) rather than
to play "attacking" moves and rely on rando
m
tactics to back
up the attack
.
Often as here random tactics do not work.
1 1 7
E:xerelse
7
Not an easy position to assess, or to analyze. 20 Bxd4 is
certainly the natural move for White, giving Black the choice
between stabil izing the pawn structure with 20 . . . exd4 or un
blocking the queenside with 20 . . . cxd4.
After 20 . . . exd4, B lack's e6 square is exposed, and he must
be very careful not to exchange his light squared bishop too
l ight-heartedly. After 21 f4 Bf6 (to prevent Ng5) 22 Re1 Be7
23 g3 White stands better and can play to increase the pres
sure with Qd2 , a5 and Ng5. Black's only real chance of play
is if VJhite rushes things too quickly, for example 23 . . . Bd7
24 h4? !
(24 t:aG)
24
. . .
Bxh4.·
Sight advantage to White then.
if
20 . . . cxd4, White's main responsibility is to keep his strong
point on e4, and indeed to overprotect it. With B lack having
no real play in the center or kingside, White can turn his
attention to the queenside. After 21 Qe2 White can consider
Rfc1 or even doubling rooks on the a-file to push the pawn
forward to induce . . . Ba6 leaving some weak light squares on
the kingside. If Black wants to force opposite colored bish
ops with 2 1 . . . Bf5, then let him
I
The bad bishop on g7 can do
little to guard the queenside, a··perennial problem in the King's
Indian.
I hope you did not
try
calculating too many variations in this
exercise! Analysis can be as much about strategy as about
tactics.
E:xerelse
8
The first point to consider is that Gufeld's suggestion of 36
. . . Kg7 37 Re6 Rf6 38 Rc6 Rxc6 39 dxc6 is unnatural; White
is voluntarily weakening his pawn structure! After 39 .
. .
Kf6
B lack should almost certainly be able to draw by picking off
the c-pawn with his king if the White king should move to the
118
queenside, and by keeping guard on g5 if White tries Kg2-
g3-g4.
A
more logical approach is to force the exchange of rooks
without compromising his pawn structure, by playing 37 Re7+
Rf7 38 ReB. Now 38 . . . Rxf4 is hopeless: 39 Ra8 Rf6 40 Ra7
Kh6 41 Kg2 1 (cutting out the last glimmer of counterplay) and
Black can resign. Black must therefore fal l in with White's
plan by playing 38 . . . Rf8 39 Rxf8
Kxf8.
Although White is an outside passed pawn u p , the win is
very difficult since the bishop is tied
to
the defence of the c
pawn, and since the kingside pawn majority is compromised
by the doubled f-pawns. The natural plan is
40
f5? 1 gxf5 41
f4; we must see why this does not work. After 4 1 . . . Ke 7 42
Kf2 White's king is in range of the Black pawn on c5, mean
i ng that 43 Bxf5 is a threat.
Thus if 42 . . . Kd6 43 Bxf51 Bxc4 44 Bxh7 1 and Whitels three
widely scattered passed pawns count for more than Black's
tightly bunched pawns. Neither though is defendi ng passively
an opinion for Biack: 42 . . . Kf6? 43 Ke 1 Kg6 44 Be2 followed
by playing the king to d3, covering the vulnerable c-pawn, .
and then playing the bishop to b5.
·
So how does Black defend? He must get h i s h-pawn to a
dark square, beyond the reach of the White bishop: 42 . . . h61
I
I
have been unable to find a White win here. A typical line is
43 Bxf5 (before Black can play . . . Kd6) 43 . . . Bxc4 44 Bc8 Kd6
45 a6 Bxa6 46 Bxa6 Kxd5. B lack's queenside pawn phalanx
is sufficient to_hold the game, as will readi ly be seen by play
ing out a few variations. A critical technical poi nt needs to be
noted here; Black draws the game if he can win the f-pawn,
even if he loses all his queenside pawns, as the bishop is the
wrong color square for the rook's pawn. In the other words,
preserving the h-pawn is of no help in White's bid for a win .
1 1 9
The ultimate useles�ness of the White's h-pawn is not so
easy to see in advance, especially on move 40, but it does
help bring · to mind
�
possible alternative plan. White must
attack Black's kings ide pawns with h4-h5, th
r
owing away the
expendable h
-
pawn
,
rather than with f4-f5. Therefore 40 h4! !
is the move. If Black now tries
40
.
. .
h5, the simplest is
41 to
gxf5 42 f4 fol lowed by bringing the king to b3 or d3
,
protect
ing the c-pawn, and then winning either the f-pawn or the h
pawn. It fol lows that Black cannot preven
t
h5
.
If 40 . . . Kf7 4 1 . h5 gxh5, White has a fairly straightforward win
by marching the king up the h-file, gobbling pawns and even
tually squeezing through the king up the h-fi le. Black must try
to preserve a pawn on g6. So: 41 . . . Kg7 ( it is also desirable
not to allow White to play h6
)
42
Kg2 Kh6 43 Be2 Kg7 44
Kg3 Kh6 45 Kh4 Bc8 (45 . . . Kg7 46 Kg5! ) 46 hxg6 hxg6 47
Bd3 Ba6 48 Kg4 BcB+ (48 . . . Kg7 49 Kg5; the Black king is
misplaced) 49 f5 ! ! gxf5+ (49
.
. Bxf5+ 50 Bxf5 gxf5+ 51 Kf3; 49
. . . Kg7 50 Kf4 ! gxfS
51
KeS)
50
Kf4 Kg6 51 Ke5 Kg5 52 d6!
cxd6+ 53 Kxd6 and White soon mops up.
There was a lot of fight left in the position when Black l
o
s
t
on
time.
··
Exerebie
9
White stands better! He has maintained his pawn chain,
since sacrifices on e5 are unsound, and he is ready to sort
out his kings ide development and take the i nitiative.
Main variations:
(a 1 3 . . . Ndxe5? 1 4 fxe5 Nxe5 1 5 Nxe5 Bd6 1 6 Qe1 !? Qd4 1 7
Bf4 Rxf4! 1 8 Nxf4 Bxe5 is unclear, �espite Black's heavy
sacrificing: There is however an unusual twist: after 1 4
Nxe5! Nxe5 1 5 fxe5, the fact that White has a pawn rather
than a knight on e5 means that
. . .
Bd6 is u nplayable
.
White
1 20
is simply a piece ahead.
(b 1 3 . . . Nc5 concedes that the pawn chain cannot be broken;
B l ack decides instead to acti·'late his own pieces. White's
most appropriate response is to get his king into safety,
avoiding random tactics. Thus 1 4 Kh2 is indicated, and if
1 4 . . . Ne4 1 5 Nfg5 knocks out Black's e4 strong-point. Black
must avoid 1 5 . . . Nf2? 1 6 Qc2, but the alternatives seem
good for White. If 1 5 . . . Nxg5 1 6 hxg5, White develops with
Bd3, etc.
Ex�rt-lse 1 0
I
do not know in how much detail this position has previ
ously been analyzed, but the critical line would seem to be
1 3 Nb3 fxe5 1 4 Qxg41 N ow:
(a 1 4 . . . Nf6 1 5 Qg7 Rf8 1 6 dxe5 Ne4+ ( 1 6 . . . d4 1 7 exf6 is
good fer White) 1 7 Kf3 d4 1 8 Kxe4 (1 8 Qxh7?? Ng5+) 1 8
. . . dxe3 1 9 Be2 and B lack's king on e8 is more exposed
than White's king on e41 P l ay is of course extremely sharp,
but probably in White's favour after for example 1 9 . . . Ne7
20
Bh5+ Kd7 {Hop_ing for 2 1 Qxf8? Qc6+ when Black is at
least equal) 2 1 Rc1
I
{Black stands well after 21 .Rd1 + Kc7
22
Rd6 Bxd6 23 exd6+ Qxd6 24 Qxf8 Qd5+ 25 Kxe3 Nf5+
26
Kf2 Qxh1 27 Nf3 b6 1 ) 2 1 . . . Rg8
22
Qf7 Qb5 23 Nf3 ! ,
and White wins the race to complete his development ( ! )
and give a kil ling check.
(b 1 4 . . . exd4 1 5 Qxe6+ Kd8 1 6 Bxd4 Nxd4 1 7 Qxb6+ Nxb6
1 8 Nxd4 B c5 1 9 Ngf3 Bg4 20 Rd1 gives White a solid extra
passed pawn. Black's bishop pair and superior develop
ment offer only transitory compensation.
(c 1 4 . . . Nxd4?? 1 5 fxe5 is simply·a blunder.
1 2 1
Some tactical sequences here which are typical of the French
Defence, and of the f4 Tarrasch in particular, in that B lack is
desperately trying to create tactiLal mayhem before White
can consol idate and make use of his considerable space ad
vantage. One should not be dogmatic, but on the whole such
position ought to favour White.
Exercise
I I
Yes ! Black stands well, thanks to a few thematic defensive
combinations:
(a 20 Bxe6 Ndxe5! 21 Bxg8 Ng4+ 22 Kg1 Nxe3 23 B xh7 Bg4'
(not 23 . . . Nxc2 24 Bxc2 Bg4 25 Qd2 ! ) 24 Qd2 Bxf3 25 Rc3
Bb4 ! etc .. White has alternatives of course, but B lack al
ways comes out ahead.
( b 20 Nhg5 Bxg5! 21 Nxg5 (21 Bxg5 Nxd4! ) 2 1 . . . Rxg5 22
Bxg5 Nxd4 and White's position col lapse�.
(
c
20 Nxh4 avoids immediate disasters, but White has little
to show for the pawn he h�s sacrificed.
Does thts mean that White's whole idea is wrong? Not so;
White's play is fine, but it is his note that is wrong. F rom the
diagram position, retract White, s last move, 1 9 Qd 1 -e2, and
play instead the more aggressive 1 9 Nhg5! An extra tempo
for the attack matters more than the bishop on e3 !
8
lack gets
mated if he tries 1 9 . . . hxg3+ 20 Kxg3 Qxe3 21 Rxh 7
+
Kg8 22
Bxe6+, so the defence of the bishop is wholly unnecessary,
yet if Black tries something else, White opens up the h-file
anyway.
1 22
Exercise
12
2 1 Bxf3 Bxg5 22 Be4! (Since 22 Bxg5 Qxd4! or 22 Bxd5
Rf5 is fine for B lack; RR - 22 Qh1 ! is also strong) 22 . . . dxe4
and now White can choose between 23 Rh8+ !? Kxh8 (23
. . . Kg7 !?) 24 Qh5+ Kg7 25 Qxg5+ Kf7 26 Rf1 + Ke8 27 Qg6+
Kd7 28 Rxf8 Qxb2 when White stil l has to prove the win, or
23 Qh5 ! Rf3+ 24 Kg2 Qxb2+ 25 Kh1 with a winning attack,
e.g. 25 . . . Bxe3 26 RhB+ Kg7 27 Qh7+ Kf6 28 RfB+ Kg5 29
Qg7 + Kh5 30 Rh8+ and mate·next move.
Exercise 13
1 7 Nb5? is a weak move, since in abandoning control of
the e4 square White allows an immediate 1 7 . . . g4! in repy. 1 8
Nxa7 is of course met by 1 8 . . . g3, and 1'8 Bxa7 is met by 1 8
. . . Rxa 7, so the threat to the a-pawn is shallow.
ECO ( E99/4), note 22, gives the strange sequence 1 7 N b5(?)
h5(?) 1 8 Nxa7 Bd7 1 9 a4 g4 20 a5 g3 21 Bb6 gxh2+ 22 Kh1
Qe8 23 Nd3 h4 24 Nf2 Nh5 25 Kxh2 Ng3 26 Rg1 h3 unclear
(Black has compensation for the sacrificed material), Piket -
V.Spasov, Groningen 1 989.
A checking of sources shows however that the position after
1 7 . . . h5 was reached by transposition, and that neither player
in fact made an obvious positional blunder. The sequence of
moves was 1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4 e4
0-0
5 Nf3 d6 6
Be2 e5 7 0-0 Nc6 8 d5 Ne7 9 Ne1 Nd7 1 0 Be3 f5 1 1 f3, and
now not 1 1 . . . f4 1 2 Bf2 g5 1 3 b4 Nf6 1 4 c5 Ng6 1 5 cxd6 cxd6
1 6 Rc1 Rf7 1 7 Nb5(?) Rf7(?) as suggested in ECO, but rather
1 1 . . . h5 1 2 b4 f4 1 3 Bf2 Nf6 1 4 c5 g5 1 5 Rc1 Ng6 1 6 cxd6
cxd6 1 7 Nb5 ! ? Rf7. Such an incident shows the need for
care when consulting reference works on the opening.
On the subject of transpositions, it is perhaps only fair to point
1 23
out that in the Korchnoi - Kasparov game in chapter 1 , the
move order to reach the position after move 8 was not the
standard King's Indian move order, but rather 1 Nf3 Nf6 2 c4
g6 3 Nc3 Bg7 4 e4 d6 5 d4
0-0,
etc.
r;xerclse 14
25 . . . Nxg2 is of course the thematic move, thoroughly ex
posing the White king, but Black is so far behind in material
that the best he can t,ope for is a draw. Thus 26 Kxg2 Rg7+
27 Kh2 Ng3! (always sacrifices . . . ) 28 Bxg3 Qg5 29 Rg1 Qh5+
30 Kg2 Be3 3 1 Rh1 Rxg3+ 32 Kxg3 Qg5+ with a draw. I have
to admit
t�::t ! ::t this
exercise not believing that Black had
"only" a draw after 25 . . . Nxg2; I wanted a further opportunity
to examine the positron. It is important to note that in the
above line 28 . . . Rxg3 is not a win; 29 Kxg3 Qg5+ 30 Kf2
Qh4+ (with White being two rooks and a piece ahead, the
king cannot be allowed to escape to e1 ) 3 1 Kg1 and now the
rational way is 31 . . . Qg3+ with a draw, since 3 1 . . . Be3+ is
met by 32 Rf2.
So if B lack is aiming for a win he should try 25 . . . Rg7. Para
doxical ly this looks even bolder than the piece sacrifice since
B lack, a whole rook down, gives White a tempo to close down
the kingside. Still, Black has his opportunities to reduce the
material imbalance ( . . . QxaB, maybe later . . . Bxc1 ), and the
White kings ide is still leaky. Kasparov gives in
New in Chess
the l ine 26 Nxf4 Bxf4 27 g4 (27 Rc7? Ng3 ! - the familiar py
thon grip - 28 Rxd7 Qh4 29 Rxg7+ Kxg7 30 Bxa7 Nxe2+ 3 1
Qxe2 Bh2+ 32 Kh1 Bg3+ and mate next move) 27 . . . Bxc1 28
Qxc1 Nf4 29 Qe3 h5 and the material deficit is under control
with B lack still having a kingside attack. Kasparov suggests
that White can still defend with 30 Rc1 hxg4 31 fxg4 Nxe2+
32 Qxe2 Bxg4 33 Qe3, but here is the only point where I can
1 24
disagree with his analysis. Why should Black exchange his
dominating knight for White's weak bishop? 3 1 . . . Qxa8 lev
els the piece count and continues to maintain the tension on
the kingside.
There is however a third alternative at move 25, mentioned
by neither Kasparov nor Nikitin: 25 . . . Bh3 !? It gives nothing
clear though after 26 Nxf4 Nxf4 27 g4: and if 27 . . . Q a8 28
Rc3!
25 . . . Rg 7 is best.
One of the few analytical exercises where even after pro
longed thought I have been able to add nothing of real sub
stance to the published analysis; but should this be surpris
ing when the analysis has been done by
a
world Champion?
Exerelse
IS
Quite a ta.1talizer! Black is only a piece down and appears
to have the White king at his mercy, but it is i ncredibly diffi
cult to find the way o
�
landing the killing blow when there are
so many tempting alternatives. The problem is that since Black
is working on a very narrow front his pieces are l i able to get
in each other's way, for example 26 . . . Be 7 27 Bf1 ! Bh4+ 28
Kg 1 and suddenly Black is wishing he had h4 for the knight
or queen.
An important practical point: Black is not worse. If u nable to
find a win he can always settle for a draw with 26 . . . Qg3+ 27
Kf1 Qh3+ 28 Kf2 (28 Kg1 ? Nh4 and . . . Rg7+) 28 . . . Qg3+.
26 . . . Ng3 is tempting, but the White king escapes after 27
Nd3 ! Qh4 28 Ke1 !
26 . . . Qh4+ 27 Kg1 is also unconvincing, since with h4 un
available for a Black knight White can stick a m inor piece on
g2, for example 27 . . . Ng3 28 Ng2.
125
26 . . . N h4 ! , as suggested by Kasparov and Nikitin, is in fact
the winning move, and if 27 Nd3 Qg3+ 2B Kf1 Ng2 ! ! The
escape route via e1 has been b:ocked off, and Black has
time to bring more pieces in to finish everything off, for ex
ample 29 Qd2 (29 Bf2 Qh2 ! ) 29 . . . Qh2 30 Bg1 Ng3+ 31 Kf2
Nxe4+ 32 fxe4 Qg3+ 33 Ke1 Ne3+ 34 Bxe3 fxe3+ and wins.
Exerelse 16
G iven that Black is playing for . . . g4, there is an obvious
presumption that 1 7 . . . Ne8, playing the knight away from f6,
i s a gross waste of time. And yet . . .
If White were to play exactly as in the game, then clearly
1 7
. . . Rf7 would appear as a great improvement on 1 7 . . . NeB.
The reason why Black preferred 1 7 . . . NeB was anxiety about
another plan, the "paradoxical push". On 1 7 . . . NeB 1 8 g4
fxg3 1 9 hxg3 Bd7 Black's pieces are reasonably wel l coordi
nated, but after 1 7 . . . Rf7 1 8 g4! White is better, much as in
the illustrative game Lebron - Renet. Black is going to spend
a tempo defending d6 before he can remove the knight.
A puzzl ing decision can thu.s be explained by the fact that
the opponent has flexibility of plans; what is good against
plan A is not necessarily 990d against plan B.
Exercise 17
White is in fact very comfortable, dangerous though B lack's
attacks may appear. After 30 . . . Qh4 31 Rff3! Rg3 32 Rxg3
fxg3 33 Qf3! White's king wil l escape without trouble via e2.
The main line i·s 33 . . . Bg7 34 Rc8+ RxcB 35 Nxc8 Ng4 36
Ne7 ! (36 Nxd6? Nf2 37 Kf1 Qh1 + 38 Ke2 Qd1 + 39 Ke3 Bh6+)
36 . . . Qxe7 37 Qxg4 and White wins the endgame.
So White did the right thing, in Cebalo's note, by getting on
with his queenside play, watching the center, and meeting
126
Black's kingside break when it came.
Exerelse 18
A lot of pawn exchanges follow, with White trying to get a
grip on the empty e4 square. 9 gxf5 gxf5 1 0 exf5
( fO
Rg1 0-
0
is also worth considering, but the attempt at exposing the
king with 1 1 Bh6 Rf7 1 2 Bxg7 Rxg7 1 3 Rxg7+ Kxg7 simplifies
the position too much) 1 0 . . . Nxf5_ 1 1 Ng3 (1 1 Nf4 Nf6 ! is u n
clear) 1 1 . . . Nf6 1 2 Rg 1 !
Often White has to be very careful in preparing the double
exchange on f5, especially when he has spent a tempo pre
paring g4 with h3, since Black has plenty of targets along the
f-fi le. Here though the tempo count is very much in White's
favour, and he will soon gain a massive grip on the e4 square,
the b 1 -h7 diagonal and the g-file, fo.r example 1 2 . . . 0-0 1 3
Bd3 Nd4 1 4 Nh5.
Exe:-clse 19
It looks at first as though Black is making a lot of ground i n
the center and o n the queenside, but appearances are i l l u
sory. B lack's big problem is that he cannot bring his knight on
h 7 into the game; none of White's pieces is similarly restricted.
25 . . . Nxb2+ is met not by Mirkovic's 26 Kc2?! , when 26 . . . N a4 !
i s dangerous (27 Nxa4 Qb4), but rather by 26 Ke1 ! After some
minor tactical chaos, White stays in control, for example 26
. . . Q b4 27 Qc2! Kxg7 (27 . . . Na4 28 Bd2 Nxc3 29 Bxc3 Q b 1 +
30 Kd2 is hopeless for B lack; g6 is coming) 28 g6 fxg6 (28
. . . Nf6 29 Bg5! Na4 30 h6+ Kxg6 31 Bh4+ Kh7 32 Bxf6 wins)
29 Bh6+! and Black has no satisfactory reply (29 . . . Kxh6 30
Rxg6 mate).
Which leaves 25 . . . Qxb2 26 Qxb2 Rxb2, assessed by Mirkovic
127
as "unclear... The simple 27 Nxe4 is probably winning for
Wh ite though; i n one move he has taken a pawn, protected a
key square
(f2)
and advanced his own attack (g6 is now dan
gerous for Black).
Since 27 . . . ReB and 27 . . . Kxg7 both lose quickly to 28 g6, it is
wise for Black to play the decoying move
L t
. . . c�. li 1�1 � lui
lows 28 Nxc3 Rxf2 (28 . . . Nxf2+ 29 Bxf2 Rxf2 30 g6 Nf6 3 1
gxf7+ Kxf7 32 h6 must ultimately be a win for White) 29 g6 !
(29 Bxf2 is unclear) 29 Nf6 (There is nothing to be gained by
an immediate check on b2; Black holds back) 30 h6! (An
important zwischenzug, threatening 31 h7+ Kxg7 32 h8=Q+
Rxh8 33 gxf7+) 30 . . . fxg6 31 Rf1 Nb2+ 32 Kc1 Nn3+ 33 Kb 1
Rb8+ 34 Ka1
Rxf1
+ 35 Rxf1 . White with his two mighty passed
pawns would seem to be winning; if 35 . . . Nb4 36 a3 Nc2+ 37
Ka2 Nxe3 38 Rxf6 with total control .
Exercise
20
If in doubt, centralize, and remember which color squares
you are strong on ! Therefore 28 . . . ReS ! followed by . . . Qe 7 i s
the appropriate plan. There i.� a lot of fight left i n the position,
and a lthough B lack's kingside is a l ittle wobbly, it should not
col lapse. For example 29 gxh8=Q+ Kxh8 30 h6 Ng5 (to pre
vent Rg7)n31 Rxf4 Qe7
32
Rf5 (32 Rxc4 Rb4 ! ) and Black i s
comfortably barricaded on the dark squares, while White is
beginning to look i nsecure.
Exercise
2 1
The intrusive White queen cannot quite be won, but there
is no easy escape either. With ac�urate play Black is very
much in control of the game.
B lack plays 1 3 . . . h6 1 4 Bxe7 Nxe7 1 5 Nf4 (On 1 5 Qf7 Rdf8
1 6 Qh5 c5! is very strong) 1 5 . . . Rdf8 (to prevent the queen
1 28
attacking the e-pawn) 1 6 N g6 Nxg6 1 7 Qxg6 Rhg8 1 8 Qh5
Qg7 and White is in massive trouble. 1 9 . . . Nxf4 is a direct
threat, while White must also be careful not to allow the ma
neuver . . . Bc6-e8.
Exercise 22
In the l ine 25 . . . Qf8 26 Bxe6 Be5, White can improve on
the suggested 27 Bxd5?1 by playing 27 Be3 1 with the threats
of Bxb6 and a capture on d5, and possibly als�Qd7. Black's
pieces are so badly tied up _on the kingside that White is
better, despite being a piece down. Unless you the reader
can suggest something else . . .
Black also has defensive problems after 25 . . . Qf8 26 Bxe6
Nxf4 27 Nxf4 Be5 28 Nxd5, giving strength to the impression
that 25 . . . Qf8 is too passive. If you want to defend by curling
up
like a hedgehog, it is better to do
s o
in front of the king,
and not on the opposite side of the board !
The outgoi ng 25 . . . Qc5! is better, the main line being 26 Bxe6
Nxf4 27 Nxf4 Bxf2 28 Bxf7 Bc6 29 Qb3 (29 Bxd5 Qxd5! )
29
.
.
. Rf8 and B lack regains the extra piece; or 29 Qb4 Qxb4 30
axb4 Be3+ 31 Kb1 Bxf4 32 Bxg8 Rxg8 where White reaches
an endgame but a prospectless one.
Even in his best games, Petrosian was occasional ly liable to
play over-defensively.
QU.rleh_
eh
ess
llooks and
for
everylaody
Russian Chess Review No.1
A successful Russian-German coprodu- ..
ction.Russian Chess Review (or, from the third
issue, International C hess Review) aims to
become the leading collection of the most
important games from the theoretical and
c�eative viewpoints, played
by
the world's
foremost chessplayers. The Wor1d Champion
Garry Kasparov writes in our foreword, ,.I wish
success to your new publication which may
bec�me a reliable guide in the turbulent stream
of today's chess theory".
430
pages with
1 200
games,
1300
diagrams
and 85 photographs.
ISBN 3-929324-00-8
International language
DM
3A 80
Russian Ches4t Reviev No.3
The "I..Jew Chess Periodical". A successful!
German-Russian cooperation Russian Chess
Reviev is an international periodical. Starting
with the 3rd issue, RCR deals also with chess
theory and convinces through its coverage and
quality. Besides its
400
matches together with
extensive commentaries by Russian grand
masters and masters each issue contains
approximately
800
additional matches without
comments. It will be publish from Sept. 1 994
.
.
430
pages,
1 200
games,
1 000
diagrams.
ISBN 3-929324-1 6-4
International language
DM
36.80
Jan Przewoznik
Teat Your Chess Fantasy
A collection of
1 20
chess exercises of varying
degrees of difficulty in which the player may
work to examine and improve this chess play
ing strength, unencumbered by stereotypical
thought patterns.
1 50
pages,
1 20
diagrams.
ISBN 3-929324-09-1
English or German language
DM 24.80
Russian Chess Reviev No.2
The resounding success of the first issue has
paved
the way for the continuation of this pu
blication, which-with its theoretical coverage
is unique in the wor1d.
420
pages,
1 000
games and 65 photos.
ISBN 3-929324-07-5
International language
K
ar
p
ov
/Gag
arin
Secreta From Russia
Chess Theory and Analysis
DM
36.80
This new publication gives a comprehensive
insight into the private laboratory of the
Russian chess school. The team of authors
wort<ing alongside Champion Anatoly Karpov
has over forty grandmasters!
480 pages,
650
books of the year!
ISBN 3-929324-1 3-X
English language
Dobosz/Oibrich/Przewoznik
Kandidatenfinale '93
Short - Timm
a
n
DM 39.80
For the first time in
21
years a western grand
master will challenge for the Wor1d Cham
pionship. But was it going to be the English
man Short or the Dutchman Timman? The
Candidates Final between the two has been
thoroughly analysed by a practical players,
Henryk Dobosz, Marina Olbrich-Sokolova,
and a theoretician Jan Przewoznik. In their
commentary they pose the question of how
Wortd Champion Gany Kasparov would him
self have reacted to the various match
situations. Also, specially for the English edi
tion, a supplementary chapter asks what
chanses Short has got in the big match later
this year.
1 60
pages,
1 20
diagrams and photographs.
ISBN 3-929324-04-0
German
lan
g
uage
D.M
24.80