A 1st printing of this paperback edition

by Chess Digest, Inc., authorized by

Sportverlag Berlin.
Chess Digest, Inc. 1992
0-87568-218-9

Send the publisher $2.00 for the New

Chess Guide that catalogs every chess
book for general sale in the United States.
You are given publishers, page counts,
notation, and critical reviews. Also in­
cluded is a free Chess Improvement
Course for Beginners up through Master
level players.

Chess Digest, Inc

P.O. Box 59029
Dallas, TX 75229
Yuri Averbakh

Chess Tactics
for
Advanced Players

Sportverlag Berlin
Averbach, Jurij L'vovic: Chess tactics for advenced players/Yuri Averbakh.
[Transl.: Ruslan Tulburg]. - 2. ed. - Berlin: Sportverl., 1986. - 326 S.:
zahlr. Diagr.
Dt. Ausg. u. d. T.: Averbach, Jurij L'vovic: Schachtaktik fUr Fortgeschrit­
tene

ISBN 3-328-00173-5

© Sportverlag Berlin
Second Edition
Licence No.: 140 355/43/86
9095
Translator: Ruslan Tulburg
Title of the German Original: Schachtaktik fUr Fortgeschrittene
Cover Design: Erika and Peter Baarmann
Printed in the German Democratic Republic
671 526 3
Introduction
We are witnessing at present an un­
dreamt-of upswing in the evolution
of the theory of chess in general and
of the theory of openings in particu­
lar. Each year, fresh tomes devoted
to the study of various opening
games or particular opening sys­
tems or even opening variants ap­
pear on the book shelves. Nowa­
days the analyses of some openings
continue deep into the middle game
and even extend to the end-game.
Not quite so explicit, but still very
thorough is the study of the end­
games. Dozens of positions, which
were once blank spots on the map
of the end-game, have been ana­
lysed in detail, assessed and in­
cluded in textbooks on the end­
game.
Turning to the theory of the middle
game we find that the progress
made here is by no means as impres­
sive, the development of this phase
being characterized mainly by a
rapid growth in the volume of data.
Books on the middle game usually
contain hundreds of positions, but
unlike their predecessors, the prog­
ress made here is reflected chiefly
by an inreasing number of examples
rather than by a more profound un­
derstanding of this fascinating
phase of the game. But the search
for new opening systems and their
elaboration by chess practitioners
have enhanced the strategy of the
middle game, because many differ­
ent types of middle game positions,
which are characterized by a gen­
eral plan or a general strategy, have
been analysed extensively and in
depth. Tactics, on the other hand,
are dominated by chiefly a matter
sizing up a concrete situation,
which makes it difficult to arrive at
generalizations.
Of course no one can maintain that
the theory of tactics has been mark­
ing time. Attempts are made to clas­
sify combinations and work out
ways of facilitating the calculation
of intricate combinations. And yet
on the whole the development of
the theory of the middle game lags
perceptibly behind the other two
phases of the game.
The explanation commonly given is
that the middle game is the most in­
volved and difficult phase of the
game. It is by no means easy to un­
derstand it and its laws. That is pre­
cisely why the middle game-more
than any other phase-calls for a
circumspect approach, for easily
comprehensible principles and for a
gradual transition from the simple
to the complex.
Anyone who has taken an interest
in engineering knows that one must
first familiarize oneself with the de­
tails of a machine before trying to
understand how it works. And yet it
is in the study of the middle game
that the basic principle of peqagog­
ics-the necessity of proceeding
from the simple to the complex-is
constantly being violated.
As a rule one begins learning the
5
middle game and chess tactics by
trying out combinations. Unques­
tio nabl y , this is one of the most fas­
cinating and exciting aspects of
chess . The mechanics of even the
most elementary combi nation are
extrem ely complex. It is advis a ble to
famil iarize oneself thoroughly
with its elements before studying
t he comb i n a t i on . Whe n chess wr i­
ters describe chess combinations
they usu ally stress the aesthetic a nd
emotional aspects) namel y the sacriH
fice, the s urp ris e effect and tht� par­
adoxical element of the game.
Without a doubt these aspects are
also important, be cause the y brin g
out the fa s ci nati n g side of t he game.
But it is even more important to un­
ravel the myste.ry of the complex
mechanics of the combinat ion and
to sho w its main elements and how
the y funct i on .
And still the attempts made so far to
explain the combination mechanism
have not yielded more tha n ·a hand­
ful of general and not ve.ry con clu­
sive conce pts, such as harmony of
the pieces or the concerted use of
forces. Besides, the debate about
what the term "combi nation" means
is still conti nu in g . Different authors
have i nterpreted the term in diffe r­
ent ways. The reader will recall that
a discuss ion launched by the Sov ie t
journa l "Sh a khmaty v SSSR" on the
topic "What is a combi n ation ?"
failed to produce a good answer, al­
though it took up a good deal of
space in the periodical.
But h ow is one to classify and study
combinations if the re is no consen­
sus about the meaning of the term?
6
Of the combination it may be said
with perfect truth that everybody is
fascinated by it, that everybody un­
derstands what it is, but that no­
body seems to be able to give an ex­
act definition of the term.
The present book is an attempt to
rescue the theoretical discussion
from this impasse. We have endeav­
oured to establish a theoretical basis
that would enable us to pursue
our study of the extremely diverse
tactical manifestations.
We have started by examining the
most common and elementary situ­
ations arising from confrontations
between isolated pieces and con­
sider possible methods of attack
and defence. Then we gradually
proceed to examine the more com­
plex facets of the game, making
sure to preserve continuity by delv­
ing deeper into positions that have
already been analysed. We soon
reach the point at which we can ex­
amine positions involving double
attacks. If we regard the term
"double attack" in a broader sense
than has been done up to now by
theoreticians, namely not merely as
a two-pronged attack, but as a com­
bination of attacks and threats, we
notice that the double attack in one
form or another is the basis of most
intricate tactical- operations.
Considerable space is devoted to
this topic to convince the student of
the truth of this statement. We soon
realize that the double attack is a
very effective device not only for at­
tacking but also for defending one's
position.
The author is convinced that this
approach makes it easier to find
one's way through the mare of tac­
tical entanglements and provides a
key to understanding the problems
of the tactical game.
The next important step is to find
out the meaning hidden behind
such mysterious terms as "harmony
of pieces" or "co-ordinated use of
forces". Our aim here is to uncover
the meaning and the essence of each
term.
This enables us to introduce the
term "elementary connections"
which we have coined. We discover
that any tactical operation no mat­
ter how complex, can be described
in terms of different combinations
of contacts. We soon realize that
the concerted use of forces boils
down to the co-ordinated use of
combinations of contacts, i. e. to
certain constellations of pawns,
pieces or squares on the chess
board. It is also demonstrated that
most combinative end-games can be
reduced to a few basic co-ordinated
combinations of contacts.
With the aid of the term contacts -
and their combinations a definition
of the combination is worked out
and a new and promising system of
classification is introduced.
But to obtain a thorough knowl­
edge of the theory of contacts and
their combinations proposed by the
author the reader will have to read
the book. It is intended to help him
fathom the sources of the combina­
tion and to understand the origin
and function of its mechanics.
Working his way through the book
the reader will realize that various
kinds of combination of chessmen
and squares on the board must be
preceded by various contacts be­
tween them. And it is precisely these
combinations aimed at specific tar­
gets that form the basis for develop­
ing combinative possibilities. We
believe that we can proceed, with
good prospects of success, not only
to acquainting the reader with the
various combinations occurring in
chess, but also to showing him how
to recognize them in their incipient
stages and how to prepare them af­
ter making a detailed study of the
theory of contacts and their combi­
nations.
In this work we have also been able
to answer a fundamental question
that has occupied the minds of chess
theoreticians for many years,
namely whether combinations are
an exception or the rule. Contrary
to the notions of adherents of the
positional school, who regard the
combination as an avoidable excep­
tion, we propose to demonstrate
that the development of contacts
and their combinations is a natural
development of the chess game.
Since the appearance of contacts
usually heralds the combination, it
evolves as a complex and various
consequence of this process.
With this work we give the recep­
tive reader a book that is in the true
sense a textbook of tactics and com­
binations. It is designed to help the
reader master the art of combina­
tion, which is the highest level of
mastery of the game. It is a pinnacle
which elevates the game to an art
and which-like any other art-is a
7
source of delight which affords the
practitioner an opportunity to take
part in the act of creation.
Compilation
of technical terms
used in the book
Before delving into the actual sub­
ject matter, the author thought it
appropriate to define the terms and
expressions used in this book. He
feels certain that this will facilitate
the study of the work and help the
reader master the theory of contacts
and their combinations.
Individual contacts- various forms
of contact that develop between the
pieces of one or both sides as well as
between pieces on the one hand and
the squares they control on the
other.
Offinsive contact- contact between
the attacking piece and the piece it
attacks.
Restrictive contact- contact that
develops between any given piece
and pieces of both players that re­
strict its freedom of movement by
being in its line of fire. Or the con­
tact that occurs between the piece
and the squares of its normal range
of movement when these squares
are attacked by hostile pawns or
pieces.
Defensive contact- contact that is
normally established between a
player's pieces and pawns for mu­
tual protection.
Shielding contact- contact that oc­
curs between pieces and pawns of
8
one colour. It occurs when one
piece is attacked and another chess­
man (piece or pawn) shields it by
stepping into the line of fire.
Queening threat- this contact de­
velops between a queening pawn
and its queening square.
Use ofthe queening threat- the con­
tact that occurs between the "king
and the queening square of the op­
ponent's passed pawn approaching
the queening square.
Co-ordinated combination ofcon­
tacts- deliberate combination of
contacts between pieces and pawns
of one side aimed at one or two tar­
gets. There are two distinct types of
co-ordinated combinations of con­
tacts: the combined attack and the
double attack.
Combined attack- co-ordinated at­
tack by pieces and pawns against a
hostile chessman, one piece or
pawn attacking the chessman, while
the other attacking pieces or pawns
restrict its freedom of movement.
This is a combination of an offen­
sive contact and a restrictive con­
nection. A combined attack may
also include shielding connections.
Two-fold attack- co-ordinated at­
tack by two pieces threatening one
and the same object (chessman or
square); or the co-ordinated attack
by two pieces against different ob­
jects; or the attack by one piece or
pawn against two objects in the hos­
tile camp. In most cases this involves
the combination of two active con­
nections.
Double attack- this term is broader
than the term "two-fold attack", be­
cause in addition to being a two-
fold attack it covers the combina­ The finale of a winning combina­
tion of attacks and threats and even tion can be a combined attack, a
the combination of two threats. The two-fold attack or a double attack.
object of a double attack can be not The finale of a drawing combina­
only a piece or a pawn, but also tion can be the creation of a posi­
an important square in the oppo­ tion in which there is no co-ordi­
nent's camp. As a rule a double at­ nated attack, but only restricting
tack occurs in the form of a combi­ contacts (stalemate, fortification,
nation of any two of the aforemen­ blocking), or in which there is an at­
tioned contacts, but it can also be a tack or a threat, but no complete
combination of strong threats, such restricting contacts (perpetual
as mating, stalemating and perpet­ check, perpetual pursuit). It goes
ual check threats. without saying that the finale of a
Combination- this term, in its ori­ drawing combination can also be a
ginal sense, refers to a co-ordinated double attack.
combination of contacts, which ne­ Stages ofattack- the posting of a
cessarily gives a player an advan­ piece in relation to the object of at­
tage, or to the conversion of a cer­ tack. We distinguish, in the order of
tain combination of White and increasing influence, between three
Black pieces into a co-ordinated main stages: the concealed threat
combination, which necessarily (also ref�rred to here as preliminary
gives one side the advantage. A dis­ threat or threat of the second or­
tinction is made between winning der), the threat and the attack.
and drawing combinations.

9
Part 1 game, involving say a lone king
against a king and a queen.
The double attack

The interrelations
between strategy
and tactics

The chessboard is a battlefield of

two opposing armies. The two com­
manding generals have to plan their White's objective is to mate his op­
operations and execute them in the ponent's king. This can be done
course of the game, i.e. they are only by forcing the king to move to
faced with strategic and tactical one of the peripheral ranks or files.
problems which they try to solve. In the position shown in Diagram 1,
The reader will recall that strategy White must first force Black's king
answers the question of what is to to move to a peripheral rank or file
be done, and tactics that of how it is and then mate it with the combined
to be done. These are the two basic forces of the king and the queen.
questions with which one is con­ The sum of the measures taken to
stantly confronted with in the achieve this end constitutes White"s
course of a chess game. strategic plan. But the ways and
Generally tactics is considered sub­ means of forcing the king to move
ordinate to· strategy. But on the to the edge of the board lie in the
chessboard the situation is some­ domain of tactics. To force the king
what different. Although in chess to the edge of the chessboard one
too, tactics ranks second to has to reduce the number of squares
strategy, it plays a very important it is free to move on. This can be
part in the game. The forces on the done by the queen alone or by the
chessboard are limited in quantity, queen acting in concert with the
there being no reserves to draw king. Let us first try the former ap­
upon. A simple mistake due to care­ proach, because it is easier to un­
lessness can result in mate or irre­ derstand.
placeable material losses, which 1 ®fs
means that any tactical operation
With a single stroke the queen cuts
can prove decisive. To get a clearer
off the king from the fifth file.
idea of the interrelation between
strategy and tactics let us consider 1 . . \tlc6 2 ®es dlb6 3 ®ds \tlc7
.

an elementary example of an end- 4 ®e6 \tlb7 5 ®d6 dieS 6 ®e7 .. .

11

The first phase of our strategic plan
is completed, the king having been
forced to move to the peripheral
rank. This was done by a special
manoeuvre with the queen, which
kept closing in on the king at a
knight's move distance. This is a
tactical method, but the whole pro-
. cess of driving the king to the edge
of the board was based on a strate­
gic concept. To continue driving
the king into the corner in this man­
ner-6 . . . d/b8 7 �d7 d/a8 8
�c7 ? ?-would of course be a tacti­
cal error leading to stalemate. Hav­
ing confined the king to the last
rank, White should now pull up his
king. The second phase, in which
the king assumes an active role, is a
purely strategic manoeuvre involv­
ing no tactical considerations at all,
for White's king can approach his
opposite nu mber along any route,
e. g. 6 . . . d/b8 7 dlb2 d/c8 8 dlb3
d/b8 9 d/b4 00c8 10 OObS d/b8 1 1
00b6 d/c8 . And now White is set for
the kill: 1 2 ®c7 or 1 2 ®e8 mate.
Let us recapitul ate in brief. The
strategic operation whose objective
was to mate Black's king with the
queen consisted of three phases :
1 Manoeuvring the queen in such a
12
way a s t o force the lone king to the
edge of the chessboard
2 Manoeuvring the king to where
the opponent's king can be check­
mated
3 Delivering the death blow with
the queen, assisted by the king.
As you see, strategy determined the
overall operation. Tactics played
only a minor role, namely in decid­
ing how to force the king to the
edge of the board, how to avoid
stalemate and how to give the actual
checkmate.
To be sure, our example illustrates
an elementary situation. With no
danger to fear from his opponent,
White's tactical tasks were quite
simple. But even here White could
have made a bad mistake that could
have cost him a win which was his
for the taking. In the more complex,
confused s ituations that usually oc­
cur in an actual game one can ex­
pect surprises with each new move,
which can give the game an entirely
new twist. No matter how well-bal­
anced our strategic plan might be,
one tactical mistake can bring it to
naught. There is an aphorism in
chess that is well worth remember­
ing : forty good moves are often not
enough to win a game, but a single
bad one is often enough to lose it.
You will have probably experienced
this paradox often enough in your
own chess career. The grandmaster
Teichmann once aptly remarked
that chess was 90 per cent tactics .
Each one of us-from beginner to
world champion-knows how true
this is. Let us consider a well-known
situation that occurred in a game
played by grandmasters to illustrate
this point.
Kotov-Smyslov
Budapest 1950
White to play
White has achieved a winning
strategic position, whilst in Black's
position there are organic weak­
nesses. White's knight on fS is
ideally posted, and his rooks are
aimed at h7. A properly executed
attack against the Black king should
end in a win for White. All White
has to do is involve the queen in the
offensive, but Black restricts its
freedom of movement by threaten­
ing to take the pawn on b4 with his
rook. White's plan therefore should
be to give up the pawn at an oppor­
tune moment and throw all his for­
ces at h7. 1t goes without saying
that such an operation should be
properly planned. The best thing to
do would be to advance the pawn to
g3 in order to prevent a possible at­
tack on the king. Instead, White
launches a headlong offensive with­
out due consideration of the conse­
quences. This is what happened :
1 '®e2 � X b4 2 � X h7 + ? � X h7
3 ®hs ® g8 4 �e7 � x g2 +
This turn of events proves that
White should have advanced his
pawn to g3 as a precautionary
measure.
5 dlf3 . . .
At first glance Bla.c k seems to be in
bad trouble. But at this point the
laws of tactics come into effect. The
next move brings about a radical
change.
5 . . . � f2 + !
This counter-sacrifice enables
White to mobilize his dispersed
heavy pieces in an attack on the
king.
6 <11 x f2 � b2 + 7 ®e2
Any other move would be fatal, but
after 7 . � x e2 + 8 <11 x e2
. .
® g2 + 9 OOd l (9 diet � aS + 1 0
OOd l ®d2 mate) 9 . . . ® f l +
White loses a rook and is forced to
resign.
Another fantastic example on the
same theme.
13
Hubner-Petrosyan
Interzonal Tournament, Bie/ 1976
Black to play
Black's position is hopeless. He is
two pawns down and there seems to
be no way to protect himself against
.El e8 and ® x d 3 . Only four moves
remain till the time check. Both
players were in time trouble, al­
though not desperately so. Black
played
1 . . . ®d6
At this point White could have
mated his opponent in four moves:
2 ® e 8 + d>g7 3 .E1 e7 + d>h6 4
®f8 + d>hs 5 .E1 x h7 mate, but he
preferred to cover the pawn and
played 2 g3 ?, Black replied 2 . . .
� x f4? although he could have
,
got at least equal chances by 2 . . .
d>g7. The knight sacrifice he offers
by him is weak, because only White
can win after 3 gf. Instead, White
belatedly mounted the very attack
that would have given him a win
one move earlier. But after 3 ® e 8 +
d>g7 4 .E1 e7 + ?? (even at this point
it would still not have been too late
to take the knight) 4 . . d>h6
.
14
Black was able to make deadly
threats, e. g. 5 gf ® x f4 + 6 � g3
®f2 + 7 � g2 ® g l mate. In a des­
perate attempt to turn the tide of
battle White played 5 �f2 � x f2 6
.E1 x h7 + hoping for 6 . dl x h7 ?
. .
7 ®g8 + and mate in two. But
Black replied with 6 . . . d>gS! and
White resigned. This end-game
provides rich material for the psy­
chologist.
Three s tages
of attack
How to checkmate the opponent's
king? To answer this basic question,
the player has to solve a number of
strategic and tactical problems.The
situation is somewhat different in a
contest in which the end of the
game is in sight, where the oppo­
nent's king has been pushed to­
wards the edge of the chessboard and
deprived of its defending pieces. In
this case giving mate is a purely tac­
tical problem. Let us see how it is
solved by considering once again an
elementary example of an end­
game in which the lone king is con­
fronted with a king and a queen.
White's king has taken up a threat­
ening position opposite his adver­
sary, so that all the queen has to do
is to deliver the final blow. If we ex­
amine all positions which the queen
can take up vis-a-vis the Black king
we realize that they can be classified
into three groups. The squares
marked "0" in the Diagram are
those from which the king is check­
mated . Squares marked "1" are
those from which the queen can
move to give mate. From squares
marked "2" the queen cannot move
direct to a mating square. Thus it
follows that the queen assumes a
checkmating position in three
moves, executing a manoeuvre in
three stages, i. e. the first stage pos­
ing no direct threat, the second
threatening mate and the third be­
ing the actual mating position.
The reader will have gathered that
from squares marked "2" check­
mate is imminent, but that the ac­
tual mating position can be reached
in two moves. It would be more pre­
cise to say that the first stage is also
a threat, i. e. a threat of the second
order, and that the second stage is a
threat of the first order. But this has
no bearing on the course of events:
with every move the queen exerts
more pressu re on the king.
One should bear in mind of course
that mate in two can be given not
only by way of a threat. Let us see
what happens for instance when the
queen is on b 1.
White to play
White has several ways of mating
the king in two. If he plays the
queen to bS, e4, e 1 or g6, the threat­
ened mate on e8 cannot be parried.
By playing his queen to h7, White
mates Black's king on 1 . . d>c8 by
.
2 ® cl. If he plays his queen to fS,
he mates the king by 2 ® f8 . There
are two more solutions in addition
to the six mentioned : 1 ® c l and 1
® c2 . These moves do not look like
threats, but Black is in Zugzwang
and is forced to vacate square c8 for
the queen by moving to aB. We
know from proble m chess that there
are only two possible lines that can
be taken in solving mate-in-two
problems, i. e. by making a threat or
by creating a Zugzwang.
The d,ifferent ways in which the
moves of the king and queen can be
co-ordinated to bring about a mat­
ing position are worth examining :
15
 8 posted o n the last rank deprives the
opponent's king of only one square
(namely d7) in the seventh rank,
while the king controls the two re­
maining squares in this rank.
The most active part played by the
king is in the third case : it cuts off
the opponent's king from three
squares, while the queen covers the
three remaining squares. A mate in
which the squares round the oppo­
nent's king are covered only once is
9
sometimes referred to by chess
problem composers as a pure check­
mate. But the queen's capacity is
not fully made use of in the given
case : a rook would suffice for the
job. Such important tactical con­
cepts as stages of attack are used
not only in mating positions. In
chess the means are just as import­
ant as the end; the means often in­
volving the elimination of the oppo­
nent's pieces or achieving material
10
superiority. Let us examine a con­
test between the king and a corn­
ered knight as a case in point.
11

In the first case, the queen has to

shoulder most of the burden. It
covers all squares round the oppo­
nent's king and then delivers the In this case too the individual stages
death blow. Here, White's king of attack can be easily distin­
only backs up the queen without di­ guished. The king attacks the
rectly taking part in the attack. knight from square b7. The king
In the second case, the queen threatens to attack the knight from

16
squares a6, c6 and c 8 . And finally,
from squares aS , bS, cS, dS , d6, d7
and d8, the king does not yet
threaten the knight, but he is poised
to assume a threatening posture. In
other word s, we see the same three
stages of attack here, but in this case
we have one piece being attacked by
another.
But our analysis of the situation
with the cornered knight is incom­
plete. The reader will have noticed
that all White has to do is post his
king on c6 , thereby cornering the
knight (the king coveres squares c7
and b6) . As soon as we realize the
special features of this position we
can extend the operating radius of
the White king. Thus the king
threatens to capture the knight
from the squ ares marked "1 ", from
the squares marked "2" the king
cannot yet threaten the knight, but
it is poised to take up a threatening
stance from here.
The different phases of attack can
also be distinguished in positions in
which no attempt is made to cap­
ture a piece, but only to put it out of
action. Let us examine the struggle
between a bishop and a knight · up with the pawn on aS; from the
posted on the edge of the board.
12
Yo u will notice that from its posi­
tion on dS the bishop has put the
knight completely out of action.
From squares marked " 1 " the
bishop does not yet threaten to put
the knight out of commission, but it
can threaten to do so on the next
move. Here again we see the three
phases of attack.
These phases can also be observed
in the king's struggle with a pawn in
the end-game. The pawn is trying
to reach the queening square and
the opponent's king is in hot pur­
su it. We can also see Some charac­
teristic tactical features here.
13
From the squares in the quadrangle
a 1 -aS-eS-e 1 , the king can catch
squares of the bigger quadrangle
a 1 -a6-f6-f1 the king can threaten
to stop the pawn . And finally from
the additional squares of the qua­
drangle a 1 -a7-g7 -g 1 the king
does not threaten to intercept the
pawn; but if White has the move it
can assume a threatening position.
The concept we have introduced of
phases of attack is a very important
one. These phases determine the na­
ture of the struggle in any position,
17
that is why they should be seen as
elementary tactical possibilities.
The more attacks and threats a po­
sition entails, the richer it is in tacti­
cal possibilities. Any position, no
matter how complex it might seem,
can be (as we shall demonstrate la­
ter on) broken down into these ele­
m entary components, i. e. attacks
and threats of the first and second
order. The terms "attack" and
"threat" are not new. They are used
by theoreticians in the study of
problems of the middle game and
by problem pundits in working out
a theory of chess composition. But
as soon as these terms were intro­
d uced, most theoreticians immedi­
ately turned their attention to the
study of difficult middle game prob­
lems in general and combinative sit­
uations in particular without using
them themselves. And yet these sim­
ple elementary concepts constitute
the unifying framework of most
tactical positions, making them
amenable to synthetic analysis. The
term "initial phase of attack" which
we have coined-and which can
also be called preliminary threat or
threat of the second order-is en­
tirely new and quite important. It is,
as . it were, a signpost showing the
direction in which the attack is to
develop. Attacks and threats can
normally be seen with the 'naked
eye', they are easy to recognize. But
a preliminary threat often escapes
notice at a cursory glance; a certain
amount of skill is needed to recog­
nize them.
The ability to recognize hidden tac­
tical opportunities is determined to
18
a considerable extent by the player's
ingenuity, his combinative perspi­
cacity and his tactical skills. The de­
velopment of these skills is an im­
portant phase on the road to achiev­
ing excellence .
Methods of defence
We have become acquainted with
the elementary means of attack.
Now we must find out how an at­
tack can be warded off. When a
piece is attacked there are four pos­
sible ways of responding. Let us ex­
amine them in their proper order.
- ch ange of posting
support from other
p1eces
Attack
blocking
counterattack
1. Change a/posting
This manoeuvre can be either pas­
sive or active. If the piece in ques­
tion is moved back under the pro­
tective umbrella of its pieces and
pawns, the defence is said to be pas­
sive. But if it is moved forward or to
the side to harass the attacker or
some other hostile piece, then it is
said to be active.
2. Support
from other pieces
In this case the piece attacked does
not leave its posting. Instead, an­
other piece is brought forward to
defend it. But in this case an ex-
·
change is possible. It goes without
saying that this technique can be
used only if the attacking piece is of
equal or greater value than the
piece attacked.
J.Blocking the
line offire
ofthe attacking piece
An attacking piece that can move
across the whole board can be
blocked by moving a piece or pawn
into its line of fire, thereby shielding
the attacked piece. As a rule the
blocking piece should at most be of
equal or lesser value than the at­
tacking piece. It is also important to
bear in mind that it should be
covered either by the piece attacked
or by some other piece.
If the blocking piece is of the same
type as the attacking piece (i. e.
bishop and bishop, bishop and
queen, rook and rook, rook and
queen) , then it threatens the attack­
ing piece when it is played to the
blocking position. Such a situation
can be regarded as a reciprocal at­
tack, in which case the initial at­
tacker h as to think of ways of coun­
tering the counterthreat. That is
why this method of defence is re­
garded as the most effective.
4. The counterattack
If the situation permits, one can dis­
regard the attack altogether and
mount one's own offensive against
a hostile piece. In such a case the
piece attacked and left unprotected
is said to hang. In this connection it
should be mentioned that hung
pieces afford much scope for tacti­
cal manoeuvres . An opponent's
piece at which the attack is aimed
should be at least equal or higher in
va-lue so as not to lose the exchange.
There is of course nothing new
about what we said about the means
of defence. One learns all these ele­
mentary methods of defence at the
very beginning of one's chess ca­
reer. It was only intended to show
how plentiful are the resources that
chess offers in the way of defence.
You will probably have noticed that
one does not always have a free
choice of all these m ethods in any
given situation. There are only two
ways of warding off an attack on
the king: one can change its posting
or block it. An attacking knight can­
not be blocked, and a piece at­
tacked by a pawn must try to evade
the attack. Let us examine a case in
which one piece is attacked by an­
other. If the former evades a con­
flict by retreating, a clash is
avoided. But if one of the remaining
methods of meeting an attack is
used, then a single battle, an ele­
m ental clash, ensues. The first thing
an attack produces is a disturbed
equilibrium. It is then restored, but
not fully, through defence. For in­
stance, in supporting a piece or
19
blocking an attack, another piece is 15

drawn into the conflict. This com­

plicates the situation. The mobility
and fighting capacity of this third
piece, which is as it were chained to
the piece it defends, are reduced .
The objective of a tactical operation
can be to burden one' s opponent
with such an immobilized piece.
Here is a classical example of an
end-game to illustrate this point. Black to play

14 Black chains White's rook to the

pawn at a6 by 1 . . . bl f6 ! If the
pawn advances then 2 . . bl a6. 2
.

\1/bs is followed by 2 . . bl fs + 3
.

dlc6 bl f6 + 4 00c7 bl f7 + 5 dld6

bl f6 + 6 00e7 bl b6 etc.
16

White's rook is tied to defending

the pawn at a7 and deprived of its
freedom of movement. In this man­
ner Black succeeds in fully neutral­
izing White's superiority and in sav­
ing the end-game. The method of
defence is very simple indeed: White to play
White's rook must be kept tied to
defending the pawn. To do this, Here White ties the rook to the de­
White's king must be driven off fence of the pawn at g2 by 1 � f3 !
whenever it tries to get near the after which Black is unable to free
pawn. himself, e. g. 1 . . . \tlc2 (the threat
An analogous, but somewhat more of 2 . bl b 1 + is parried by White
. .

difficult defensive method, which is by moving his king behind Black's

also based on the tying down of a
piece, is shown in the next diagram.
king before the rook can give
check) 2 00c4 d/d2 3 \tld4 di e t 4
00e3 00f1 . Black is determined to
fight for his freedom, but White re­
stores the original situation by 5
� e2 + 00e 1 6 � f3. Hence a draw.

20
A unique s ituation on the theme of advantages. This also applies to
tying down pieces is shown in Dia­ blocking. The fighting capacity of
gram 1 7. an interposed piece may be im­
paired even more than that of a tied
M. Liburkin, 1946 one.
Conclusion ofa study A piece used for shielding its king
against an attack can be chained to
17 it so firmly that it loses all its mobil­
ity and firepower. Even experienced
masters often tend to forget this.
Let us consider the following very
instructive example.

Makogonov-Chekhover
Tbilisi 193 7

18

Black to play

What should Black do ? If 1 . . . e S

then 2 � f7, a n d Black is unable to
cover his pawn and avoid move
repetition. His king is securely
pinned in the corner, for a move by
it would deprive the rook of the
square eS in view of the knight fork
at c6. This means that Black h as no Black to play
choice but to move the rook.
White was not worried by 1
1 .. . lle2 + 2 dld3 lle 1 3 dld2 1:1 fS because, seeing as the hostile
1:1 e4 4 dld3 1:1 eS 5 dlc3! . . . rook is tied to defe nding the king
This fine move again forces Black and is unable to attack the piece on
to decide what to do next. By 5 . . .
f6, he thought he could counter the
1:1 cS + he makes the last attempt to move with 2 lld8 . But White forgot
improve his position, but after 6 that his queen too was rendered
dlb4 he is forced to turn back : completely powerless. After 2 . ..
6 . . . 1:1 eS 7 00c3! ® h4 + !he had to acknowledge de­
feat because he lost his queen.
Let me stress once again that a tied
piece i s a serious positional handi­
-cap and that a tactical operation
aimed at tying an opponent's piece
can be a means of gaining certain

21
 19 Euwe-Alekhine
Amsterdam 1935

21

To force one's opponent to burden

himself with a pinned piece can be
the objective of a tactical operation.
White to play

A. Troitzky, 1924 Black's bishop is pinned, but Black

threatens to extricate himself by
20
1 . . �b5 . White frustrates this at­
.

tempt by 1 r! a2 !, a brilliant move.

Now 1 . �b5 can be answered
. .

simply by 2 cb to cover the rook.

Since Black was unable to free him­
self, White won easily by advancing
his pawns on the K-side.
When a piece protects any piece
other than the king, the protected
piece being stronger than its defen­
White to play and win der, then the latter's fighting capac­
ity is diminished because its retreat
After 1 �d2 + Black has to defend will result in loss in material. But
himself by 1 . . bl g5, thereby pin­ one should remember that in excep­
.

ning his rook. White's task now is tional situations the knot can be
to capture Black's pawn without al­ severed by sheer force.
lowing his opponent to u npin his
rook. This is how he goes about do­
ing it: 2 �f7 a4 3 ctlh2 a3 4 ctlh3 a2
5 � x a2 et/ x h5. For a moment
Black thinks he is saf�. But then
comes 6 �f7 + , and Black suc­
cumbs to a deadly pin, e. g. 6 . ..
bl g6 7 ctlh2 or 6 . . . ctlh6 7 �e8 !

22
Tarrasch- Tshigorin
Petersburg 1893
White to play
The knight at f3 is pinned , but after
1 � x f6 + gf, White disregards the
pin and plays 2 � x eS ! This means
that the pin was only apparent, be­
cause after 2 . . . � x d 1 , Black is
checkmated in two moves by 3
� x f7 + and 4 �h6, which shows
that the reply 2 . . . de 3 ® X g4 was
forced and that White won a pawn.
Here is a very recent example. This
version of the English opening was
played by Razuvayev-Kupreitchik
at Dubna in 1 970 : 1 c4 e5 2 �c3
�c6 3 �f3 fs 4 d4 e4 5 �g5 �f6.
Instead of moving away the knight,
White lets himself be drawn into an
exchange of blows by 6 dS ef 7 de fg
S ed + .
White did this in order not to lose
any pawns. He had anticip ated
8 . . . ® x d7 and 8 . . � x d7.
.
But the unexpected happened.
8 . � x d7 ! ! and having lost a
. .
piece, White was forced to resign.
The scope of defensive moves is
broader if initially there is no at­
tack, but only threats. In this case
the same four defensive methods
are used, but here they are em­
ployed as preventive measures.
1. The evacuation
One can for instance leave the
danger zone before the attack �akes
place. A classical example of th1s de­
_
fensive method is the evacuation of
the king during the opening game
by castling. Or one can ta �e the ac­
tive line of play by advancmg a
piece (this does not apply to the
king) early in the game in order to
launch an attack before the enemy
does or to assume a threatening
posture.
2. Support
Let us assume that one of our pieces
is in danger of being attacked. If
there are no pieces or pawns in its
vicinity which could support it, they
could be moved up in advance to
give support when it is needed. But
if there are pieces within reach
which can give support, they can be
drawn up before the attack materi­
alizes.
23
3. The blockade
If a player has to block an attack, he
can move up pieces or pawns, but if
they are already available then they
can be posted to block the files,
ranks and diagonals in question in
advance. If we have warded off a
threat before it materialized, our
opponent can still carry out his at­
tack, but it would be confined to the
blocking piece, which in this case
too would result in a pin.
4. Anticipating
the opponent's
active operations
One can initiate offensive opera­
tions or occupy the most forward
rank with pieces, depriving one's
opponent of a chance to do so first
and threatening the opponent's of­
fensive pieces even before they can
engage in offensive operations.
Questions of anticipating threats
and preventing activities on the part
of the opponent belong more in the
realm of strategy than tactics. Some
of them are dealt with in the open­
ing - in mobilizing forces and oc­
cupying the centre - others in the
middle game when drawing up a
plan of the game.
We shall deal mainly with attacks
and threats (of the first and second
order) that are of a marked tactical
nature and which are directly con­
nected with the opponent's aggres­
sive operations.
24
The second assault wave
The first assault has been parried.
The defence was sufficient to ward
off the offensive. The piece at­
tacked has been given support or
shielded by another piece. But the
struggle is not over. If the opponent
has reserves, he will throw them
into a fresh offensive thrust or
mount an offensive. The combat
zone expands.
Let us first consider a situation in
which the piece under attack is sup­
ported by another piece. In this case
two kinds of offensive tactics can be
used:
1. Second attack on a
defonded piece
We shall call this a two-fold attack
because a second piece is involved.
If for some reason the piece under
attack cannot be withdrawn or de­
fended by another piece, it is lost.
2. Attack on the
defonding piece
This kind of attack is even more
dangerous, for the defending piece
cannot move away leaving its
charge u nprotected. If in such an at­
tack the defending piece itself can-
. not be protected by another piece or
by a pawn, the entire defence bul­
wark collapses like a house of cards.
Basically, this too is a two-fold at­
tack, the only difference being that
two pieces are attacked: the piece
being defended and the defending
p1ece.
If the piece being attacked is
blocked by another piece then there This is a true two-fold attack. All
are two possible kinds of attack :
1. A second attack
on the blocked piece
from another direction
If the attacking pieces act along one
line, then the attack can prove to be
a two-fold attack, because both the
blocked and the blocking piece are
threatened.
2. A second attack
on the blocking piece
these devices used in the second as­
sault wave are represented in the di­
agram below.
Particularly interesting situations
arise when the piece being attacked
is left unprotected and a counterat­
tack is mou nted instead. Different
responses are conceivable, e. g. one
can attack the other hostile piece or
one can withdraw one's own piece,
but the best course to take is to
move the piece to a safe place, at the
same time attacking one of the op­
ponent's pieces. This way two of his
pieces are in danger, only one of
which he can save in a single move
however. As a result of a two-fold
attack he loses material. Here is an
example illustrating this point.

-support
--{ second attack on the defended piece

-l
attack on the defending piece
Attack

blocking
--{ second attack on the blocked piece

second attack on the blocking piece

25
Maciewski-Averbakh oughly familiar with the mechanics
Polanica Zdr6j 1976 of its action and with its causes in
order to make effective use of this
24
powerful weapon.

An excursion
into the past

The two-fold attack-or more pre­

Black to play
In this position Black attacks
White's knight at dS by 1 . . . c6 .
White disregards the attack and
threatens square f7 by 2 il f3 ? But
Black answers 2 . . � eS. In this
.
cisely a special case of the fork-as
an important and dangerous offen­
sive method was recognized by our
forefathers at the beginning of the
era of modern chess.
To prove this let us take a glance at
one of the first textbooks on chess
entitled "Libro da imparare giocare
a scacchi" by the Portuguese author
Damiano of Altemira, which ap­

m anner Black covers the pawn at f7

peared in Rome in 1 512 . The book
and attacks the rook at f3 , the
�njoye d great popularity, 'Yas pub­
knight dS hanging anyway. Here _ several editions and made
hshed m
we have a typical case of two-fold
a substantial contribution to the
attack. This situation cannot be
spread of chess in Western Europe.
remedied by 3 � f6 + d/g7 , be­
Damiano devotes a whole chapter
cause again two of White's pieces
are threatened. The result is that �o the fine points of the game, giv­
mg many examples of the two-fold
White has to reconcile himself to a
attack. We shall find it instructive to
loss in material.
study them.
Although the two-fold attack is a
more complex offensive method 25
�han the simple attack, its effects are
more lasting. It can develop from an
attac � of two pieces on one, or of
two pteces on two pieces. It goes
without saying that a single piece
can also execute a two-fold attack.
The two-fold attack is considered
one of the most effective tactical
methods in chess. For this reason
we shall examine all its basic fea­
tures. It is very important to be thor- Black t o play

26
This is the first example given by
Damiano. The author cautions the
reader not to take the pawn at h3
because after 2 El x e6 + and 3
� f4 + , Black loses a piece.
26
Black to play
In this position both players pro­
mote their pawns at the same time:
1 . . f4 2 aS f3 3 a6 f2 4 a7 f1 ( ® ) S
.
a 8 ( ® ) , but Black captures White's
short- lived queen by a two-fold at­
tack : S . . . ® b l + 7 dlas ® a 1 + .
27
White to play
D amiano notes that to White's 1 g7
Black can reply 1 . . . El x a2, be­
cause after 2 g8 (®) Black has a
two-fold attack 2 . . . Elg2 + and
3 . . . El X g 8 .
Although this i s true, on analysing
the position we notice that White
s hould still play 1 g7, but in re­
spo nse to 1 ... El x a2 he should
continue 2 Elb6 + dieS 3 Ele6 + !
instead of 2 g8 (®) ?
The immed iate attempt 3 Elg6 is in­
adequ ate because 3 . . . Elg2 + 4
d/hs El h2 + s d/g4 Elg2 +
(draw) , because the king cannot
abandon the rook. But now Black
cannot capture the rook, because 4
g 8 ( ® ) + and S ® x a2; but if the
king steps aside, the decisive move
is 4 Elg6 .
28
White to play
D amiano gives this example to de­
monstrate another knight fork : 1
® x eS ® x eS 2 �f7 + and 3
g xeS.
The next position illustrates the
two-fold attack by a rook.
27
 29
White to play
1 � x e6 + and if 1 . dl x e6
. .
then 2 � a6 + and 3 � X h6, win­
nmg a ptece.
30
White to play
This example illustrates the capture
of a rook by a two-fold attack: 1
� d8 + . Black's rook is tied to the
queen and may not leave it, but
1 . dla7 is followed by 2 � x e8 .
. .
The theme of pinning in conj unc­
tion with a two-fold attack is illus­
trated by the following elementary
example.
28
31
White to play
The pawn at b7 is reliably pinned by
the bishop, making it possible to
give mate in two : 1 ® x a6 + and 2
® x b7 mate .
A more complicated example on the
theme of pinning is shown in the
next diagram.
32
White to play
The only answer Black has to 1
�fs + is 1 . �d7, but by playing
. .
2 � x d7 ! rl x d7 White exchanges
his rook for the bishop in order to
burden his opponent with a pinned
rook, which is then captured in the
following manner : 3 � cS rl g7 4
�dl.
 33 Ph. Stamma, 1737

35

Black to play

Here we see the lifting of a pin and White to play

the capture of a rook : 1 . . .
t!tJ X a 1 + ! 2 m X a 1 ab. In conclu­
After 1 Eld8 + mb7 2 Elb8 + !
sion , Damiano gives an example
m X b8 3 � c6 + White captures
showing how a fork can come in
the queen with the aid of a fork.
handy in queening.
The fork-a special case of the two­
34
fold attack-was studied carefully
by our forefathers.

The fork

At a very early stage in our chess ca­

reer we become acquainted with the
very simple concept of two-fold at­
White to play tack. The fork-a simultaneous at­
tack by one piece or pawn on two
White distracts the knight c6 and units-is nothing other than a spe­
gets a queen by means of a two-fold cial case of the two-fold attack. The
attack 1 � d4 + . fork is particularly dangerous, be­
Here is another example from the cause two pieces are attacked simul­
distant past. taneously and the opponent can
save only one of them in one move.
It is generally assumed that forks
are created only by pawns or
knights. Here for instance is an in­
teresting position in which a fork
m akes a pawn stronger than a rook.

29
G. Kasparyan, 1940
Conclusion ofa study
36
White to play and win
How can one win in this position? If
1 � h8 + , Black answers 1 . . .
d/ x h 8 2 d 8 (®) + d/h7 . But if
White does not sacrifice his rook
and plays 1 e6 instead, then Black
answers 1 . . . � g7 2 r1 c8 r1 a 1 ,
threatening mate by 3 . . . r1 h 1.
The solution is brilliantly sinwle : 1
� e8!! � x d7 2 e6 and Black re­
signed . The move 1 r1 e8 is vt>ry im­
portant. To any other move (for in­
stance 1 . . � c8) Black \VOuld, af­
. tacks" with a bishop or rook against
ter 2 e6, offer a sacrifice 2 ... !l d 8
t o save himself.
The next position demonstrates the
strength of a knight fork.
30
H. Rinck, 193 5
37
White to play and win
Even the long diagonal proved too
short for the bishop to escape from
the knight's stubborn pursuit. See
for yourself: 1 � f7 �a 1 2 d/b 1 ! If
the bishop tries to flee to f6 or g7,
White replies with 3 � d6 + and 4
� e8 + , but if it is played to c3 or
d4, then White replies with 3
� d6 + and 4 � bS + .
Basically, these attacks (by one
piece against two others) are in no
way different from the "fork at­
two pieces.
38
White to play
White plays 1 g8(®) + and after account of 2 11b7 � a6 3 11 a7
1 . . dl x g8 forks the rook by 2
.
2 11bs . . .
� dS+.
White threatens a two-fold attack
39 by 3 11 cS, thereby forcing the
bishop to withdraw.
2 . . . � d 8 3 11ds � e7
The threat was 4 11 d6, whilst 3
� h4 would have been followed by
the fork 4 11 hS + ;White's decisive
reply to 3 . . . � f6 would again
have been 4 .El d6.
4 11 d7 dlg6 5 B c7, and White
White to play WinS.
Even the king can carry out a fork­
This position is similar to the one like attack.
above. White continues with 1
a 8 ( ® ) + and, after 1 . . 11 X a 8 ,
.
A. Troitzky, 1896
captures the rook b y 2 � f 3 + . Corrected version 1922
Conclusion ofa study
L. Kubbel, 1923
41
40

Draw
White to play and win
Black is two pieces up, but White
Here White manages to capture succeeds in capturing one of them
one of his opponent's minor pieces by means of a "king fork".
with the aid of a rook fork :
1 c7 � b7 2 c8 ( � ) ! ! . . .
1 11b3 � c6
In this manner White gets Black to
1 . . � d7 loses immediately on ac- play his bishop to c8 . 2 c8 (®)
.

count of 2 11b7 and 1 . � c7 on

. . would have ended in a loss, because

31
Black's reply would have been a
fork 2 . . . �d6 + .
2 . . .� x c8 3 cbb6 ! . . .
White threatens to attack both of
Black's pieces. There is no way out
of this dilemma, because 3 . . . � d6
would still be followed by 4 cbc7
with a two-fold attack. It goes with­
out saying that the powerful queen,
thanks to its extraordinary ability to
carry out two-fold attacks, is parti­
cularly dangerous . .
H Rinck, 1949
White to play and win
White's position looks precarious
indeed, Black threatening mate by
1 . . . rlg l . But by 1 ®d4 ! White
not only parries the threat, but
himself poses a threat to Black's
rook and knight. Black loses imme­
diately by both 1 . . . n g6 (because
of 2 ®d3 + ) and by 1 . . . rlg3 (on
account of 2 ®ds + cbc3 3
®e5 + ) . The best reply is for the
knight to move towards the king
without delay :
1 . . . � b4
32
B u t Black's rook still stands quite
insecure, and White manages to
convert this to his advantage by a
series of keen-witted moves.
2 ®e3 + cba4 3 t!tla7 + cbb3 4
® f7 + cba4 5 ®d7 + dlas 6
®d8 + d/bs 7 ® b 8 + d/c4
All Black's attempts to hide his king
behind the knight prove futile.
8 ®f4 + ct>cs 9 ®fs + ! . . .
The telling final blow ! No matter
where the king turns, it and the
rook are "forked". For instance, if
Black plays 9 . cbc4 or 9 . .
. . .
ebbs , White plays 1 0 ®fl + , but if
the former plays 9 . . 00c6 or
.
9 . . 00d5, White replies 1 0
.
® f3 + .
The discovered check
One of the most thoroughly ex­
amined cases of two-fold attack is
the discovered check, in which two
pieces take part in an attack.
A. Troitzky, 1896
Conclusion ofa study
43
White to play and win
I ,et us start off with an elementary
·
example . White plays 1 Elc8 �
threatening to queen. 1 . . . .EI x a7
is followed by 2 \tlb6 + .
The mechanics of this two-fold at­
tack is simplicity itself. One piece
gives check, while the other attacks
any hostile piece or pawn. Since the
opponent has to protect his king,
his piece is lost.
One does not always notice imme­
diately that a discovered check can
be given. The reader will see from
the following position that occa­
sionally it takes time to recognize
such a situation.
lsakov-Nihtin
Correspondence game 1947
44
B lack to play
Black's position looks critical. He is EI X f2 + 3 OOh l � g2 + 4 \tlg l
one piece down and White is threat­ EI x e 2 + 5 00h2 � f3 + . In the end
ening to give mate in two by 1
� x d7 + EI x d7 2 .Eib 8 . And yet
he has a wonderful manoeuvre in
store to save the situation. It is
based on discovered check.
1 ® d 3 + � �.
0 0 0
Black sacrifices his qu een to set the
stage for a discovered check. If the
king sidesteps, Bl acks follows up
with 2 . . . ® x b 1, threatening the
capture of White's bishop at c l . But
if White decides to take Black's
queen at d 3 , Black captures White's
queen after 2 . � x c6 + and
. .
3 . � x a4 and ends up with one
. .
pawn up.
Kazhdan-Horowitz
New Yo rk 1939
45
Black to play
In this position Black would have
achieved a marvellous win had he
played 1 . . . � X f1 + 2 tfr1 X hS
the discovered check would have
enabled him to recapture the queen
and to win a piece.
33
Furman-Smyslov
Moscow 1949
46
White to play
White played 1 ®b2 !, threatening
to attack the pawn at g6 with his
knight and to checkmate his oppo­
nent in short order. If Black replies
2 . . . hg, he is checkmated by 3
Ei h 3 . But if 2
. . &J x g6 then 3
.
Ei g3 + . There is no defence against
the attack on g6. After the best re­
ply possible in these circumstances,
i. e . 1 . . . � c4, White plays 2
� X g6 + @ X g6 3 Ei X c4 + tfttg7
4 @ X g7 + 00 X g7 5 ij c7 + 00f6 6
f4 and achieves an overwhelming
positional advantage with a pawn
up.
The two-fold attack in the form of
discovered check or a discovered
attack coupled with check can also
be used as a means of defence.
34
A. Herbstman, 1948
Conclusion ofa study
47
Draw
Black has a substantial advantage i n
material b u t i s unable t o exploit it.
1 Ei f5 + 00e6 2 Ei eS + 00d6
One might think the game was ·
over, there being apparently no way
of giving a meaningful check. But
White springs a surprise : 3 r:l ds + ! !
the rook is unassail able because if
Black plays 3 . . c:1d x dS, he leaves
.
himself open to a discovered attack
with a check : 4 c4 + , and the queen
is lost. That is why the king tries to
flee.
3 . . . 00c6 4 � c5 + !! \Jdb6 5
� bS + ! ! 00a6 6 � aS + ! !
No matter on which square Black's
king tries to capture the rook, it is
constantly faced with a discovery
and a check, which costs Black his
queen. Thus White manages to end
the game with a draw.
The discovered check is part of a
well-known tactical manoeuvre,
which we refer to as the transposi­
tion of the two-fold attack from
one point to another.
 48
White to play and win
White forces mate in four : 1
�X h7 + d:Jh 8 2 � g6 + 00g8 3
® h7 + 00f8 4 ® x f7 mate.
How can one defend oneself
against a discovered check ? This is
shown in the following example.
Tshigo rin- Gunsberg
Havana 1890
Game version
49
White to p l ay
Black has set up a discovered check, petual check.
but has no time to implement it, be­ The reader will no doubt know that
cause after 1 ® e6 + 00h8 2 � x f5 ! discovered check is the mainspring
White begins to harass the black
queen. If for instance 2 . . . ®gS
then 3 f4 ! ® x fs 4 &:1 x f5 � x fs 5
m X g2, and White wins.
An interesting position in which the
discovered check proves to be
harmless again is shown in the next
diagram.
Averbakh-Bondarevski
Moscow 1948
50
White to play
Black' s position looks ominous, be­
cause the bishop at e3 is hanging
and there is a threat of 1 ...
� xg2 + 2 d:>xg2 � c3 + by
which Black captures the queen.
White has only one, but a fully ade­
quate, way of dealing with the situ­
ation: 1 � x e4 ! � x e4 2 � g5! He
attacks the bishop at e4, at the same
time blocking the dangerous g-file.
White's defence is based on the fact
that 2 .. . � x g2 3 � x e7 is not
dangerous, because after 3 . . .
� f3 + 4 OOfl � g2 + Black has
achieved nothing more than a per­
of such a complex tactical operation
as the "double mill".
35
Torre-Lasker First of all the king has to parry a
Moscow 1925 series of checks:
1 � x e2+ cbe 1 2 � b5+ � e5 3
51
Et x e5 + � e2 4 � x e2 + OOf l
The situation has changed and
Bl ack has to ward off further checks
by the bishop.
5 Et e4+ � d3 6 � x d3 + Et e2 7
� x e2 + dl e l
Now it i s the rook's turn to molest
the king again. We see the outlines
of an original, exquisite finale tak­
White to play
ing shape.

Here we have a classical example of 8 � g4+ dlf1

the double mill, whose wheels be­ Wh ite has made mincemeat of a
gan to turn after 1 � f6 ! -® x h5 2 string of Black's pieces, but what
� x g7 + 00h8 followed by 3 should he do now ? He decides to
� X f7 + 00g8 4 E!g7 + 00h8 5 try to win the game by checkmate.
� X b7 + 00g8 6 E!g7+ 00h8 7
� g5 + 00h7 8 � x h 5 . White won 9 00d2 Ij X g4 1 0 Ij X g4 h t ('t!tJ) 1 1
after ach ieving superiority in mate­ Et c 4 , a n d there i s no defence
rial. against the mate.
By now you are probably convinced
Here is a somewhat different ver­ of the effectiveness of the dis­
sion of the double mill, which might covered check. But the next exam­
be called the "meatgrinder". ple is even more delightful.

Hoch, 1973
I.
W Mees, 1973
Conclusion ofa study
52 53

White to play and win White to play and win

36
At first glance it seems rather futile
for White to try to win with his
rook against a queen. The best line
to take is to exchange his rook for
the queen and call it a draw. But in­
stead he moves back his bishop: 1
� d7 ! ! Wherever the queen moves
it succumbs to the rook by dis­
covered check. White has no choice
but to expose it to the bishop's at­
tack again by playing 1 . . . �h3!!
But White captures the queen by
the following manoeuvre : 2 i:l fS + !
00b4 3 EXf4 + , thereby reaping the
fruits of the discovered check. A re­
markable position !
By means of double check the rook
gradually forces the king to retreat
into the corner a8 , where in the end
it is checkmated. 1 EXf2 + dle3 2
i:l f3 + dle4 3 EX e3 + d/d4 4
EXe4 + d/ds 5 EXd4 + di eS 6
ridS + dlc6 7 rlcS + d/b6 8
rlc6 + dlb7 9 rlb6 + dla7 1 0
rlb7 + dla8 ( 1 0 . . . d/a6 1 1 rla7
mate) 1 1 n a7 + dlb 8 1 2 n a8
mate.
And here is a similar mating attack,
which occurred in a game played in
Paris in 1 922.
55

The double check

Double check is a vigorous variant

of the two-fold attack. This method
is particularly effective, because the
king is attacked by two pieces si­
multaneously. The following prob­
lem lucidly illustrates the moment­ White to play
ous strength of the double check.
1 ij X f6! � X d 1 2 ij g6 + 00h7 3
A. White, 1919 rlg7 + d/h8 4 rlh7 + d/g8 5 rih8
mate.
54

Mate in twelve

37
Barcza-Bronstein
Moscow-Budapest 1949
56
B lack to play
Taking advantage of th e strength of fensive position still further by tak­
the double check, Black achieved
material superiority :
1 . . . � X d3 2 &J X f5 3 � X e 1 ! !
This is the crux of the matter. White
has no time to undertake anything, If h e moves t o h 7 , White wins b y 5
because there is the threat of mate
by double check on f3.
3CQf1 � c2 + 4 �cl El xc l + 5
c.l>e2 � d4 + 6 <i>d2 � b3 + . White
resigned.
An ingenious mating system is
shown in the following example.
Friedstein-A ronin
Moscow 1949
57
White to play and win
38
Black's position seems to be not
bad . He is threatening with
one of the two possible moves:
1 . . � x g4 + and 1 . . ® x e 3 .
. .
But what happened was 1 �h5 + !
The rook is unassailable because of
2 � g6 mate .
1 . . . dlh7 2 � g6 + digS
In these circumstances the best de­
fence, because if he had played
2 . . . dlh8 or 2 . . . c.l>g7 , White.
would have replied 3 � d 4. But
White manages to improve his of­
ing advantage of the dangerous
confrontation between the king and
the rook.
3 El x f6 ! ® x e3 4 � f? + ! dlfS
·
�g8 + and 6 El x h6 mate .
5 �e6 + dle7 6 Elf7 + and mate in
two .
Here i s another example showing
how the attacker fails to convert a
double check.
Koblenz-Ragozin
Moscow 1945
58
Black to play
By discovered check 1 . . Il f3 + ?
. W Neustadt, 1929
2 � x eS .El x d3 Black captures the
queen, but after 3 � x h8 White 59

sees his chance and uses it. The cor­

rect continuation would have been
1 . ® x e4 ! ! The salient point is
• .

that White may not take the queen

with his bishop because of 2 . . .
J:l f2 + ! (the deadly double check !)
3 Cbh 1 Il h2 mate. But if White
takes the queen at e4 with his
queen, Black replies with 2 . . .
0 X e4 + 3 � X eS .El X eS , after
White to play and win
which he is a pawn up and has good
winning prospects. By playing 1 Cbd 1 White forc� s
Black's bishop to seek protection
from the king, because 1 . . . �b2
leads to an immedi ate loss due to 2
The two-fold attack
� c2 + (discovered check), whilst
on a defended piece 1 .�a3 would result i n 2 � c2 +
. .

(fork) . That is why Black replies

If the target of attack is covered, 1 . . � d2. White must now repeat
then the number of pieces attacking his attack on the bishop . He does
.

it must outnumber the defending this with a well-planned manoeuvre

pieces by one unit if the attack is to by the knight: 2 � c6 ! While getting
be successful . Thus if the target is set to deliver the final blow with the
covered by a single piece then two knight, White prevents the black
pieces are sufficient to attack it. It bishop from leaving the endangered
goes without saying that the target d2 square. Black has no choice but
must first be immobilized or the op­ to play 2 . . . Cbd 3 . This is followed
ponent must be forced into a situa­ by 3 � e5 + , and depending on
tion in which he has no time to find which way the king flees-to e3 or
a safe refuge for the piece. c3-White wins a piece by a fork on
The following two examples illus­ square c4 or by a discovered check
trate the mechanics of an attack o n on f3 .
a defended piece.

39
T. Gorgiev, 1929
Conclusion ofa study
60
White to play and win
Here we see the capture of a rook in
similar circu mstances. After 1 00 d6,
Black is forced to play 1 . . . b:l cS to
give the rook protection from the
king, which has a limited operating
radius. By playing 2 � d 3 + , White
forces his opponent to answer 2 . . .
dlb6, after which the pinning with 3
� e3 clinched matters.
The attack on the opponent's castle
by two pieces is dealt with in detail
in Part 2 in the section entitled "At­
tack on the king" . At this point we
shall examine only one case of such
an attack.
40
Miidler-Uhlmann
Bad Liebenstein 1963
White to play
White wanted to capture a pawn by
playing 1 b:l c7, thereby threatening
the bishop at b7. This enabled Black
to unleash a typical attack on
square g2 : 1 . . . E!e 1 + 2 00h2
b:l h 1 + ! The rook has to be taken,
but with what piece ? If White plays
3 � x h 1 , he exposes square g2
and Black gives mate by 3 . . .
® x g2.
But if he takes the rook with his
king, his pawn at g2 is pinned and
Black again attacks square g2 by
3 . . . ®h3 + ! 4 OOg l ® X g2 mate.
The attack on squares f2 and f7 in
the opening game warrants a special
discussion. The reader will recall
that these squares are the weakest
points of the two sides, because
they are defended only by the king.
Thus, to attack them successfully
one needs only two pieces, namely
knight and queen, bishop and
queen or bishop and knight. We
shall deal with these questions in
detail in the chapter entitled "The
double attack in the opening game".
But here I should like to acquaint
you with a not so familiar variation,
which convincingly demonstrates
the strength of the two-fold attack.
It would be advisable to be familiar
with this opening, because it occurs
to this d ay even in master-class
games .
1 e4 e5 2 � f3 d6 3 d4 � d7 4 � c4
�e7?
62
The bishop's move looks natural.
Since Black is worried by the pos­
sibility of the knight's moving to g5,
he decides to post his bishop at e7.
But while covering square g5, Black
overlooks another point, which is
equally important. White immedi­
ately takes advantage of this.
5 de ! de? 6 ® d 5 ! . . .
This position deserves a diagram.
63
Only six moves have been played,
but the attack on f7 by two pieces is
already irrefutable. After 6 . . .
� h6 7 � x h6, Black is a piece
down. Obviously 5 . . . � x e5 6
� x e5 de would have been better,
but this would have been followed
by 7 ® h 5 and a double attack on
the pawns at f7 and e 5 . Black has to
reply with 7 . . . g6 and accept the
loss of a pawn.
An attack on a defended object can
also be a means of defence. An ex­
ample is shown in the following dia­
gram .
G.Kasparyan, 1949
Conclusion ofa study
64
Draw
White's pieces are attacking the
knight, but unfortunately for
White, its capture would result in a
disadvantage for him. 1 � x a4
� x a4 2 00 X a4 00e7 etc. That is
why White first plays 1 f6 !, after
which he threatens to capture the
knight at a4. The knight seeks
safety in flight : 1 . . . � b2. There is
no use trying to escape to the other
side, because after 1 . � b6 2
. .
41
<1/cs � c8 3 � d7 � a7 4 dlb6, the Lasker-Euwe
knight is trapped . Nottingham 193 6
White's king is hard on its heels : 2
d:;>c3 � a4 � In this way the bishop 65

vacates square d 1 for the knight.

White proves equal to the task by 3
� f3 � He makes timely preparations
for a renewed two-fold attack. Af­
ter 3 . . . g d 1 + 4 d/d2 it becomes
clear that Black's knight has to turn
back on account of 4 . . . g f2 5
dle3 � h3 6 �g4 � g 1 7 dlf2 .
In this way White succeeds in bring­
i ng about a draw through repetition
Black to play
by alternately exposing Black's
knight to a two-fold attack on This model case has long since
squ ares d 1 and a4. come to be regarded as classical.
Black has a somewhat better game,
but he can hardly win. Nevertheless
he apparently thought it was worth
The two- fold attack
a try, and instead of moving back
on two targets his knight he went 1 . . . � aS ?,
thereby allowing himself to be
The discovered check, which we drawn into tactical complications .
have already discussed , is a good Let' s see what happens then : 2 b4 !
example of the two-fold attack on � x b4 3 � c2 ! Now both Black's
two targets. But this time we shall pieces are un der attack and one of
examine positions in which any two them is inevitably doomed.
pieces-with the exception of the
king-are subj ected to two-fold at­ L. Kaiyev
tack. The king can naturally also
take part in the attack. 66
It should be borne in mind that such
a two-fold attack can easily ensue in
defence through counterattack,
when the attacked piece leaves the
end angered zone and itself attacks
the hostile piece.

White to play and win

42
Black' s bishop is tied up covering
the knight at e 8 . Taking advantage
of this White succeeds in pinning
and then capturing it by a two-fold
attack.
1 d;>b4 � d 1 !
Black's only chance of offering re­
sistance is by counterattacking.
1 . . � c6 would be very weak be­
. Having taken the knight at c7,
cause of 2 g e 5 , after which he
wo uld i mmediately lose a piece .
2 g e3 . . .
Attacks the bishop and threatens to
give check at f8 . Black's reply is
forced .
2 . . . � h5 3 El h2 g f6 4 g g4 ! . . .
Finishing up with a two-fold attack !
4 . . . g X g4 5 El X h5 + d;>g7 6
EX g5 + , and White wins.
Even if Black had chosen to defend
A. Kakovin, 1941
himself by 3 . . . g g7, White would
have replied with a two-fold attack
by 4 g f5 and won the game after
4 . . . g x fs s EX x h5 + .
Hohler- Tcherniak
Heidenheim 19 59
67
White to play
Weak and uncovered pieces often
pave the way for two-fo l d attacks.
Take a look at what happens in the
position shown in the d iagram, for
instance . First of all , through a
series of exchanges , White elimi­
nated Black's cover.
1 � X a6 El X a6 2 � X f6 � X f6
which had been covering th e queen,
White mounted a two-fold attack :
3 gd5 ! . . .
Black is forced to move his queen to
d8 (or d6) and finds himself a piece
down after 4 g x f6 + 'l!t1 x f6 5
EX X d 7 .
A n d now l e t us examine s o m e ex­
amples showing that a substantial
material superiority can be equ al­
ized by two-fold attack.
68
D raw
Black is a rook up, but his knight is
hanging, although it cannot be
taken yet because of the fork 1 . . .
EXc3 + . By playing 1 d;>b4 White
launches a two-fold attack. There is
43
only one thing Black can do, d4, which it cannot do now due to
namely 1 . . . � e4. White replies the pin 4 � h2 .
with another two-fold attack : 2
3 . . . Et e4 +
� g3 � By playing 2 . . El x d5,
.

Black once again resorts to a "fork" (3 . . . � e3 + is i neffective on ac-

(3 . . . Eld4 + ) . But White makes
yet another two-fold attack : 3 cbc4
� f6 4 � e4 ! thereby achieving a
draw (4 . . . � f5 5 � d6 + ) .
A. Gurvitch, 1931
Conclusion ofa study
69
count of 4 cbfs rl eS + 5 cbf6 re­
sulting in a draw) .
4 cbf5 � h4 5 digS and the whole
thing can start all over again.
The next study shows how one can
effectively extricate oneself from
such two-fold attacks .

T Gorgiev, 1929
70

Draw

Here too, Black is a rook up, but by

playing 1 cbe7 White attacks two White to play and win
pieces simultaneously. This results
in a strange chase in which the king At first everything looks quite
is pursued by the Black rook. straightforward :
1 . . rl d 5 2 cbe6 rl eS +
. 1 � f6 + cbh7 2 � g7 + d/h6 3
� f7 d/g6 4 � f8 . . .
The rook defends itself furiously.
The forced moves have led to a
3 cbf6 ! . . .
two-fold attack, both of Black's
The king fearlessly exposes itself to pieces being threatened. What now ?
discovered check, White being Black unexpectedly replies with a
aware of the danger facing Black's two-fold attack of his own.
bishop. 3 cbf6 is actually the only
4 . . . � c6 ! 5 � x d8 cbg7 6 rl e8
move to make, because 3 cbd6 � b2
cbf7 !
would put White at a disadvantage,
since this would give the bishop This creates a situation similar to
enough time to move from squ are

44
that discussed in the previous dia­
gram. The king intends to keep pur­
suing the rook incessantly. But
White finds the vulnerab le spot in
his opponent's defence system.
7 ri h 8 d/g7 8 � f6 + ! cj) X f6 9
r1 h6 + , and White regains the in­
itiative, because this last two-fold
attack tips the scale in his favour.
V Korolkov, 194 7
Conclusion ofa study
71
Draw
Despite the fact that Black is a piece
up, he can bury all his hopes of win­
ning the game if he loses the pawn.
The line of play pursued by White,
who keeps attacking either the
pawn or a piece, is based on the fol ­
lowing idea :
1 dlb4 � c3 2 dlc4 ! � e2 3 dld3
�dl
It would have been a weak move on
the part of Black if he h ad played
3 . . . �b5 + , because the. result
would have been 4 � c4 � x c4 + 5
dl x c4 dlf5 6 dld3 and an immedi­
ate draw. But here we see Black
shepherding his forces in the other
direction and White sticking to his
old method of attack :
4 dld2 � c3 5 dld3 ! � b5
Black is prepared to repeat the man­
oeuvre, but to 6 dlc4 he replies not
with 6 . . . � a4 , but with 6 . . .
� e 2 + and wins. But White antic­
ipates this and pl ays 6 � c4 !, which
is a two-fold attack on a piece
covering a pawn. He masters the
s ituation with this manoeuvre and a
draw is not far off.
At the beginning of the chapter we
said that we would not be consider­
ing any positions in which the at­
tack is aimed at the king. I should
like to make ju_st one exception.
Chekhover-Kan
Leningrad 1933
72
White to play
Here we see a typical case of a two­
fold attack. In parrying the threat to
his queen by White's rook move to
a 1 , Black attacks the white queen
by playing to f6 . How should White
continue ? Bearing in mind that the
king must be protected when
checked, we easily d iscover the
45
right way to proceed : 1 ®ds + ! Af- queen, which seemed so safe behind
ter 1 . . ld x dS 2 � x dS + dlh8
. the rook.
3 ld x a2 White is a rook up. S uch a two-fold attack can also be
mounted in conju nction with a pin
on the diagonal.
The two- fold attack 74

In COllJ UllCtlOn
. . .

with a pin

Situations in which a piece is sub­

j ected to a two- fold attack and in
which it is pinned into the bargain
are particularly dangerous for the
player who is on the defensive. This
is lucidly illustrated in the examples
below. Black t o play

73 Instead of moving his king to the

side after 1 . . � c6 + , White de­
.

cided to play 2 � e4 and after 2 . . .

®fs was forced to resign. To his
misfortune he stands not only to
lose his bishop at e4 but also to be
checkmated at square fl . If this had
not been the case, White could have
replied 2 ® X c6, which would have
broken the pin and Black's plan
would have been frustrated.
Black to play

Gendel-Sushkevitch
White cannot play his king to the
Moscow 1956
first rank in reply to 1 . . .!d. a2 +
.

on account of the threat of his 75

queen getting pinned. He has no

choice but to shield the king with
his rook by playing 2 l;t e2 and al­
lowing his rook to be pinned. Black
continues his attack with 2 . . .
® e3 ! This creates a tragicomical
situ ation in which White being un­
able to provide additional cover for
his rook, is forced to take Black's
rook, thus inevitably losing his Black to play

46
Here we immediately see that the 1 f7 00g7 2 f8(®) + ! . . .
tactical operations ought to be fo­
White's intentions are still unclear
cused on the pinned and tied bishop
at this point. Why does he sacrifice
at g3. And this is just what Black
the pawn ?
does by playing 1 . . . El hg8 . White
can only respond with 2 b;l d3. How 2 . . . 00 X f8 3 bj fl + . . .
can Black intensify his assault? A
Black's king naturally seeks cover
sacrifice is out of the question, be­
behind the rook :
cause if 2 . . El X g3 + 3 b;l X g3
.

bl g 8 then White takes the rook at 3 . . . 00e7

g8 with impunity because his queen
What will White do now ? Why not
is covered.
try 4 El fe 1 !
You probably understand the un­
derlying idea now. It is essential 77
that the queen should be supported
for this tactical operation to be
crowned with success. This means
that it has to be driven from square
b3 by 2 . . . El ab 8 ! 3 'f!t1 c3 . And now
Black can strike at g3 : 3 . . .
bj X g3 + 4 El X g3 bj g8 5 bj X g8
&:! X c3, and Black wins.

A. Mandler and I Konig, 192 4

76
White to play and win
The situation does not look very
tense. Although both rooks are un­
der fire, they cover each other and
at first glance Black seems to be in
no immediate danger. But let us see
how things develop.
Again a situation similar to that
shown in example 7 3 . Black can of
course keep checking the king with
his rook until the king moves to
square a4. But then he has to take at
e2, and loses his second rook after
White has recaptured his rook and
given check. Black can try to flee to
the other side by 3 . . . 00g7. This is
followed by 4 bl g2 + 00h6 5
El h l + bl h3 6 bl gh2 !, and White
again achieves his objective of win­
ning a rook.
The two-fold attack coupled with a
pin is not only a very effective of­
fensive but also a defensive method.

47
A. Gurvitch
Conclusion ofa study
78
White to play
In this famous finale White attacks
Black's rook at d4 by a second piece
by 1 d/c3. Black seems to have re­
stored the balance by playing 1 . . .
dieS to give his rook additional
cover. But after a waiting move with
the bishop (2 � f2 or 2 � g 1 ) we
realize that it is not only the rook at
d4 that is permanently tied. If the
king makes a wrong move, Black
loses material, and his other rook
can only shuttle back and forth use­
lessly between dS and d8. Despite
his truly overwhelming superiority
in material (two rooks against a
bishop) Black cannot win. A unique
position !
Sometimes a piece i s already pinned
and all one has to do is organize a
two-fold attack on it.
48
Alekhine-Capablanca
Competition held in 192 7
79
White to play
Black's rook has strayed into the
enemy camp, and White cuts off its
retreat by playi ng 1 � cS . What fol­
lowed was 1 . . . � x cS 2 de (natu­
rally not 2 bc because of 2 . . . -® aS,
and thanks to the attack on the
p awn at a3 Black comes off the
hook) 2 . . . 't!tl e S . In an attempt to
save his rook Black gets entangled
in a dangerous pin. Black must not
allow him to extricate himself : 3 f4 !
't!tl g7 4 � x e4 de, and instead of
the somewhat delaying move 5
00f2 , White could have immedi­
ately decided the game in his favour
by playing 5 n f2 ! with the unpre­
ventable threat of 6 n c2 .
Trifunovic- Golombek
Amsterdam 195 4
80
White to play
By playing 1 g4, White immediately
threatens to capture a piece. Black
replies with 1 . . . g S , thereby
blocking the pawn's further ad­
vance . But White retains the initia­
tive by 2 h4 !, a strong move. If one
of White's pawns is taken, White
can play 3 gS , whilst 2 . d/g6 is
. .
followed by 3 � e4 + , and White
checkmates Black's king in three
moves after 3 . . �- x e4 4 fS +
. defence for Black, 2 . . d/g8 being
d/h7 5 bl x f7 + and 6 ®g7 mate .
The p i n is used skilfully in the fol­
lowing examples for mou nting two­
fol d attacks.
Tshigorin-Allies
Moscow 1901
81
White to play
In reply to 1 hg + Black was forced
to play 1 . . ® x g6 (if Black had
.
played 1 . d/g8 , Wh ite would
. .
have gained advantage by 2 bl f7
and an irrefutable threat of playing
the queen to the h-file) . White has a
brilliant reply to this move : 2 bl fS �
He takes advantage of the pin with
a view to giving Black the decisive
check at hS. There is no s atisfactory
.
refuted by 3 bl f8 + .
W Bron, 1963
82
White to play and win
49
White's pieces gang up on the king Simagin-Zagorianski
forcing it to seek protection behind Ivanovo 1944
the queen.
83
l� c3 + dle7 2 ® e s + dld8 3
� aS + b6
At last the queen manages to afford
protection to the king. But the lat­
ter's hapless placing behind the
pawn at b6 (again that calamitous
pin) enables White for the first time
to unleash a two-fold attack.
4 ® cS ! 'cl'Yb8
Black to play
The best possible reply. If Black had
played 4 . . . dieS instead, he would Black's position is rather difficult.
soon have found himself check­ White's second rook threatens to
m ated : S 'cl'Ye3 + dlf8 6 �b4 + join his first one on the 7th rank.
d/g7 7 ®gS + . But Black finds an interesting de­
The game then took the following fence :
course :
1 . . . ld f7 2 ld h8 + d/d7 !
S � x b6 + dle8 6 ®e3 + dlf8 7
� cS + d6 84

Black tries to put another obstacle

in the bishop's way, but a second at­
tack sweeps it aside.
8 ® e S ! ®d8 9 � X d6 + d/g8 1 0
®g3 + d/h8 1 1 � e S + f6 1 2 ®gS !
Black has no defence against this
third blow.

This has brought about a mutu al

The mutual
two-fold attack
This extremely tense situation in
which two pieces on each side at­
tack two hostile pieces deserves
special consideration.
two-fold attack. All four rooks are
confronting each other, but their
exchange would not bring about
any particularly interesting devel­
opments, e. g. 3 ld X c8 ld X g7 or 3
ld x f7 ld x h 8 . But White finds a
way of turning the advantages of his
position to account.

so
3 � c6 + ! ! . . . White is a pawn up, but his king's
side has been weakened and there is
A move fraught with possibilities. If
the annoying threat of Black's play­
the rook takes the bishop on c6 ,
ing 1 . l'bf3. The correct move
then White can take the rook on f7 .
. .

would have been 1 El c2, which

If Black takes the bishop with his
would have enabled White to play
king, White immediately replies 4
his bishop to g2 in response to the
El x c8 + and then captures the
queen being played to f3 . But White
Black rook on f7 . If Black takes the
wanted to solve the defence prob­
bishop with his knight, then White
lem once and for all and played 1
plays 4 El x f7 + . Which means that
� c4 ? in order to exchange the
the bishop may not be taken by any
bothersome bishop. What followed
piece and the king is forced to move
was
away.
. . . � x c4 2 l'b x c4 El d 1 + !
3 . . . \ll e 6
The second assault wave is ap­ 86

proaching.
4 El h6 + Ei f6 5 � d7 + !
The final phase is also based on the
two-fold attack. White achieved a
qualitative superiority and easily
converted his advantage. Mutual
two-fold attacks require precise cal­
culation and circumspection, for
they can easily boomerang if not
handled with due care.
Alatortsev-Konstan..t inopolski
Tbili(i l 93 7
85
White to play
Taking advantage of the rook's
contact to the queen, he thickens
the plot by mounting a mutual two­
fold attack in order to derive an ad­
vantage.
3 \ll g 2 l'b x c4 4 El x c4 El X a1 ,
and Black is a rook up.
You will recall that a similar situa­
tion occurred in the position shown
in diagram 77, in which pinning was
involved. In both situations a rook,
an entirely innocent bystander in no
way involved in the events taking
place, became the victim of an in­
exorable fate.

51
M utual two -fold attacks can occu r
in a variety of ci rcu mstances both in
offensive and defensive operations.
Special attentio n should be paid to
them in situations involving recipro­
cal attacks and pinning.
Burn-Marshal!
Telegraphic match
USA -England, 1911
87
W h i te to play
By playi ng 1 � x dS ! White used
the tw o - fold attack not only to free
h i m s e l f from a pin but also to gain a
pawn, e. g. 1 . . . ® x d2 2
� X f6 + gf 3 EI X d 2 .
Mar6czy-Bogolyubov
Dresden 1936
White to play
52
In view of his threat to captu re the
pawn at c2, Bl ack thought his pawn
at c6 was unassailable. Despite this ,
White continued with 1 de ! disre­
garding the threat. White's defence
is based on a m utual two-fold attack,
which occurred after 1 . . � x c2
.
2 EI X d 8 + EI X d 8 3 � X f7 + !
89
Black must reply with 3 . . ® x f7
.
after which White takes the bishop
at c2 . The operation initiated by
White had to be calculated with
great precision.
The check with the bishop on f7 af­
ter 1 . . . El x d 1 + 2 EI x d 1
� x c2 was a mistake, because
White would have himself become
the victim of a two-fold attack by
3 . . . <J/f8 ! 4 � x g6 � x b3. Only
3 ®a2 ! ! was correct. This ingenious
solution enables White to remove
his queen from the line of fire. His
reply to 3 . . . E H8 is 4 EI d7 and
squ are f7 can no longer be covered.
You will surely have recognized
how the player responsible for the
mutual two-fold attack extricated
himself from it. He usually resorted
to a check to give one of his pieces
just enough time to leave the d anger
zone. Here ::tre a few more exam­ Here White executes the mutual
ples. two-fold attack by playing 1
� x g6 ! After 1 . ® x g3 White
. .

K. Richter- Winz first plays 2 � x e7 + and then 3

Berlin 1 95 7 El. x g3, capturing a piece. Even if
B lack had chosen the best possible
90 reply under the circumstances, i . e.
1 . . � dS, White still wins by 2
.

� x f8 + .

Thomas-Euwe
Nottingham 1 93 6

92

White to play

By playing 1 El x e6, White re­

gained a piece with the aid of a mu­
tual two -fold attack. His operation
is based on the fact that in reply to
1 . ® x f6 he is able to give check
. .

by 2 El. c 1 + thereby saving his Black to play

rook.
Black played 1 . � e6, apparently
. .

Afatortsev-Zamikhovski assuming that after 2 El. x d6

Moscow 193 1 El x d6 3 ® x d6 El. d 8 a mutual
two-fold attack is created from
91 which he extricates himself by 4
® X c7 .El X d l + with advantage in
m aterial. But White found a differ­
ent way of defending himself and of
protecting his position against ha­
rassment by the rook. The attempt
to recover the piece by means of a
pin after 4. . ® x d6 5 El. X d6
.

� f8 was frustrated by the dis­

covered check 6 � f6 + .
White to play

53
The double attack R. Reti, 1928

94
Up to now we have examined a
wide variety of cases of two-fold at­
tack, and you might have noticed
that we never used the term double
attack. We believe that the two-fold
attack is only a special case of the
double attack. We want to broaden
the concept "double attack" to
cover both attacks and threats of
the first and second order. Let us
consider a few examples. White to play

F. Saavedra, 1895 This is the finale of an equally well­

93
White to play
In front of us we have the finale of a attack on the bishop and a threat to
well-known study. By playing 1
dlb3, White attacks the rook and
threatens checkmate on c 1 . It is not demonstrates another variant of the
difficult to see that we have the
same situation here as in the two­
fold attack : It is not possible to
parry the threatened checkmate and
to save the rook in one move. This
particular double attack consists of
an attack and a mating threat.
known study. First of all White
plays 1 e7 forcing Black's bishop to
move to the square bS. Then he at­
tacks the piece by 2 dieS .
This move constitutes an attack on
the bishop bS and in addition it
poses the threat of the king entering
the quadrangle of the pawn on g4.
Black is unable to refute his oppo­
nent's two-pronged plan with a
single move and the game ends in a
draw. We can justifiably regard the
move 2 dlcS, which consists of an
stop the pawn as a double attack.
The finale of the following study
double attack.

54
L. Kubbel, 193 7 White plays 1 ®b3 thereby attack­
95 ing square d S , which is covered
only once, with two pieces. At the
same time he threatens to capture a
pawn after 2 � x f5 rl x fS by 3
rl x e4 . One can easily see that the
double attack in this case consists of
a two-fold attack and a threat of the
second order.

T. Lawson, 1925
97
White to play and win

In this position White first plays 1

cl, thereby forcing the bishop to
take the pawn ( 1 . . . � x c7) and
then he attacks it by 2 d)c6. The
bishop may not leave the diagonal
a5-d8 because of the threat of
checkmate 3 d)b6. That is why
Black has to play either 2 . � aS
. .

or 2 . . � d 8 . But then White has

.
White to play and win
another threat in store : a two-fold
attack by discovered check by Lawson's study shows a double at­
means of 3 d)bs + or 3 d)d7 + . tack consisting of a combination of
You will notice that this double at­ two threats. White begins with 1 f7
tack consists of an attack coupled cbg7 2 �e7 ! Since the pawn threat­
with the threat of a two-fold attack. ens to queen, the reply 2 . . d) x f7
.

is forced, after which the second

Keres-Sliva threat of the capture of the knight
Goteborg 1 955 by 3 � b4 is realized.
96 98

White to play White to play and win

55
We have examined s ituations simi­
lar to the one shown here before.
White's move 1 EX h8 is a double at­
tack consisting of two collateral
threats, i. e. queening and-if Black
should choose to defend himself by
taking the pawn with his rook­
mounting a double attack on the
king and rook.
Balashov-Biyiasas
Manila 1976
99
White to play
White sacrifices his rook for a
bishop by playing 1 El x d7 in order
to reply to 1 . ® x d7 with 2
. .
®g4, threatening mate on g7 and
the capture of Bl ack's queen by 3
� h6 + . White's move 2 ®g4 con­
sisting of two threats-a mate and a
two-fold attack-is also a typical
case of a double attack.
56
A. and K. Zarytshev, 192 8
Conclusion ofa study
1 00
White to play
White plays 1 dld7. He doesn't
seem to threaten anything. Black's
pawn has moved far ahead, and
White's pawn may not queen be­
cause of the threatened check at fs
by the bishop. But if we take a
closer look things are qu ite differ­
ent. In reality 1 dld7 is a double at­
tack consisting of two threats of the
second order. White is about to
threaten the pawn at bS and at the
same time to queen his own pawn.
If Black hesitates to advance the
pawn, it will be checked after 2
dld6 � fS 3 dieS . Black must antic­
ipate these threats before it is too
late .
Let us assume he plays 1 . . .
� fs + . This is followed by 2 dld6
b4. There is no other defence
against 3 dieS . This is the double at­
tack, which-analogously to exam­
ple 94-consists of an attack and the
threat to catch up with the pawn.
Consequently Black was virtually
powerless against the double attack,
which started with the first move
consisting of two threats of the sec­
ond order, although this was not so
apparent at first glance.
I hope you will agree with me that
both two-fold attacks and double
threats, namely those of the first
and second order, as well as combi­
nations of attacks and threats can
be regarded as double attacks, al­
though the methods of defence
against them may vary. S uch an ex­
pansion of the term double attack
enables us to make the fol lowing
extremely broad classification. A
double attack can be
1. a pure two-fold attack;
2. a combination of an attack and a
two-fold attack ;
3. a combination of an attack and a
threat;
4. a combination of a two-fold at­
tack and a threat;
5. a combination of two threats of
the first order;
6. a combination of two threats of
the first and second order ;
7. a combination of two threats of
the second order.
The threats, too, can "vary consider­
ably. They can be threats of a
checkmate, threatened attack,
threats of a two- fold attack, threats
of queening or of stopping a queen­
ing pawn, etc. In addition, as you
know, these threats can be of differ­
ent orders .
Defence against
the double attack
We have seen that there can be a
wide variety of double attacks, and
the range of appropriate defensive
techniques is equally extensive . If,
for instance, a chessman covered
only by one piece is attacked by two
pieces , one can use an elementary
defensive technique, which is to dis­
patch another piece to its defence
by restoring the balance between
offence and defence. This can be
done to give support or to unite
pieces and/ or pawns designed to
frustrate the effect of one or l?oth of
the attacking pieces by interposition
or shielding.
We have already discussed such
methods as support and imerposi­
tion in the case of the single attack.
In principle these defence variants
in more complex situ ations differ in
no way from the simple case . In
both cases, the mobility and activity
of the supporting or interposed
pieces suffer because they are tied
and pinned. Double attacks di­
rected at two pieces give rise to typi­
cal defence systems. Let us examine
them more closely.
J Kling, 1849
1 01
White to play and win
57
As the diagram clearly shows, Black that the best way to defend yourself
has set the stage for a double attack : against a double attack is by coun­
he is attacking the rook and his tering it with a double attack of
queen is poised to checkmate the your own.
king on square a l . White's situation
seems to be critical. But there is a
A. Kakovin, 1 961
way out : 1 Ela 4 ! This is an effective
Conclusion ofa study
way of protecting the rook, because
it not only enables it to leave the 1 02
d anger zone but also to pose the
threat of a checkmate on square h 3 .
White takes advantage of the fact
that the rook is unassailable after
1 . &1 x a4 2 Elh3 + dle4 3
. .

El h4 + thanks to the two-fold at­

tack. Black can parry the checkmate
by 1 ® c8 , but after 2 El h3 +
. . .

® X h3 3 Ela3 + his queen never­

theless succumbs to a two-fold at­
tack.
We can derive the following rule
from the above : The two-fold at­
tack remains ineffective if one of
the pieces under attack can d odge
the attack and in its turn pose some
dangerous threat. As illustrated
above, this can either be the threat
of a checkmate, of capturing a hos­
White to play a nd win
In this position White must protect
himself against the threatened
double attack 1 . . . 00g3. He solves
this problem by playing 1 � c7 !
This move is in preparation for a
two-fold attack : 1 . 00g3 2 � g6 !
. .

This is the whole point of the exer­

tile piece, of queening, etc. The as­ cise ! White surrenders a piece tem­
sailed piece can also extricate itself porarily only to regain it without
by giving check to the opponent's delay after 2 . � x g6 by dis­
. .

king. Let us return to Kling's study. covery check 3 fS + .

I am sure you will have noticed that
White's move 1 Ela4 ! was actually a
d ouble attack, because in addition
to threatening the queen White has
also set up the threat of a check­
mate. Try to fix firmly in your mind
Although peculiar, the method of
d efence shown in the next magnifi­
cent study, in which a double attack
is used to refute a double attack, is
characteristic of such positions.

58
S. Kaminer, 193 5
1 03
White to p l ay and win
After 1 h7 � h 5 ! White unexpec­
tedly found himself in a precarious
situation . If he decides to queen 2
h8 (®), Black plays 2 . . . � x g6 +
3 cba t � e7 !, after which the
deadly threat of a double attack
crops up through 4 . . . � f6 + .
White is unable to block the dia­
gonal, because 4 � f3 � f6 + 5
� e5 + is followed by 5 . . . cbe7 !
with a bad position. How can White
parry this threat? We know that the
best m ethod of defending oneself
against the double attack is by
mounting a double attack of one's
own. But how can this be done ?
Finding a solution is anything but
easy.
2 � f4 ! ! . . .
White forces Black to remove his
knight by attacking his bishop. For s urprising moves-moves that seem
the time being it is unclear what this entirely out of the question at first
changes, because the threat of a
d ouble attack remains . Let us see
what happens then :
2 . gf 3 h 8 (®) �g6 + 4 di a l
. .
� e7 5 � f3 � f6 + 6 � eS + cJde7
1 04
Now we realize that the knight was
·
not sacrificed in vain, because in
taking the knight, the pawn at g5
had to vacate the diagonal h4-d 8 .
White elegantly makes use of this
circumstance by playing 7 ® h4 ! !
By pinning the bishop he eliminates
the possibility of a strike on e5 and
threatens to capture the pawn at f4 .
7 . . . � x h4 is followed by a deci­
sive double attack 8 � x g6 + and
9 � X h4, which gives White the
victory.
In this case, White used a pin to
repel a double attack, but it was a
well-planned double attack that
brought him victory.
Please bear in mind that the most
sight-can occur in such tactically
critical situations.
59
Alekhine- Verlinski Ed. Lasker-Marshall
Odessa 1 918 New York 1924

1 05 1 06

White to play White t o play

Black intends to mount the double

Black had j ust assaulted both of
attack 1 . . . ®e5 attacking the
White's rooks by an effective ma­
rook and threatening checkmate on
noeuvre with his queen.White cannot
h 2 . If, for instance, White plays 1
take the queen because of the
.El e 1 , Black's attack after 1 . . .
threatened checkmate on f1 . And
® e S 2 OOgl E1 b2 decides the issue.
White cannot extricate himself
White found the only defence possi­
whatever he does : He can play
ble in this predicament : He covers
neither 1 .El cd 1 because of 1 . . .
both targets with his queen : 1 ®e3 !
® x d6, nor 1 � c4 because of
® e s 2 ®g l .
1 . . . � x c4, nor 1 .El d d 1 because
If Black plays 2 . . . ® x fS then 3
of 1 . . . ® x e5. But he found a
Ei fl ® g6 4 Ei e l E1 e8 5 ® e 3 , with
brilliant move which solved all his
good prospects of a draw. After
problems : 1 ® d 1 ! ! thanks to which
2 . . . e3 3 bl e 1 e2 4 � f3 ® x fS 5
he can assert his superiority in mate­
� g4 ® ds + 6 ® g2 ® x cS 7
rial. If Black takes the queen on d 1 ,
® x e2 White would also have had
White can take his queen o n c7. But
good chances of defending himself
i f he replies with 1 . . . ® aS , then 2
successfully.
® d S is the most expedient move . In
this case White was able to use the In some situations it is also possible
simp lest method of defence against to refute a double attack by moving
the double attack on two pieces, one of the pieces attacked out of
viz. by covering both of them in a danger and posting it to cover the
single move . But this move was not other piece or some other endang­
easy to find. ered target.

60
Reshevski-Euwe
The Hague 1948
Game variant
1 07
Black to play
A double attack has been launched.
White is attacking the knight on d7
and threatening checkmate on h7.
Despite the acute d anger, Black has
an adequ ate defence : 1 . . . � e4 ! In
this manner White's queen is di­
verted to squ are e4. If the knight
takes on the squ are e4, the threat is
suspended for a moment and Black
has time to reply 2 . . . B. fcS . But to
2 ® x e4 B lack answers 2 . . . � f6 !
attacking the queen and covering
square h7 at the same time.
In view of a further double attack
by 2 � d7 ! !, 1 . . , � f6 ? instead of
1 . . . � e 4 ! woul d have been a mis­
take. Now 2 . . . � e4 is useless be­
cause of 3 � x f6 + gf 4 ® x e4,
and \X!hite wins.
Smyslov-Zita
Prague 1946
1 08
White to play
Bl ack, who had just p l ayed B. e7,
pins all his h opes on threatening
square f2 . But White was able to
prove that the resulting double at­
tack held no dangers for him. This
is how the game continued :
1 t!!J X e7 ! t!!J X d4 2 t!!J d8 + \l>h7 3
® x c8 . . .
,
White has calculated every move
and is confident of success.
3 . . . EI X f2
The stage is set for a discovered
check, but Black cannot avail him­
self of it.
4 ® x c4 . . .
By taking the pawn, White at the
same time covers the rook. 4 ® c5
would have been a much weaker
move because of 4 . . .El x g2 + ! 5
.
dl x g2 ®d2 + and 6 . . . ® x c l .
4 . . . ®d2 5 &:t c2 + !
Destroys all illusions the opponent
may have had. Through this ex-
61
change, wh ich is also based on a
double attack, White simplifies the
p osition and asserts his superiority
in material.
5 . . ® x c2 6 El X c2 El
. x c2 7
� e 4 + and White wins.
The possibility that the attacker's
situ ation m ight, for some unforeseen
reason, deteriorate after a double
attack is also conceivable.
After 7 .El. x d 1 c2 ! the pawns queen
in short order. You will note that
7 . . . c2 is also a double attack by
which Black attacks the rook and
threatens to queen at c l .
The dangerous passed pawns gave
Black a decisive advantage after at-
lowing White to make the double
attack. Other factors can also be of
considerable i mportance .

Geller-Averbakh S. Isen �gger> 19 59

Kiev 195 4
1 10

1 09

White to play and win

Black to play
Black threatens to eliminate White's
In this position we see White's
last pawn by 1 . . . eS. White's at­
threat to mount a d ouble attack, but
tempt to prevent th is by 1 � d7 is
Black decides to let him try it. The
countered by Black with 1 . . . �bS
game then took the following
and his double attack enables him
cours e :
to capture a knight.
1 . . . El X eS ! 2 de 'l.!i::J X e5 3 'l.!i::J X e 5 White is not deterred by this and
� X e5 4 El X gS + d; h 7 5 El g X eS continues with 1 � d7 ! It becomes
apparent after 1 . . . �bs 2 d;ds
White has captured a rook through
� x d7 3 � c7 + dlc6 4 � e S that
the double attack, but Black's
Black is in Zugzwang and is forced
pawns on the queen's wing are
to surrender his bishop.
worth much more.
A similar, although somewhat more
5 . . . be 6 El. b S .El. d 1 ! , and White complex situation is shown in the
resigned. next diagram.

62
]. Fritz, 195 1
111
White to play
White's chances of winning do not
seem very good in this situation :
The knight at b7 hangs and Black
threatens to play 1 . . . � x h3 2
� x h3 � f4 and to capture White's
last pawn (this is a threat of the sec­
ond order) . The game takes the fol­
lowing course :
1 � d6 � X h3 2 � fS + . . .
The check must be given pr�cisely
at this point because the king can­
not move to h S .
2 . . . digs 3 � x h3 � f4 4 � d4 !
White tries to save his pawn by
threatening a double attack.
4 . . . � X h2 5 � f3 + dif4 !
Black, too, pins all his h opes on the
double attack 6 . . . dlg3 , which
can be mounted as soon as Black's
bishop is taken by the knight. But
White has thought further ahead .
6 dlf2 ! ! . . .
The bishop suddenly finds itself i n a
hopeless situation. After 6 . ·. .
� g 3 + 7 dlg2 Black has no satis­
factory way of continuing the
game.
Black cou ld also have taken a dif­
ferent line and played 4 . . . dlh4
before taking on h2 . In this case 5
� fl ! � x h2 6 � f3 + dlg3 7 dle3
would have followed, and Black
would again have lost the bishop as
a result of Zugzwang.
To conclude this chapter, let us
consider an example in which the
course of events is understandable
without much comment.
Bernstein-Smyslov
Groningen 1946
1 12
Black to play
Black has a winning position. But in
attempting to speed up the course
of events he decided to play 1 . . .
b 2 ? 2 &l x b2 ! r:! h2 + relying on a
double attack. Only after White's
surprise move 3 dlf3 did he con­
cede a draw as a result of stalemate.
63
Forcing
the double attack
We have already noted that the
double attack is one of the most
dangerous methods of attack. But
we also know that one can antic­
ipate it if one recognizes the threat
early enough . The double attack is
particularly dangerous if it occurs
suddenly and is forced , leaving no
time for counter- measures . In the
present chapter we shall find out
how and why such a situation oc­
curs.
1. The attack
One can easily imagine a situation
in which the king is attacked and
forced to submit to a double attack. double attack.
The situation shown in the next d ia­
gram is typical of such cases.
K. Geller, 18 73
1 13
White to play and win
A series of skilfully executed ma­
noeuvres enables White to force the
king into a fork :
64
1 �g6 + ctde4
Black has no other option , because
1 . . . ctdfs loses i mmediately be­
cause of 2 � e7 + , and after 1 .. .
ctde6 2 � f6 + White gives check
with the knight on e7 or e5, de­
pending on whether the king moves
to d5 or d 7 .
2 ® h 4 + ctde3
Again the only possible reply, be­
cause any other move would res u lt
in a fork either by the knight or by
the queen.
3 ®e 1 + ctdd4 4 � g 1 + , and wh er­
ever the king turns he cannot escape
a double attack.
In the position shown in the next
diagram the king covers a long d is­
tance before it succumbs to a
Z. Vecsey, 193 5
Conclusion ofa study
1 14
White to play and win
1 � e7 + dlg7
1 . Otherwise the king is check-
mated in two moves.
2 ®g8 + <l>f6 3 � d5 + \t>e5 4
®g7 + ! . . .
Only the last move shows that even
in the centre of the board the king
has only a narrow path to move on
if a double attack is to be avoided.
But even this path leads to a preci­
ptce.
4 . . . \t>d6 5 ®e7 + d)c6 6 ®c7 +
d)b5 7 � c3 +
The king can be driven off not only
by direct attack but also by a threat­
ened attack.
A. Troitzky, 1896
1 15
White to play and win
If White wants to win he must free
the rook covering the pawn on a7.
But how can this be done if Black's
king keeps hiding behind his rival's If Black plays 2 . . . 00d7 in order to
back? And yet White retains the up­ move up closer to the dangerous
per hand by driving Black's king
towards a disastrous double attack knight by 3 � c5 ! di e S 4 �b4 and
by threatening check.
1 ctdf4 ctdf2 2 ctd e4 d)e2 3 ctdd4
ctdd2 4 ctdc5 ctdc3
There is nothing better. If 4 . . .
El e t + 5 ctdb4 El b l + 6 dla3
El a 1 + , then ? ctdb2 and there is no
escape from 8 El d 8 + . But now we
have arrived at a position which we
are familiar with (cf. diag. 43) in
which White wins by a discovered
check.
5 .bl c8 ! El x a7 6 ctdb6 +
The king can be driven on to a
dangerous square not only by
threatening to attack it. This can
also be achieved with various other
kinds of threat. In the study shown
in the next diagram, for instance,
the king runs into a fork only in the
course of a conflict with a hostile
pawn .
A. Troitzky, 1924
116
White to play and win
1 f6 00d6 2 � X d 4 � d2 +
pawn , then White cuts off the
captures it then .
3 dl e 2 � e4 4 � e5 + ctd d 7 5 f7
·'!de7 6 dle3 . . .
We now realize that the knight has
no suitable escape route, White's
65
decisive reply to 6 . . . � cS being 7 W Bran, 192 7
� d6 + and to 6 . . . � g5 7 � f6 + .
118
P ressure on the king is often cou­
pled with an attack on other men,
for instance as shown in the next
two d iagrams.

A. Troitzky, 1914
117

White t o play a n d win

1 b:l c8 ! ® a3 2 � d4 + dlb6 3
b:l b8 + OOcs 4 bj bs + 00d6 s
B ds + 00e7
It looks like Black's king has ma­
naged to escape from the danger
White t o play and win zone , but . . .
6 b:l aS ! ® X aS 7 � c6 + or 6 . . .
In this position White must try to
® d6 7 � fS + , in both cases White
fork the king and the queen . The
captures the queen.
first move is not hard to find.
S. Kaminer, 1925
1 El b7 ! ® g 8 !
119
The strongest possible response.
The queen tries to put as much dis­
tance as possible between itself and
the king.
2 � eS + dieS 3 b:l b8 ! . . .
If Black's queen had been on a 8 ,
this move would have settled mat­
ters on the spot.
3 . . ® h7 4 b4 + dld6 s b:l h 8 !
.

White t o play and win

® X h 8 6 � f7 + , and White wins.
A similar combined attack on the
king and queen also leads up to a
double attack in the following
study.
A combined attack by the king and
the b ishop results in a double attack
on Black's knight. This is how the
situation develops :

66
l � c4 � c5 +
If the knight had moved to b 8 ,
White would have captured i t after
2 d/ x e7 + 00g7 3 d:>d6.
2 cbds � a4 3 �b3 � c3 + 4 cbc4
� b 1 5 � a2 � x a3 + 6 d:>b3 � bS
Up till now the knight has been
moving along a precisely prede­
fined p ath thereby avoiding the
double attack. But at last Black has
ru n out of defensive resources and
White wins by 7 cba4 + .
The next diagram illustrates a
double attack that materializes as a
res ult of a counter-attack.
Podgayets- Vladimirov
Chelyabinsk 1975
1 20
Black to play
Black thought there was still time to
defend himself against the threat d6
by playing 1 . . h6. But White
.
nevertheless played 2 d6 !, and after
2 . . . hg 3 de ® x e7, the pawn on
g6 was without cover, so that the
double attack 4 � X g6 gave White
a qualitative advantage.
2. Zugzwang
Zu gzwang- i. e. the obligation to
m ake a move-plays an impo rtant
role in the end-game. Zugzwang
c a n also be used for bringing about
the double attack.
H Rinck, 1 91 6
1 21
White to play and win
White's objective is to force, by
Zugzwang, the hostile queen to
move away from its king on the 8th
rank. White manages to do this by
skilful rook manoeuvres.
1 El h7 + ! . . .
1 El f7 would be a mistake because
of 1 . . . ®d6 !, as a resu lt of which
Black's queen would be out of
d anger.
1 . . . digS 2 � he7 cbhs 3 � bc7
digS
This is the toughest defence.
Weaker would be 3 . . . ®g8 4
cbfl ! ® f8 + 5 � f7 � g 8 6 r! a7 ,
because after 6 . . . ®e8 White
would bring his threat to bear : 7
r! h7 + cbg8 8 � ag7 + cbf8 9
r! h8 + .
67
4 � ca7 cbh8 S rl f7 -&1 e 8 + By 1 � a S , White deprives the op­
ponent's king of its freedom of
If S . . . ® g 8 , then 6 cbf1 , and
movement, for any move it makes is
Black is in Zugzwang.
immediately punished by a double
6 \t>f2 \t>g8 7 rl g7 + \t>f8 8 rl h7 attack. Thus, only the bishop re­
\t>g8 9 r:l ag7 + \t>f8 1 0 rl h8 + , mains free to move : 1 . . . � h4 2
and White wins. \t>g4 � e 1 3 cbh3 ! and after that
Black is in Zugzwang. Now every
A classic example of a double attack
move made by the king or the
is shown in the following well­
b ishop results in a double attack.
known study.
L. Kubbel, 1 909
R. Reti, 1922
1 22 1 24

White to play and win

White to play and win

After 1 g d4 + \t>cs 2 \t>h 1 ! ! Black Here, White manages to capture

is in Zugzwang, because any move
the rook by a double attack thanks
by the bishop results in its loss as a
to Zugzwang.
result of the fork.
1 � cS l:l c8
L. Kubbel, 1928 If Black plays 1 . . l:l b 8 , then 2
.

1 23 rl h8 + cSdc7 3 � d6 + . Neither can

the course of events be changed by
1 . . . cSdcS 2 � a7 ! aS 3 cbd 1 a4 4
di e t a3 5 cSdb 1 a2 + 6 et/ x a2, be­
cause after this Black is in Zug­
zwang.
2 � b6 + dle8 3 � c7 ! aS 4 did 1 a4
5 diet a3 6 \t>b 1 a2 + 7 \t>a 1 ! d/f8
8 l:l h8 + , and White wins the
game.
White to play and win

68
3. The exchange
One can easily imagine that the
double attack can also be forced by
an exchange. Here is a very simple
example.
1 25
White to play
By a series of exchanges White
clears the way for its rook to the
eighth rank: 1 bl x e8 + ® x e8
( 1 . . . � X e8 2 � h7 !) 2 � X b 5
� x b 5 3 bl d 8 + , a n d the double
attack has become reality.
Berger-Schmid
1 26
White to p lay
This exampl e is taken from Lasker's
"Manual of Chess". White succeeds 3 . . . ® g5
in converting Black's insecure posi­
tion to his advantage through a
series of exchanges and then in
l au nching a double attack.
1 g4 ! � g6 2 gf � X fS 3 � X e4
� X e4 4 � c3 � X f3 5 � X d5
� X d 1 6 � X cl + . . .
The upshot of the game, which has
been forced by an exchange, is that
White achieves material superiority.
6 . . . \t>f7 7 � X a8 � f3 8 de and
White wins.
Kan-Lowenfisch
Tbilisi 193 7
127
White to play
The uncovered Black b ishops
p rompted White to work up a
double attack on them by means of
an exchange.
1 � X d4 ! rl X d4 2 � e2 ! fl X d 1 3
® x dl . . .
These forced exchanges enabled
White's queen to move to line d,
from where it can penetrate to d7
and attack the two bishops.
69
Black wants to protect one of the vary considerably depending o n the
bishops with the queen. Bad was position, but th is is often neglected
3 . ® x e4 because of the "fork"
. . in practice.
4 � f3 . Particularly in tactically critical po­
sitions involving attacks and threats
4 h4 ! ® f6 5 e5 ! . . .
these relative values are only of lim­
Forces the queen to abandon the ited validity. Very often the
b ishop. strength of a particu lar man is meas­
u red by the role it plays in a planned
5 . . . 'd1J g6 6 tm d7 � X g2 7 d/ X g2
tactical operation. If in order to
� X h4 8 'diJ d 3 ! 'diJ X d3 9 � X d 3 ,
checkmate your opponent's king it
a n d White wins.
proves necessary to demolish its de­
We have seen how the double at­ fensive system, then the material
tack was forced through sacrifices. sacrificed for this purpose is well
The next chapter is devoted to the spent, because one is repaid a hun­
s acrifice as a means of forcing the dredfold. The same applies to the
d ouble attack. double attack. If a double attack
promises to bring a material advan­
tage then it is worth sacrificing
something in order to ensure suc­
The s acrifice cess. The whole trick consists in
first sacrificing something and then
S acrifices are , of course, an effec­ getting something in return . Such
tive means of forcing the double at­ sacrifices always involve a certain
tack. If for no other reason, they are amount of risk, because if the oppo­
effective because as a ru le they nent thinks of a way to frustrate the
come as a surprise to the opponent. plan, the sacrifice is made for no­
Their unexpected ness and sudden­ thing. This means that such sacrifi­
ness is due to the fact that we tend cial ploys should be considered
to be too conscious of the relative thoroughly and often in breadth
values of the pieces in normal situa­ and depth before it is decided to re­
tions. sort to them.
At a very early stage of learning the A sacrifice that forces or helps bring
fundamentals of the chess game we about a double attack can bring
learn that the queen is much more about widely different results . It can
valuable than the rook, that the steer an opponent's piece to a cer­
rook is stronger than the knight or tain square on which it can be sub­
bishop, which in turn are consider­ jected to a double attack immedi­
ably superior to pawns in value. ately or later on.
These truths have become second The following examples from
nature to us. And yet we know that games played in top-level competi­
they are not immutable, that they tion are characteristic.

70
Averbakh-Horberg
Stockholm 19 54
1 28
Black to play
The awkward position of White's
queen, rook and bishop prompts
Black to sacrifice a rook in order to
launch a double attack : 1 . . !:l e t !
. mobilized in a double attack.
Black lures the queen to square c l
in order t o capture it immediately
after 2 &:1 x cl by 2 . . . � x e2 + 3
r:l X e2 'df:J X c l + .
The rest is easy : 4 00f2 � a6 5 � d 3
&:1 x a l 6 � x a 6 ® d l . White gave
up.
Pirc-R. Byme
Helsinki 19 52
1 29
Black to play
In this situation, too, the deploy­
ment of White's pieces prompts
Black to sacrifice the exchange in
order to mount a double attack.
1 . . � f5 + 2 OOd3 n x c3 + ! 3
.
00 X c3 � e3 4 r:l X f2 � d 1 + 5
00d4 � X f2
The difficulty of this tactical opera­
tion consisted in making a correct
assessment of the end-game. After
that B lack had no difficulty in win­
ning the game. Such a combination
of sacrifices and double attacks is a
standard procedure which one
should be thoroughly familiar with .
H ere is a classic example of a sacri­
fice by means of which pieces are
Em. Lasker
1 30
White to play and win
To lure Black's rook to c 8 , White
first sacrifices his rook by 1 n c8 + !
r:l X c 8 , after which he also gives
away his queen to recover all the
lost material and more in a single
m ove through a double attack : 3
be( � ) + ! !
71
A series of sacrifices designed to
draw the queen into a fork is de­
monstrated in the following old po­
sition.
B. Horwitz, 1881
1 31
Black rejects the first sacrifice of the
queen because this would have been
followed by 3 � x c6 + and 4
� x d8 with victory.
3 � c4 + ! rl x c4 4 ® aS + ! !
\tl X aS
For all that White still managed to
force Black to accept the sacrifice.
This time he is defeated as a result
of a slaughter brought about by a
double attack.
S � x c4 + d/b4 6 � X d6 + and 7
� x fs .

Kikovic-Forintos
Budapest 195 7
1 33
White to play and win

1 ab(® ) + 'f!!:t x b8 2 � d6 +
® X d6 3 ij X d2 + 'f!!:t X d2 4
� x f3 + and 5 � X d2 with vic­
tory.

E. Pogosyants, 1959

1 32
White to play

White realizes the relevant sacri­

fices as follows :
1 il X c8 + ! . . .
In this manner the rook is lured to
square c8 .
1 . . . il X c8 2 d7 . . .
This sacrifice steers the king to d7.
White to play and win
2 . . \tl X d7 3 � g4 +
.

Here the sacrifice precipitates two

White recovers the material he has
double attacks. j ust sacrificed with interest and
1 � eS + et> as 2 ® d8 + cl/bs compound interest.

72
Eliskases-Kan advantage of the position of Black's
Game variant king and queen by 2 El x f7 + � By
eliminating the pawn at f7 White, in
1 34
response to 2 . . El x f7 , is able to
.

continue with 3 � e6 + and 4

� x cl. However, if Black had re­
jected the sacrifice by playing 2 . . .

00g8, events would have taken the

following course :
3 El g7 + ! 00h8 4 El X h7 + d/g8 5
El g7 + dlh8 6 El x g6 by winning.

Tolush-Antoshin
Leningrad 1 95 6
White to play
1 36

In this case, too, White succeeds in

involving two of B lack's pieces in a
double attack.
l El X f7 ! El X d6 2 El X g7 + !
dl x g7 3 de + and 4 El x d6.
A s acrifice designed to eliminate the
defence is frequently resorted to be­
fore a double attack. Here is a typi-
cal case. Black t o play

Kupper- Olafsson Black's pieces appear to be hovering

Zurich 1 95 9 threateningly over Black's king. But
the ominous confrontation of the
1 35 two queens enabled Black to strike
back with a s acrifice followed by a
double attack.
1 . . . El X d3 !
Much to his distress, White has to
accept this sacrifice with his rook
because his queen is busy covering
the p awn at g2. But now the queen
is without protection. After 2
El x d 3 El e 1 + 3 dlf2 � e4 + 4
White t o play
m X e 1 ® X g6 Black captured the
After 1 � X g7 m X g7 White took queen and soon thereafter won the
game.

73
Ljubenski-Stepan iec
Poland 1 95 5
1 37
Black to play
The situation is very tense. White
has j ust staged a double attack and
Black is threatened with mate by
El a4 and the knight at d4 hangs .
Drastic measures are called for.
Black could , of course, move the
kn ight back to c6 in order to ward
off the check with the rook by � a S .
But he fou nd a better way out. If he
gives check on c2 with his queen ,
Black cannot create a knight fork
on f3 because the pawn on g2 con­
trols this square. But perhaps the
pawn can be eliminated. As soon as
you have studied this position you
will have no difficulty finding the
effective sacrifice that ends the
game immediately :
1 . . � f3 + !
.
Due to the mate in one move by the
knight or the queen, the king can­
not retreat, but if White plays 2 gf,
Black decides the issue by 2 . . .
® c2 + , followed by 3 . . .
� x f3 + forking White's king and
queen.
74
Rabinovitch-Chekhover
Len ingrad J 933
Game variant
Black to play
The pawn on f2 is White's sole pro­
tection against the double attack by
the bishop and the rook on squ are
g2. How can it be eliminated ? With
the aid of a sacrifice : 1 . . ®g3 ! It
.
has to be accepted by 2 fg, but this is
followed by 2 . . El x g2 + 3 dlh 1
.
El dd2, and there is no escape from
a mate .
A sacrifice might also be resorted to
for opening a file or rank in order to
stage a double attack.
Bogolyubov-Capablanca
New Yo rk 1 924
1 39
Black t o play
A classic model. Black first sacri­ Lowenfis eh- Ryumin
fices his knight by 1 . . � x d4 2
. Moscow 1936
cd only to recover it by 2 . . .
!l 8 x cS. It becomes clear that 3 de 141

is worthless due to the double at­

tack 3 . . . ® x cS + by which the
rook at c 1 is captured .

Alster-Betak
Prague 1956

1 40

W h ite to play

White played the knight back to g3

despite the fact that he had a splen­
did opportu nity to engineer a
double attack in the form of two
threats of the second order by mak­
ing a sacrifice. The correct continu­
White to play ation, which would have immedi­
ately sealed Black's fate, was 1
White sacrificed a bishop by 1 � f6 + ! gf 2 ef. White has dislo­
� X h6 in order to dislodge the de­ cated the defensive position of his
fensive system of Bl ack's king. After opponent's king and gained a
1 . . gh 2 ®g3 + , White's queen
. threatening pawn on f6 by means of
moved speedily to the diagonal the sacrifice. As a result two threats
h2-b 8 , which is dangerous for the have been created with one stroke,
rival queen, making it possible to i. e. 3 ®g3 + followed by 4 ® g7
answer any move by the king with 3 m'ate and 3 ® x f8 + ! dd x f8 4
� g6 + and 4 ® x c7. EX d8 mate. There is no defence
against them.
The next example also illustrates
the destruction of the opposing
king's fortifications and the open­
ing of files for a subsequent double
attack as a means of ending a game
victoriously.

75
W Richter- Casper
Benshausen 1 9 75
1 42
White to play
Black had j ust moved his knight to
b4 and attacked the pawn on a2. He
was expecting White to reply 1
bl c7, which could have been fol­
lowed by 1 . . . ® x a2 2 � f6 +
dlg6. But events took an entirely
different cours e :
1 � f6 + ! gf
The knight had to be taken, but this opens a file for the rook on d 8 . At
laid bare the king's position.
2 EI c7 + dlg6 3 ® d l !
This double attack-White is at­
tacking the rook and threatening
mate on g4-became possible
thanks to the sacrificed knight.
76
Sto/yar-Averbakh
Leningrad 1 93 8
1 43
Black t o play
The sacrifice shown in this example
was made for the three-fold pur­
pose of opening a file, pinning and
smashing the defence. Its ultimate
purpose is the same, i. e. to launch a
double attack.
1 . . � X g2 + !
.
By sacrificing the bishop, Black
the same time White's queen is
pinned when it takes the bishop ·and
thus becomes u nable to defend
square f2 against a check by the
knight.
2 ® X g2 EI X d 1 !
This has resulted in a mutual double
attack. White cannot play 3 ® x c6
because of 3 . . EI x e 1 + .
.
3 r:l x d l � f2 + 4 dlg l ® x g2 +
5 d/ X g2 � X d 1 6 i:l X b4
r:l X c2 + 7 dlgl EI x b2 and Black
won the game.
Decoy attack
You are probably familiar with the
technique of decoy in which a piece
(or a pawn) defending a certain
chessman or square is distracted
from its defensive task. This tech­
nique is also used for staging double in the next study.
attacks. Let us consider an elemen­
tary example.
Chekhover-Sokolski
144
Black to play
White defends himself against the
threatened mate on f2 with the
queen from e2. If Black attacks the
queen by 1 . . . � a6, he eliminates
the cover by a pin and forces imme­
diate capitulation. In this case the
double attack consisted of an attack
on the queen and the mating threat
on f2 . The decisive factor was that
the attack itself was two-fold, be­
cause as a result of it the queen is
pinned. Otherwise it could have
moved away from square e2 with-
out abandoning the point f2 .
Consequently in this case, too, the
inevitability of the events that took
piece was achieved with the aid of a
double attack.
A similar course of events is shown
A. Troitzky, 1 924
1 45
White to play and win
There are two ways in which Black
can defend himself against White's
queening after 1 h7, but White wins
in either case by distracting the hos­
tile rook thereby preventing it from
fulfilling his duties . In both cases
the distraction is effected by a
double attack :
(a) 1 . . . E1 h2 2 E1 f1 + ctdd2 3
Ei f2 + E1 X f2 4 h 8 (®)
(b) 1 . . . E1 d8 2 E1 c6 + d/d2 3
Ei d6 + E1 X d6 4 h 8 ( ® ) .
The tactical diversionary manoeu­
vre shown in the next diagram is
somewhat more complicated.
77
L. Sa/kind, 1 9 15
146
White to play and win
Both White bishops are under at­
tack, and Black threatens to take
the bishop on b2, giving check,
and then to return to the original
position to prevent the pawn from
q ueening if White's b ishop on e 8
m oves anywhere. For this reason
White first plays 1 h7 + ! thereby
d rawing Black's king away from the
pawn on e7 and forcing it to move
to a white square .
1 . . . d/ X h7 2 � a4 rl X b2 + 3
� c2 + ! . . .
Thi� brilliant check brings about a
double attack. White threatens to
queen and attacks the hostile king.
The bishop has to be taken , but af­
ter 3 . . rl x c2 + 4 cbf3 it is clear
.
that the king cannot return to the
8 th rank because its own pawn on
c5 is in the way. Black soon runs out
of checks :
4 . . . rl c3 + 5 00f4 B c4 + 6 cbfs .
78
Chekhover-Kaspa ryan
Yerevan 1 93 6
1 47
White t o play
White's queen attacks the knight,
which is covered by Black's queen.
By playing 1 ®h8 + White can
drive Black's king from his hideout,
but he does not achieve anything
s ubstantial. But White can also
launch a double attack by playing 1
h 5 . He is still attacking the knight
and threatening to checkmate the
king by 2 ® h 8 . Black's queen can­
not cover the knight and parry the
threat of a check. After 1 . . .
® x hs 2 ® x e6 White wins.
An analogous case of d iversion by a
double attack is shown in the next
example given by G. Lisitsin.
White to play
By playing 1 -®fs White threatens Smyslov-Lyublinski
mate o n h7 and keeps the sights Moscow 1 94 7
trained on the bishop on f6 at the
same time. To protect himself 1 50

against mate, Black has to reply

with 1 . . g6, but this deprives the
.

bishop of cover and results in a loss.

I n this case a mate threat diverts the
pawn.
One frequently finds diversion by
d ouble attack being used in practi­
cal chess. Here are a few examples.

Hogolyubov-Sultan Khan White to play

Prague 1 93 1
H ere, White first sacrifices a knight
1 49 by 1 � x f7 ! � x f7 in order to
mount a diversionary attack by 2
� b6 ! . If the bishop is taken by
2 . . . '® x b6, then the game is de­
cided by 3 'f!f:J x f7 + 00h7 4 h S .
T h a t is why Black's reply is 2 . . .

® d 7 . What happened then was : 3

� X d8 00h7 4 � X f7 'f!f:J X d8 5
� g6 + , and Black resigned.

Makogonov-M ikenas
White to play
Tbilisi 1 94 1
· l 'he knight on f4 saves Black from
1 51
1 11 ate in three ( 1 � e2 + 00a5 2
rl a7 + � a6 3 � X a6 mate ) . That
i .� why White brings about a double
.l t.tack by the d iversionary move 1
tX d 5 + � By attacking the king he
fl l rces Black to capitulate immedi­
.1 t.cly, because to 1 . . . dla6 White
n·plies with the decisive 2 � c8 + .

White to play

79
In reply to 1 t!t1 d7, Black played 1 53

1 . . � x f3 hoping that the threat­

.

ened mate by 2 bl g8 + ? t!tJ X g8 3

� x g8 C x g8 would force White
to return the queen immediately.
But White made a diversionary
thrust 2 r1 g7 !. Black may not touch
White's rook because the rook e8 is
under attack. Thanks to the threat 3
n X h7 + Black is forced to play
2 . . � gS , after which White cap­
White to play
.

tures a p1ece.
After 1 .bt c l + dlb 8 2 ®b4 + dla8,
Alekhine-Nestor
White gave mate in three by taking
Trinidad 1 93 9
advantage of the unfortunate plac­
1 52
ing of Black's king ( 3 � f3 + !
n x f3 4 ® e4 + ! ® x e4 5 .bt c8 ) .
Please note the peculiar features o f
these two diversionary sacrifices. I n
the first case, b y attacking the king,
White forces the rook to leave the
e-file and to leave the square e4 un­
covered. In other words 3 � f3 + is
an attack with a threatened check
by the queen on e4, which in itself is
White to play a two-fold attack. After removing
the bishop, White realizes the threat
White carried out an original op­ of this two-fold attack in order to
eration on the theme of diversion : 1 divert Black's queen from covering
.bt c8 !. By attacking Black's rook, he the square c 8 .
threatens to give mate by 2 ® f8 +
thereby forcing Black to play 1 . .
.

n X c8 . But this is followed by an

elegant diversionary thrust 2 ® e7 ! ! . The two-fold
A double attack has thus material­ double attack
ized : firstly White threatens a swift
mate by 3 d8 (®) + and, secondly 3 In the course of a game a situation
® x e6 followed by 4 de(® ) mate. can arise in which one player un­
Black capitulated because he was leashes a double attack, but his op­
unable to parry both threats. ponent is able to defend himself
The next position comes from a against it successfully. But the con­
game of the grandmaster Duras. flict is by no means over. A critical

80
situation ensues in which a two-fold Belenki-Pirogov
double attack may prove decisive. Moscow 1 958
This development in its simplest
form is illustrated in the followin g 1 55

study.

H. Rinck

1 54

Black t o play

In this position the rook on g 1 pro­

White to play and win
First of all White moves a pawn to
the 7th rank thereby threatening to
queen :
l h7 ! (not, o f course, 1 a7 ? R
� g8 � '
and Black saves himself) 1 . .
2 a7 E1 a l
tects White's king against mate on
g2 and the q ueen protects it from
mate on hS. A single move, wh ich
brings about a double attack, is
enough to end the game immedi­
ate I y : 1 . . . b'! e 1 �
F Sackman, 1 91 0
A. Ch eron, 1 965
. 11 h 1
1 56

For the time being Black i s able to

ward off the two threats, each pawn
being attacked by a Black rook. But
White frustrates Black's plans with
a single move .
3 b'! d 1 !
By attacking b oth rooks simultane­
ously, White severs their contacts
with the pawns. Black has no de­
Draw
Fence left : one of the pawns q ueens
i nexorably.
After 1 c7 E1 c6 2 e7 .E1 he6 we have
And here is a middle game position a familiar situation : White's queen­
in which the same two-fold double ing threats are parried by Black's
attack decides matters. rooks. The first impression is that

81
White might as well throw in the
towel. But White has a surprise up
his sleeve :
3 � d6 ! ! . . .
Viewed superficially, this move is
incomprehensible : White, who
hardly has anything, gives away
even that.
3 . . . l:l c x d6 + 4 00c4 l:l c6 + 5
00d5 ! !
1 57
This is the whole point of White's
plan. By sacrificing his knight,
White is able to launch a second
double attack. This is a remarkable the h-pawn in two moves) . The first
development. Black has two rooks
against two isolated pawns on files c second by the bishop. By means of
and e, but for each pawn he has to
give a rook.
If Black had taken the knight with
the other rook, then 3 . . .
r1 e x d6 + would have been fol­
lowed by 4 cfde4 n e6 + 5 cfdd5
with the same result. But what
would have happened if Black had
not captured the knight straight
away and played 3 . . . 00 x h7 in­
stead?
Then the only line of play leading
to a draw is 4 cfdd4 ! tl C X d6 + 5
82
00c5 ri c6 + 6 00d5 00g7 7 e 8 ( ® ) !
(if 7 c8 ( ® ) ? then 7 . . . l:l x c8 8
00 x e6 r1 e8 ! 9 00d7 00f7, and
Black wins) 7 . . . r1 cd6 + 8 00c5
l:l X e8 9 00 X d6.
Nenarokov- Grigoryev
Moscow 1 923
1 58
Black to play
Again a situation which we have en­
countered several times in this chap­
ter. Black has two threats, i. e. a
threat of the first order (queening
the pawn on d-file in one move) and
one of the second order (queening
threat is countered by the rook, the
the impressive move 1 . . � d6 ! !,
.
by which the coverless bishop is
placed at the point of intersection of
the lines of fire of the opponent's
rook and bishop, Black destroys
their co-ordination and wins the
game. Strictly speaking 1 . . . � d6
is actually a three-fold attack in
which Black attacks the bishop on
cl and threatens to queen on the cl­
file and to play h3-h2.
K. Behting only a bishop, but any other chess­
man, save the king of course, can be
1 59 placed at the point of intersection of
the hostile pieces.

A. Troitzky, 1 926
Conclusion ofa study

1 60

White to play and win

In this position the knight on d4 is

making two mating threats, i. e. on
c2 and b3, but for the time being the
bishop and the rook are able to
parry them. How can White disrupt Black to play
the co-ordination of Black's pieces
and plan a new double attack? A White has unleashed a double at­
diversionary sacrifice is the answer : tack-threat of 1 e 8 ( ® ) as well as 1
ed(®)-but Black finds an ingenious
1 l:j h3 ! tl X h3 2 f3 . . .
move to counter it by playing 1 .. .

Nothing is to be gained yet by 2 ld. d4 !. As you can see, Black uses

�d3 because of 2 . tl h 1 + , that the most promising line of play by
. .

is why the rook has first to be lured refuting White's double attack with
to the f-file. a double attack of his owm, 2
ed(®) being frustrated by the rook
2 ll X f3 3 0. d3 ! ! . . .
and 2 e8(®) is foiled by the dis­
. . .

And here is that blow again which covered check 2 . . . O.bs + .

prevents co-ordinated action of White's king dodges the threatened
Black's pieces and which immedi­ discovered check by moving to 2
ately decides the issue, because d>c l , but Black nevertheless plays
3 . . ll X d3 is followed by 4 � b3 2 . . . �bS. This is followed by the
.

mate, and 3 . . ld. x d3 by 4 � c2

. last double attack by 3 � d7 ! . The
mate. two queening threats become acute
again and Black is unable to ward
It is worth mentioning that in the
them off.
case of such a two-fold defence not

83
Koskinen-Kasanen
Helsinki 196 7
161
Black to play
It is almost inconceivable that such
a confused position occurred in an
actual game. White's king is in a
precarious situation, although for
the time being it is protected by its
pieces from direct threats, the rook
on h2 preventing a mate by the
knight on c2, whilst the bishop and
the queen prevent a strike against
b4. But by a manoeuvre already
fam iliar to us, Black manages to dis­
rupt the co-ordination of the enemy
pieces : 1 . ® x b4 + ! ! 2 � x b4
. .
Id. d2 ! !, after which White is con­
fronted by the insoluble problem of
parrying the two threatened mates
on b4 and c2. If he plays 3 Id. x d2
he is given mate by 3 . . . ab, and if
he tries 3 ® x d2 his pieces get in
each other's way. By playing 3 .
.· .
� c2 + Bl ack d iverts the queen
from covering square b4 and forces
a mate anyway.
The ju stifiable question arises
whether there is any defence
against the two-fold double attack.
It turns out that in special cases
84
even the two-fold double attack is
not fatal , but only if.the defending
player has a new strong threat. Just
s uch a situation is shown in the fol­
lowing example.
Diirsch- Ta rrasch
Nuremberg 1908
Game variant
1 62
Black to play
Whose position is preferable? It is
not easy to answer this question off
the cuff. On the one hand , White
has j ust mounted a double attack­
Black's p awn on h7 and rook c8 are
hanging-on the other hand Black's
queen along with the knight are
threatening mate on f3 and d 1 , of
which the first is parried by White's
queen and the second by the rook.
Black found an instructive possibil­
ity of refuting White's threats and
launching a two-fold cou nter-offen­
sive of his own .
1 . . Id. c7 !
.
This creates a critical situation for
White. No piece may take the rook
because of mate, at the same time
White's queen is attacked, and if it
sid esteps, White loses his rook on Particularly
d7. But White rises to the occasion. strong moves
2 bS � ! . . .
w-hen we spoke about the resources
Notice how each move radically af­
of defence against attacks we
fects the prospects of both players
' pointed out that one of the most ac­
posing difficult problems for each
tive methods of defence is the mu­
to solve. At the moment 3 � f8
tual attack or cou nterattack. For
threatens mate, for which reason
now the attacker himself immedi­
·

Black may not take any of the

ately has to turn his attention to
pieces being attacked. In this com­
warding off an attack. Now that we
plicated situ ation Black fou nd no­
have acquainted ourselves with the
thing better than 2 . . . ®d 1 + 3
double attack, we can say that a
El X d 1 El x b7, after which the
counterattack in the form of a
chances of winning were about
double attack is a still more effec­
equal. Although 2 . . . ® e2 would
tive method of defence. We have al­
not have brought Black more than a
ready come across such situations
draw because of 3 El x h7 + !
(see diagrams 70, 1 0 1 , 1 02 , 1 0 3). In
(b X h7 ( 3 . . El X h7 4 � f8 +
the present chapter we shall ex­
.

ddhs 5 ® x h7 + ) 4 ® x c7 + .
amine the question of defence by
It seems that 1 . . . El c7 is not the
double attack a little closer.
strongest move and that the modest
continuation 1 . . . El h8 deserves
f Hasek, 1923
attention. Black would indeed have
Variation ofa study
�een able to pose some very effec­
tive threats had he chosen to cover 1 63
his pawn on h 7 . He would have
been able to reply to 2 b S , for in­
stance, with 2 . . . ® fs ! with the in­
tention of playing 3 . . . ® b 1 + . Af­
ter 2 � b 2 ® h 3 ! 3 �g7 + ddhs 4
l:l dS + � fS , Black would have re­
tained his offensive. But given the
be st possible reply, i. e. 2 � c l ® h 3
3 � x e 3 , White's chances of de­
fending himself successfully would
have been good. Black to play

Whi �e repl ies to 1 . . . a 1 ( ® ) + by

playmg 2 � e S ! . This is not merely a
defensive move.
White not only attacks the newly
acquired queen but also threatens

85
mate by 3 �f7. Thus, we have here
a case of defence by a double at­
tack. After the forced sequence
2 . . . ®a2 3 c7 ®c4 4 c8(®) +
® x c8 5 �f7 White achieves his
objective after all and gives mate.
The next study again shows a
double attack.
H. Mattison, 1 930
164
White to play and win
White's bishops are threatened. It
seems that White will not be able to
extricate himself from this situation
without losing material. But White
has an interesting possibility.
1 rl dS ! .bi x dS 2 � e3 ! ! . . .
1 65
86
Let us pause to examine this posi­
tion. It is not difficult to see that this
is a double attack. White, who has
brought his bishop h6 to safety, at­
tacks the rook and threatens mate
by playing his bishop to f7 . Black
can defend himself by 2 . . . rl gS or
2 . . .bi d6. Black's rook tries to get
.
out of reach of the bishop. But
White plays 3 �f7 + bt g6 4 �h2 !
and the rook is lost. Black would
not have fared any better by playing
3 . . . �h6 after 2 . . .bi gS 3
.
� f7 + because of 4 � e 8 !
It remains to be said that 2 �e3 is
the only move that leads to success.
If White plays 2 �d2, Black saves
himself by an effective move : 2 . . .
rl f5 !, because 3 � X f5 results in
stalemate. In the case of 2 � f4, on
the other hand, Black can play
4 . . . �g4 ! after 2 . . . rl gS 3
� f7 + rx g6 4 �h2 .
And here are a few examples from
tournaments.
Heemsoth-Heissenbutter
Federal Republic ofGermany 1 958
1 66
White t o play
The situation is very tense. Black
has attacked the knight on c2 and
th reatens mate. It seems that White
has to go on the defensive. But on
taking a closer look one notices that
White could checkmate his oppo­
nent in two moves by 1 Il x h7 +
followed by 2 t!tJ g7 if his own rook
on gS were not in his way. The most
expedient way to convert this situa­
tion is by covering the mate on c2
and launching a double attack at
the same time : 1 Il cS ! !. He coun­
ters Black's 1 . . t!tJ x cS by bring­
. ing himself from a check by the
ing home his second threat : 2
ll x h7 + 00 x h7 3 ti!J g7 mate.
Kretschmar-Laue
Eisenach 1 95 1
1 67
White to play
White can capture his opponent's
queen with his knight, but this only
leads to a draw with a perpetual
check. How can White protect him­
self against this threat?
1 h4 ? ?, for instance would be very
bad because of 1 . . . Il h t + ! 2
00 X h 1 t!!J b 1 + 3 00h2 ti!J g l + 4
ct>h3 ti!J h 1 mate. Weak is also 1
® x f4 Il h 1 + ! 2 00 x h 1 (some-
what better is 2 ct>g3 ti!Jg6 + 3 � g4
hg) 2 . . . ® b 1 + 3 00h2 ti!Jgl + 4
00g3 � f2 mate.
This means that the most important
thing for White is to neutralize the
dangerous bishop on b6. You will
notice that the bishop plays a part in
the threatened perpetual check, but
that it also prevents the rook from
penetrating to square d 8 . As soon as
we realize that the bishop is over­
loaded with duties, we find the ele­
gant move 1 t!tJ c5 ! !. Whilst protect­
bishop, White simultaneously
threatens mate on c8 and f8 with his
queen. What is more, the queen
may not be taken because of 2
Il d8 + with an inevitable mate. In
effect White has three threats (the
two immediate threats 2 t!!J f8 + and
2 ti!Jc8 + and-after the queen is
taken-the indirect threat of 2
Il d 8 + ) .
Mar6czy-Romi
San Remo 1 93 0
1 68
White t o play
It is not so easy to unravel this situa­
tion. White's queen is threatened,
87
but it may not be touched due to the
check on g8 threatened by the rook,
followed by e7 -e8 ( ® ) + . But
Black threatens 1 . . . .bi x g2 + 2
,bi X g2 ® X g2 + 3 d/ X g2
.bi x h6, which promises Black a
piece advantage. How should
White defend himself? It would be
bad, for instance, to play the queen
to g S , because Black would reply
with a decisive series of blows 1 . . .
.bi x h3 + ! 2 dlg 1 ®b6 + . Nor
does 1 .bl g8 + .bi x g 8 2 ®hs +
dld7 3 e8 (®) + .bi x e8 4 ® x e8 +
dlc7 look promising.
White's position looks critical. And
yet there is a way out-and not only
that but even a way to victory. By
playing 1 ®hs ! ! White is able to
stage a magnificent double attack,
threatening a devastating double
check by 2 .bi g8 + , against which
Black has no adequate defence. The
finale took the following course :
1 . . . .bi x g2 + (no longer d anger­
o us) 2 ,bi X g2 + ,bi X h5 3 ,bi X b2
,bi X h 3 + 4 d/ g 1 ! .bl h7 5 .bl h2 !
.bi g7 + 6 dlf2 .bi g8 7 .bi h6 dlf7
(loses at once, and even other
moves do not come to the rescue) 8
e 8 ( ® ) + .bi x e8 9 .bi h7 + .
88
Taimanov-Zhukhovitski
Leningrad 1 94 9
Game variant
1 69
White to play
White has already carried out a
double attack. His queen has at­
tacked Black's rook and his pawn is
threatening to queen. But for the
time being White cannot profit
from the double attack, because his
opponent is poised to make a strong
threat by playing 1 . . . � g3 + to
be followed by 2 . . . ® h6 m ate.
Thus, White has to find a move that
thwarts the mate threat and keeps
the double attack going. These con­
d itions are satisfied by 1 � fS ! al­
though White has to reckon with
Black's replying 1 . . . .bi d2 ! by
which Black attacks the queen,
b rings his rook to safety and pre­
vents the pawn from queening.
2 ® X d2 ! ! . . .
This is the punch-line. White sacri­
fices his queen in order to press
home his threat of queening his
pawn.
2 . . . ® x d2 3 .bl ad 1 ® a 5 4
.bi ge l ! � d4.
No better would be 4 . . . ®d8 5 two threats : 2 ® e 1 mate and 2 d:>f2
,El x e2 hS 6 Et e7, and Black is corn­ mate. The only defence against this
pletely helpless. is 1 . . . ® f8 + , but then the third
threat (of the second order) materi­
5 Et e8 d:>g8 6 kl x d4. White won
alizes : 2 d:>g3 + d:>g 1 3 ®g2 mate .
eas ily.
Now let us j ust see why 1 ® d S ?
In all examples examined White de­ would not have prod uced the de­
fen ded himself by conjuring up sired results. In this case Black
double attacks by means of masterly would still have had a defence. By
well masked moves. Unlike moves threatening stalemate he would
that constitute pure double attacks, have transposed his queen to the g­
moves aimed at staging double at­ file by 1 . . . ®g8 ! and saved him­
tacks as a means of defence could self.
be regarded as particularly strong In this example it was not difficult
moves. Incidentally, one should not to find the decisive move 1 ®e4.
only look for and find such moves, The threats were by and large obvi­
but one should be in a position to ous. But this is not always the case.
recognize them in advance. Very often one has to have a thor­
Pure attack moves containing three ough understanding of the position
threats simu ltaneously, two threats i n order to be able to find the
and one attack, one threat and two decisive move embodying various
attacks or even three attacks can threats .
also be included in the category of
Let us examine the following study.
particularly strong moves.
Let us first consider some positions V and M. Platov, 1909
in the field of end-games.
1 71

E. Pogosyants, 1976

1 70

White to play

Can you imagine this position oc­

White to play and win curring in a game between average
players ? After 1 � f6 d4 White is
White plays 1 ® e4 ! thereby posing convinced that he cannot prevent

89
the appearance of a black queen. flee by 3 . . . <bd2, White follows
For this reason he replied 2 � f3 up with a double attack by playing 4
and after 2 . . . a 1 ('i!tt ) continued � b3 + . There is only one line of
with � x d4 + � x d4 4 � x d4. defence against the threatened
After 4 . . dl x d4 5 \t>g4 dl x d 3 6
. mate, i. e. 3 . ®as, but White
. .

digS d:>e4 7 d:>h6 d:>fs 8 et> x h 7
d:>f6 the final game ends in a draw.
All this is uncomplicated, forced
and colourless. But this study calls
for White's victory. This means that
we have to look for other more so­
phisticated solutions. It is only fair
to warn the reader that the moves
leading to victory are extremely dif­
ficult to find. To be sure, the first
move with Black's response- 1 � f6
d4-is correct. But instead of 2 � f3
White plays 2 � e2, because Black's
king cannot take the knight yet. But
what remedy is there for 2 . . .
a t (dtt ) ?
Here White comes u p with a parti­
counters this with a double attack
threat of the higher order: 4
� X d4 + ! \t> X d4 5 � b3 + and
the queen is lost after all.
As you see, White's move 3 � c l ,
which was difficult to find and even
more difficult to anticipate, embod­
ied a mating threat and two double
attack threats, one of the first, the
other of the second order.
Also in the followin g well-known
study the finish is decided by a par­
ticularly strong move.
A. Zeletski, 1 933
1 73

cularly strong move. In what ap­

pears to be utter disregard of com­
mon sense he plays 3 � c l ! ! . . .

1 72

Whit e to play an d win

After 1 � a6 + �b8 2 'l!tt g 3 + ®aS

3 �b7 + ! � x b7 White plays 4
� d7 ! This move is an attack and
creates a mating threat by 5 � b6.
But we soon realize that the knight Black parries both threats with
is unassailable due to 4 �gS + and 4 . . . dtt d s . But White follows this
that by playing his knight to c l up with a masked threat of a higher
White threatens m ate b y playing his order: He plays 5 dtt b 8 + ! and gives
bishop to gS. At once Black's king mate after 5 . . . � x b8 by playing
finds itself in a cage, for if it tries to 6 � b6.

90
In the foregoing studies the third
threat was connected with a sacri­
fice. But this does not necessarily
have to be the case.
E. Pogosyants, 1 976
1 74
White to play and win
In this situation it is 1 � f5 + that
decides the game. This move is first
of all an attack and it also poses the
threat of a double attack, i. e. 1 . . .
«;1le8 is followed by 2 � d6 + , and
finally after 1 . . . «;1lg8 White
threatens to give mate by 2 � h6 +
and 3 � b2 + .
The following study illustrates the
combination of particularly strong
m oves with double attacks.
L. Kubbe/, 1 93 9
1 75
White to play and win
It is easy to guess how the situation
shown in the diagram will end.
White has not got a single pawn,
that is why if he wants to win he has
either to mate or capture Black's
rook. Since Black threatens to play
1 . . «;1lc7, White has to hurry in
.
order to build up a m ating network
around his opponent's king in time.
Black's possibilities are restricted
most effectively by the discovered
check 1 � f6. Let us first examine
this move. It is a sudden double at­
tack. Apart from the check, White
threatens a fork 2 � e 8 + if 1 . . .
«;1lc7. But Black can still play 1 . . .
c!lc8 . After 2 � cl + Blatk natu­
rally does not proceed to b8 be­
cause of mate in two moves, but re­
turns to d8, after which White has
achieved a mere perpetual check.
There is a much stronger double at­
tack :
1 � f8 + ! . . .
I n this way White also prevents the
king from moving to c7, but the
king has two possibilities : retreating
either to c8 or to e 8 . Let us examine
them in that order.
1 . . . «;1lc8
White has a particularly strong
move in reply : 2 Il c 1 + . It attacks
the enemy king and poses two si­
multaneous threats : if 2 . . . \!/d8
then 3 � e6 + , if 2 . . . ®b8 then 3
� d7 + followed by 4 rl a 1 mate.
1 . . . ®e8
In this case the first player also has a
particularly strong move, i. e. 2
� e6 ! While attacking the rook,
91
White poses two threats simultane- Zhivayeva-Gaprindashvili
ously: if 2 . . . E H7 then 3 Et d 8 Tbilisi 1 9 76
mate o r i f 2 . . Et h7 then 3 Et d 8 +
.

\t)f7 4 � gS + . Black has no choice 1 77

other than to play his rook to g 8 .

But then White springs a masked
threat, or more precisely, a threat of
the fourth order. He forces mate in
fou r : 3 � c7 + d:lf7 4 E H1 + 00g7
5 � e6 + 00h6 6 Ei h l mate !

A isenstadt-Margolis
Leningrad 1 95 7
1 76
Black to play

Black' s position seems critical.

White to play
By the particularly strong move 1
E! a8 ! the first player instantly
achieves a winning position in an
impressive show of skill. He attacks
the rook e8 and poses two collateral
threats : the mating threat by 2
® f8 + Et x f8 3 E! x f8 and the
threat of a double attack by 2
®e6 + ! If Black takes the rook
then the game is decided by 2
® d S + ® f7 3 ® x a 8 + and mate
on the next move . The only defence
- 1 . . . ® f7 - is thwarted by 2
® d 8 as well as 2 ® c6 with the loss
of the rook.
The next example shows the inter­
esting case of a triple attack.
Mate is threatened on c7 , and 1
El d7 is followed by 2 ® cs + \t)ds
3 ® x a7 with decisive threats . And
yet Black has deliberately got her­
self entangled in this situation, hav­
ing foreseen the opportunity of
launching a triple attack by sacrifi­
cing a piece, thereby taking advan­
tage of the insecure posting of her
opponent's king. She played
1 ...� x d 4 + ! ! 2 cd ® h4 +
And there it is-the triple attack !
Black is attacking the king, the
bishop and the pawn on d4. White
saved her bishop by 3 � g 3 , but this
was followed up by 3 . . . ® x d4 +
4 00g2 ? . . .
The radical turn of events confused
the opponent of the world cham­
pion woman chess player, so that
she failed to adopt the best line of
defence. Supported by the threat­
ened mate, White should have
played 4 OOfl . Bl ack would have
had to answer with 4 . . . n d7' be-

92
cause 4 . . . � e3 + 5 \t>e2 is worth­Needless to say, it may be that we
less . After 4 . . . ® x a t + 5 ® f6 too shall be examining examples of
� e3 + 6 \t>e2 � x c2 7 ® x d4 s uch opening blunders, which are
� x d4 + Black would be one normally cited in literature as exam­
pawn up, but after the reply 8 \t>d3 ples of what one should not do in
White cou ld have continued giving openings. But our aim is somewhat
stiff resistance. different. We examine the two-fold
But the game ended, after 4 d!g2 ?, attack as a tactical method , whereas
the organization of the co-ordi­
on the fol lowing note : 4 . . . � e3 +
5 \t>h3 (White apparently thinking nated attack on f7 forms the basis of
that her king was out of d anger, but White's strategy in many Open
. . . ) 5 . . . ®g4 + ! 6 fg hg doubleGames. Our intention is to familiar­
check and m ate ! ize you with examples of the open­
ing game in which the attack on f7
is of a tactical nature and which
h ave nothing to do with the strate­
'fhe doub le attack
gic concertion of the opening, i . e.
in the openin g in which the attack comes s uddenly
as a result of rash moves on the part
You will no doubt know that the ba­ of the defender.
sic thought behind m any open As you know, White's strategy in
games is organizing an offensive on the Sicilian Defence hinges on cer­
the weak point f7 in Black's camp. tain moves in the centre and on the
As soon as the f-file is opened-as it king's side. Only occasionally does
often is with the King's Gambit-or a battle flare up around f7 as well,
the bishop appears on c4 in the first but this is a tactical development.
few moves-as with the Italian For this reason we have culled a few
Opening, the Evans Gambit or the examples of sudden attack on f7
Scottish Gambit-Black must be on taken from this opening.
his guard and keep a cautious eye
on his opponent's moves that m ight Sicilian Defence
be aimed at the weakest spot in his Thomas-Shapiro
position . The same applies to the A ntwerp 1 93 2
fwo-Knight Opening, in which
1 e2-e4 c7-c5
·

White openly tries to attack this

2 � g l -f3 � g8 -f6
square with two pieces . The open­
3 � b l -c3 d7-d5
ings just mentioned have been ana­
4 e4 X d5 � f6 X dS
lysed in detail and all or nearly all
5 � f t -b5 + � c 8 -d7
examples of inadequate Black de­
fence allowing White to take the Let u s try to find our bearings in
point f7 by storm have found their the events that have taken place on
place in textbooks on openings a the board . Black's move 3 . . . d5 is
long time ago. not exactly the best. As a result of

93
this premature thrust, the knight on
dS is in a precarious position, which
White wants to use to his advantage
by giving check with his bishop on
bS. The correct answer to this check
is, without fearing the doubling of
pawns, the deployment of the
knight to c6. But after the move
with the bishop the knight on dS
hangs, which creates favourable
conditions for tactical operations.
6 � f3-e5 ! � dS X c3 ? ?
1 78
Black would like to exchange his
knight, but this natural move proves
a mistake and results in a loss. Rela­ Black's plan is to'put pressure on
tively better was 6 . . � x bS 7
. White's central pawns.
®f3 f6 8 � x b5 fe 9 ® x d5, al­
though in this case, too, White had
the advantage.
7 ®d 1 -f3 !
And here we have the two-fold at­
tack on f7, which guarantees White
success. Black has no adequate de­
fence left.
7 ... f7-f6
8 ®f3-h5 + g7-g6
9 � eS X g6 ! d)e8 -f7
1 0 � g6-e5 + !
Black resigns. After 1 0 . . . c!Je6 the is weakened. In other words, the
94
easiest way to victo ry is 1 1 s.":. c4 +
� dS 1 2 � X d5 + CO X d5 1 3
� f7 + with the capture of the
queen.
And here is a relatively recent ex­
ample from master-class competi­
tions.
Sicilian Defence
A ronin-Kantorovich
Moscow 1 960
1 e2-e4 c7-c5
2 � g l -f3 g7-g6
3 c2-c3 b7-b6
Black tries to experiment in the
opening. It is, of course, very easy
to thwart such a plan. But in my
opinion such a strategy is admissible
provided that it is carried out with
precision and that a sharp eye is
kept on the opponent's plans and
threats.
4 d2-d4 �c8-b7
5 �fl -c4 !
White protects the pawn by threat­
ening the point f7. Black should
have gone on alert already when
W hite placed the bishop on c4 .
5 ... d7-d5
Black opens the game. This move in
no way fits in with the sluggish de­
fence system he has chosen. On top
of that it has two serious tactical
drawbacks : f7 becomes even more
vulnerable and the diagonal a4-e8
move 5 . . . dS is a mistake and puts 7 �fl -c4 h7-h6
Black in a difficult position. 8 � g5 X f6 g7 X f6
6 e4 x d5 � b7 x d5
180

This taking of the pawn loses imme­

diately. Better would have been
6 . . . � f6 or 6 . . . �g7 and to ac­
cept the loss of the pawn.
7 � d 1 -a4 + � d5-c6
If Black had played 7 . . . � c6 he
would not have fared better because
of s ·� eS rl c8 9 � bs t!i:tc7 1 0 � f4
-'1 h6 1 1 �g3 with loss of material.
Up to this point there was nothing
1 79
worth commenting on in the open­
ing. An extremely dynamic, modern
variant of the Sicilian Defence was
played in which White quite often
sacrifices a piece on e6, d5 or b5. In
view of the fact that White's pieces
are ominously aimed at Black's po­
sition, Black would be well advised
to be especially careful so as not to
overlook some dangerous threat
from his opponent. White has
8 � f3-e5 ! chosen to exchange on f6 so as not
T'he decisive manoeuvre ! If Black to lose any time. The only correct
takes the queen, White follows up answer was 8 . . . .� x f6. But now
with 9 � x f7 mate, or if 8 . . . �cl White sacrifices the bishop on f7
then the easiest reply is 9 � x c6 and immediately proceeds to moun­
� x c6 1 0 d5 with piece advantage. ting a decisive attack with two
Seeing this, Black resigned. pieces, taking advantage of the ga­
ping weakness of square e6.
Sicilian Defence
"J � c4 X f7 + ! <ile8 x f7
Voorema-Zakharyart
1 0 t!i:t d 1 -h5 +
Minsk 1 964
Black resigned. If 1 0 . . . 00g8 then
1 e2-e4 c7-c5
1 1 � e6 with an attack on the queen
2 � g 1 -f3 d7-d6
and mating threat by 1 2 ®g6 + .
3 d2-d4 c5 X d4
4 � f3 x d4 � g8-f6 And here is an example in which a
5 � b 1 -c3 a6-a6 premature attack on f7 led to pre­
6 � c l -g5 � b8-d7 cisely the opposite result.

95
French Defence White has consistently stuck to his
Lehmann- Teschner plan and made a fork. But he neg­
Bad Pyrmont 195 0 lected a very important detail :
1 e2-e4 e7-e6 10 . . . � e7-b4 mate.
2 d2-d4 d7-d5
Double check and mate !
3 � b l -c3 d5 X e4
4 �c3 x e4 g b 8 -d7 In the next ex-ample f7 is only one of
5 �gl -f3 g g8 -f6 two objectives at which the attack is
6 �f3 - g5 aimed.
As in a similar version of the Caro­
Spanish Game
Kann Defence (see below) White is
Mohrlok-Kramer
planning a q uick attack on f7 . But
Varna 1962
here the sacrifice is useless, which is
why Black provokes his opponent 1 e2-e4 e7-e5
with the next move to carry out his 2 g g l -f3 g b 8 -c6
plan. 3 � f 1 -b5 a7-a6
4 � b5 -a4 d7-d6
6 .. � f8 -e7
5 d2-d4 b7-b5
7 �g5 X f7 ? ? d!e8 x f7
6 � a4-b3 g c6 x d4
8 �e4-g5 + d!f7-g8
7 � H3 X d4 e5 x d4
9 �g5 x e6
8 c2-c3 d4-d3
White's plan was to attack the
This move is a sheer waste of time.
queen and to gain control over
The balance is redressed by 8 . . .
squ are c7. On top of that he wanted
�b7.
to capture the rook on a 8 . But there
was a little flaw in his reasoning . . . 9 a2-a4 � c 8 -d7
1 0 a4 x b5 a6 x b5 ? ?
9 ... ® d 8 -e 8
1 0 � e6 x c7
1 82

1 81

96
 \pparently a move pbyed automat­
this position Wh ite sue­
i r a l l y . In
reed s in b u n c h i ng a double atuck,
" h i c h i m m e d i ately d ec i des m atters.

l l �d 1 - h S � d3 -d2 +
1 2 � c l x d2

Bl a c k re s i g n e d . 1 2 . . . g6 is fol­
l nwed by 1 3 e d s , threatening mate
l l l l f7 a n d attacking the rook on a 8 .

In a more· recent example taken

l'ro m a game p l ayed by two grand­ Usu ally 1 2 � f4 is played at this
l l l a st e rs, a n u n expected strike was poi nt. White tries to adopt a new
n 1 ade on f2 . plan, wh ich includes prep-arations
for castling long, but h i s p l an s u f­
( :aro-Kann Defence fe rs a surprising fiasco. The cause of
Suetin - Kholmov fai l u re is the insecure posting of
1/udapest J 9 76 White's kn ight on e 5 .

l e2-e4 c7-c6 12 . . . � c5 x f2 + �
2 d2-d4 d7-d5 A bolt from the b l u e � After 1 3
3 � b 1 -c 3 d S X e4 d) X f2 � X eS 1 4 ® X e 5 � g4 +
4 � c3 x e 4 � b 8-d7 Black recovers the piece by a fork
5 � f l -c4 � g 8-f6 and rem a i ns one pawn up.
6 � e4-gS e7-e6
7 ® d l -e2 You will recall that cond itions for a
double attack are p a rt ic u l a rl y fa­
I n this version of the C aro- Kann v ou ra b le if the two s i des decide to
I >cfence , Wh ite deploys his pieces
h ave an excha nge by attack and
h�· :Hta c k i n g f7 . cou nte rattack.
7 � d 7-b6
Irregular King 's Pawn Opening
In this manner Bbck refutes the
Rusakov- Verlinski
t h re a t 8 � x f7 d/ x f7 9 f!t1 x e6 + .
Moscow 1 948
H � c4-d 3 h 7 -h6
1 e2-e4 e7-e5
The pawn d4 may not be taken be­ 2 c2-c3 fl b 8 -c6
c a use of 9 � 1 f3 followed by 1 0 3 d2-d4 fl g8 - f6
(:J eS with a two-fold attack on f7 . 4 � c 1 -gS h7-h6
S � gS-h4 g7-gS
!.) � g5-f3 c6-c5
6 � h4-g3 e5 x d4
I O d 4 x cS � f8 X cS
7 e4-e5 d4 X c3
l l fl f3-eS �d 8 -c7
1 2 � c l- d 2 The two opponents have d i sre-

97
garded all cliches in their opening Spanish Opening
moves. By sacrificing his pawn 8 Hort-Zhelyandinov
� x c3 White could have profited Havana 196 7
from an active position . But he
1 e2 -e4 e7-e5
chose to have an exchange instead,
2 � g 1 -f3 � b 8-c6
misj udging the possible conse­
3 � f 1 -b 5 a7-a6
quences to boot.
4 � b 5 x c6 d7 x c6
8 e5 x f6 ? c 3 X b2 5 0-0 f7-f6
9 ® d t -e2 + 6 d2-d4 eS X d4
7 � f3 x d 4 c6-c5
1 84
8 � d4-b3 ®d8 X d t
9 b H1 X d l � f8-d6
1 0 � b3-a5 !
The line adopted by White is ac­
cording to a recipe by Fischer, who
used it su ccessfu lly in a number o f
games. T h e intention behind the
knight's move to aS is to make
Black's development of his queen's
side more difficult.
10 . . . � c8-g4 ?
White was banking on eliminating
the dangerous pawn on b2. But Black tries to solve his development
events took an entirely different problems by tactical means.
course .
1 1 f2-f3 0-0-0
9 ... ® d8-e7 ! !
1 0 f6 x e7 � f8 -g7 ! 1 85

White resigned.
A tragicomic situation. White is a
q ueen and a knight up, but he can
only prevent the pawn from queen­
ing by giving away his own queen,
after which Black's material superi­
ority wins.
The next two examples show play­
ers who have been "hoist with their
own petard". Black had planned to answer 12 fg
They show how the player setting a with 1 2 . . . � x h2 + , but there
trap for his opponent ends up as the was a most unpleasant surprise in
victim . store for him.

98
1 2 e4-e5 ! 9 . .®
. x d4 is refuted by 1 0
� bS + .
A double attack of decisive impact.
' I 'he loss of one of the two bishops This survey, which covered only a
hcing inevitable, Black resigned. small group of openings, proves
that one has to recko n with the pos­
French Defence sibility of a double attack from the
Kotkov-Akopyan first few moves on, i . e. even in the
Krasnodar 1 966 development phase. S uch factors as
i nsecurely posted pieces, reciprocal
I e2-e4 e7-e6
attacks and various pins increase
2 d2-d4 . d7-d5
the danger of double attacks.
3 � b l -c3 d5 x e4
Forks, discovered checks and
4 � c3 x e4 � c8-d7
double checks and even more com­
The problem of developing the p licated cases of the double attack
queen's bishop can also be solved i n - all these things can occur in the
1 h i s peculiar manner. opening stage. However, the double
attack normally occurs as a result of
5 � g l -f3 � d7-c6
a bad mistake, of underestimating
6 � f l -d3 � g8 -f6
the opponent's plan, miscalcula­
7 � e4 x f6 + ® d 8 X f6 ?
tion, carelessness and sometimes
B l ack tries to lure his opponent into s imply as a result of thoughtless,
a trap. He plans to answer 8 � g 5 mechanical playing.
w ith the two-fold attack 8 . . .
,\.� x f3 and if 9 ® x f3 then 9 . . .

'd:IJ x f3 with equ alization.

The double attack
H � c l -g5 � c6 X f3 in the end-game

1 86
Many chess-players entertain the
mistaken notion that tactics is
something that concerns the mid­
dle-game, while the en d -game is
governed by strategy and that tacti­
cal features occur only in excep­
tional problem situations. Despite
the fact that many positions dis­
cussed in the preceding chapters to
illustrate various aspects of the
double attack were end-game situa­
l) ®d t -d2 ! tions, we have decided to devote a
I H ack had overlooked this possibil­ special chapter to the theme of
His queen is trapped because
i t y. double attack in the end-game. We

99
set ourselves the task of showing
that tactics , especially the double at­
tack, is part and parcel of end-game
strategy and that in m any cases it is
one of the basic methods of che ss
play.
Y ou will, of course, know that in
the end-game the king is relatively
safe, as a result of which tactics in
the sense of a double attack on the
king rarely succeeds in achieving its
purpose. Instead, a d ifferent objec­
tive is pursued in the end-game: the
queening of pawns. Tactics in gen­
eral and the double attack in parti­
cular plays an outstanding role in
dealing with problems of queening.
Let us begin our analysis with ele­
mentary pawn end-game positi ons ,
in which the king and an inside
pawn are fighting against the lo ne
opposing king. There is a simple
rule covering such positions. If the
pawn has crossed the middle line,
then it is enough for the stronger
player's king to stay in front of the
pawn in order to ensure conversion.
If the pawn has not yet crossed the
middle line, then the strategy is to
advance the king first and then the
pawn. This means that the stro nger
player's king must first penetrate
into the hostile territory, while the
weaker king should try to prevent
this.
Let us follow this duel of the kings.
1 00
White's king is threatening to out­
fl ank his opposite number. To pre­
vent this , Black's king must go into
opposition. But this is not enough.
Let us assume it is Black's move. In
this case he must step aside and
White carries out one of his threats :
1 . . ddds 2 ddb4 ddc6 3 ddc4 ddb6
.
4 ddd s ddc7 5 ddcs ddd7 6 ddb6 etc.
But if it is White's turn to play, then
Black's king can foil the White
king's advance and the game is
drawn, e. g. 1 ddb3 ddbs or 1 ddd3
\tlds , etc.
Thus we see that in this end-game
White carries out a double attack
(in the form of two threats) , and the
task of the second player consists in
parrying it. In this way the double
attack constitutes the essential tacti­
cal element of this elementary end­
game.
The double attack as a combination
of various threats forms the basis of
the game in many other, more diffi­
cult end-games with pawns .
Let us consider a famous study as
an example.
R. Reti, 1 92 2 about to create the threat of sup­
porting its pawn . Hence it is easy to
u nderstand that by moving to
square g7 the king creates two
threats of the second order. It is
showing signs of creating two dif­
ferent threats, i. e. to support his
p awn and to stop the hostile pawn.
If Black answers with 1 . . h4 then
.

the respective quadrangles of

Black's pawns shift to
e l -e4-h4-h 1 , d l -dS-hS-h l
and c l -c6-h6-h 1 . After 2 cbf6
I > raw
the king is already threatening to
give its pawn immediate support
I .et us try solving this p roblem only and at the same time making ar­
with the aid of the terms "attack" rangements of creating the threat of
and "threat" . stopping the opponent's pawn. If
White's pawn is two moves from the B lack wants to parry the immediate
qu eening'rank, but Black's king is threat he has to continue with 2 . . .
i n its quadrangle and is threatening d/b6. Again the squares of the pawn
t o attack it. At first glance White's on c6 shift, because the king can
ki ng is hopelessly far behind Black's only support it from squares d6 and
pawn and unable to support his own d 7 . White plays 3 dieS , again with
pawn . two objectives in view . He creates
l�ut let us analyse the Black king's the two direct threats of supporting
i nterrelations with the two pawns. his pawn and of stopping his oppo­
· I 'he king could obviously intercept nent's pawn. Black has no defence
t he pawn if it could get to the qua­ to counter them. After 3 . . . h3 4
d rangle d 1 -d S -hS-h 1 . This d/d6 h2 5 c7 the game ends with a
means that the king could threaten draw.
10 stop the pawn if it were in the As you see there is nothing super­
quadrangle c l -c6-h6-h 1 ; which n atural in this study. It complies
means that as soon as it enters the with the inherent logic of chess and
q u adrangle b 1 -b7-h7- h 1 the its solution rests o n the repeated ap­
king can threaten to stop the pawn. plication of the double attack.
N ow let us analyse the White king's
i nterrelations w:ith its pawn. The
king could support its pawn from It can be demonstrated that the
.� qu ares e8 -e7 -e6-e5, and from method of triangular movement so
�quares f8-f7-f6-f5 it would common in the end-game with
1 h reaten to do so. And finally from pawns is a special case of the double
squares g8 -g7 -g6-g5 it would be attack.

101
 189
White to play
White has two threats, i. e. that of
queening one pawn and that of
penetrating to b6. But for the time
being Black is able to refute them
successfu lly, because 1 d:>cS is fol­
lowed by 1 . . . 00c7 but 1 00d5 by
1 . . . dlc8 2 00d6 00d8 . In these
circumstances White makes a wait­
i ?g m �ve: 1 00c4 . He retains the op­
. 4 00b7 00d6 5 dl x a7 00c7.
tion oi makmg e ither threat. But we
realize at this point that pinned
against the edge of the board, the
Black king itself is forced to upset
the equilibrium. After 1 . . . 00c8 2
00d5 or 1 . . . 00c7 2 00c5 White
succeeds in pressing home a threat.
By its nature a double attack is also
a cha_r acteristic method of pushing
the kmg to the edge of the board in
the end-game with pawns.
The following example shows how
to go about it.
102
I. Maiselis, 1921
190
White to play and win
White must capture the pawn on a7.
If he tries to accomplish this in a
straightforward manner by pro­
ceeding along the straight line
f7 -a7, the game ends in a draw : .1
00e7 dlc3 2 00d7 00d4 3 00c7 00c5
White's objectiv,e is to stage a
double attack, combining the threat
of capturing the pawn with that of
not letting Black's king come near.
This is achieved by the following
series of manoeuvres by the king 1
00e6 00c3 2 OOdS ! 00b4 3 dlc6 00a5
4 00b7 OObS 5 00 x a7 00c6 6 00b8 '
and Black's king arrives too late.
Hence, the conclusion to be drawn
is that the double attack as a combi­
nation of various threats is an essen­
tial tactical method, which is very
often used in end-games with
pawns.
The double attack also plays an im­
portant part in end-games with
knights, primarily because the de­
coy as a tactical method is a funda­
mental technique used in this end-
1�ame in order to achieve pawn su­
per iority.
191
White to play
After 1 \tle6 � b4 2 d7 \tlc7 it
woul d be a mistake to play 3 d/e7
hecause of 3 . . . � d5 + . The cor­
rect move is 3 � a6 + ! , which de­
coys the knight, so that after 3 . .
.
� x a6 4 \tle7 the pawn can queen.
Another reason why the double at­
tlck is so important in this end­
glme is that the knight fork be­
comes a particu larly effective wea­
pon at this stage of the game.
A. Philidor, 1777
1 92
White to play
Without the king's help but with the
help of the fork, the knight defends
itself successfully against the king
and the pawn :
1 � c l + \tlb2 2 � d3 + \tlc2 3
� b4 + \tlb3 4 � d3 !
Black's king has managed to push
the knight aside. But despite the fact
that the knight dodges to the side, it
checks the black pawn's advance by
threatening to mount a double at­
tack by 5 � c l + . Hence a draw.
Occasionally the knight manages to
catch up with the pawn with the aid
of the double attack.
193
Black to play
Black can play 1 . . � b3 2 b7
.
� c5 ! Black carries out a double at­
tack (i . e. an attack and the threat of
a two-fold attack : 3 b 8 (®) � a6 + )
and saves himself.
The following study is an excellent
example of a double attack in an
end-game with a knight.
1 03
V. Chekhover, 1 93 8 From an analysis byA. Alekhine, 1934
1 95
1 94

Draw
Squares marked with a cross are
squares which result in the loss of a
Black to play
Alekhine sees this as a win and uses
the following variant to prove it:
1 . . 00e2 2 f4 00e3 3 fS dle4 4 f6
.

pawn if Black's king steps on them. dieS 5 b S . You will notice that the
White's forces are separated, but thrust 5 bS is basically a double at­
Black is unable to take advantage of tack because both pawns are threat­
this because the area around the ening to queen. But Black has an
Black king is "mined". It may not appropriate answe r : He plays 5 . . .
step on the c-file or on squares b 3 or \t>e6 ! thereby invalidating White's
bS. If Black should, e. g. play 1 . .
. double attackbyturning the tables on
dlb3 then 2 fle4 fs 3 � g3 f4 4 him. The bishop takes over the task
fle2 f3 5 fld4 + and the f-pawn is of checking the f-pawn and the king
lost. For this reason Black cannot j ust manages to cope with the task
win despite his superior force. of preventing the advance of the b­
pawn. A draw is inevitable after 6 b6
In end-games with bishops, espe­
d/ d7 7 dlbs dlc8 8 d/a6 dlb 8 .
cially when two pawns are pitted
against a bishop, the whole game
L. Centurini, 1 847
very often hinges on the double at­
1 96
tack, namely on the combination of
two queening threats . And con­
versely, the defence against these
two threats is based on the correct
assignment of tasks to the king and
the bishop.

White to play and win

1 04
In this position, too, the first F Sackman, 1 909
player's strategy is based on two
1 hreats, i. e. on the double attack. 197

On the one hand White threatens to

interpose the bishop on c7 and, on
1 he other hand, to move it over a7

and b8 and from there to drive

Black's bishop off the diagonal
h2-b8 . Events can then take the
following course :
I �h4 d/bS ! 2 �f2 d/a6 !
Black's king tries to repulse both
White to play and win
threats .
.) �cS ! . . . White manages to queen thanks to
a double attack.
We shall see presently that the en­
tire secret lies in this waiting move. 1 �c8 ! . . .

.) . . . �f4 White already threatens a double

attack by 2 e7. The bishop has to
The bishop is forced to leave its
move aside, since an end-game with
hideout.
pawns is hopeless for Black.
4 �e7 dlbs 5 �d8 dlc6
1 . . . �f1 2 e7 �b5 3 c4 + ! . . .
Black's king has returned in time,
This first do uble attack forces the
but it is powerless against the decoy
king to take the pawn .
attack which immed iately decides
the game . 3 . . . <11 x c4 4 � a6 !
6 �gS ! . . . And the second double attack de­
cides the game because the pawn
[n this manner White succeeds in
cannot be stopped .
bringing his bishop to a7 with the
It is easy to see that in a confronta­
aid of a double attack (attack plus
queening threat) . tion between a knight and a bishop,
a double attack is a very effective
6 . . �h2 7 �e3 OObS 8 �a7 00a6
method . In the ch apter devoted to
.

9 �b 8 �g 1 10 �g3 �a7 1 1 �f2,

the fork we have given a number of
and White is winning.
examples wh ich you can review.
Here you will be acquainted with a
model in which a game centred on
the pawn is dominated by the
double attack.

105
H. Mattison, 1914 now comes the second and this time
decisive fork:
198

8 � d7 + \tla7 9 � x c5, and Black

can lay down his arms.
The role played by the double at­
tack in tournaments has been illus­
trated by n u merous examples (cf.
diagrams 43, 76, 9 8 , 1 1 2 , 1 1 5 , 1 45 ,
1 5 4, 1 57) . Let u s consider two
more.

White to play and win Em. Lasker, 1 890

Black threatens to achieve an imme- 199

diate draw by 1 . . . �f3 or 1 . . .

�e2 . That is why with his first
move, in which he sacrifices a
pawn, White closes the important
diagonal.
l c6 ! dc 2 a6 �f3
For a draw Black has only to play
c6-c5, but . . .
3 � g5 ! �d5 4 � e6 ! . . . White to play and win

The first double attack. White

In this well- known study White
threatens both 5 � cS , after which
succeeds in pushing back Black's
the pawn passes, and the fork 5
king and eliminating his pawn by a
� c7 + Black must choose the
series of double attacks.
.

lesser of two evils.

4 . . . c5 5 � c7 + \tld7 5 � x ds
\tlc8
If 6 . . . \tlc6 , then White's king en­
ters the c-pawn's quadrangle by 7
dlg3 . That is why Black tries to at­
tack the White pawn as quickly as
possible.
1 dlb7 bl b2 + 2 \tla7 . . .
This move merits attention. It con­
tains two threats, i. e. that of queen­
ing and of forcing back the oppo­
nent's king. Hence, basically it is a
double attack.
2 . . bl c2 3 Et hS + (the threat is
.

carried out) 3 . . . dla4 4 \tlb7

7 � b6 + \tlb 8
Et b2 + 5 dla6 ! (another double at­
There is no alternative, because tack) 5 . . . Et c2 6 Et h4 + \tla3 7
otherwise White plays 8 a7 . But dlb6 Et b2 + (with 8 Et x h2 as an

1 06
imminent threat) 8 OOaS El c2 9
El h 3 + 00a2 1 0 .E1 x h2 ! , and
White is winning.
And here is an instructive example
taken from a tournament.
Galic- Vukcevic
Yugoslavia 1 9 75
200
Black to p l ay
Black launches a double attack.
1 . . . e3 !
The rook is u nder auack and the
pawn threatens to queen. White
loses the game immediately if he
tries to parry the attack by 2 f4 , be­
cause this is followed by 2 . . . e2 3
00f2 El e4 4 OO e l El e8 ! and White
finds himself in Z ugzwang. Hence,
White has no choice but to take the
rook:
2 .E1 x d4 e2 3 bl d6 + . . .
White wants to lure Black's king to
the e-file in order to threaten with a
double auack of his own.
3 . . . 00f7 4 .E1 d7 + 00f6 5 .E1 d6 +
00e7 6 .E1 d3 . . .
All this seems very logical, because
White plans to play 7 El e3 + in re­
sponse to 6 . . . e 1 (®) . But at this
point there is an unforeseen devel­
opment.
6 . . . e1 (�) + ! !
White had to admit defeat. The cor­
rect continu ation was 6 El dS !
e 1 (®) 7 El x fS and a drawn game.
What part does the double attack
play in the end-game with queens?
It can be demonstrated that in this
case, too,it is an effective weapon
which he lps the stronger player to
hide his king from checks and to
promote his pawn.
Let us examine the following situa­
tion, for example.
20 1
Black to play
Black's queen gives a series of
checks. But White's king is able to
find a safe haven by taking advan­
tage of his opposite number's post­
ing and of a double attack, e. g. :
1 . . . ® d 5 + 2 OO e l ! (checks on
the e-file cannot be given because of
the double attack 3 ®e2 + ) 2 . . .
® h 1 + (2 . . . ® aS + 3 OOfl
® b s + is again followed by a
double attack by 4 ® e2 + ) 3 d/f2
1 07
®h2 + 4 ®g2! :tnd the pawn pro-
ceeds to queen because 4 . . .
®h4 + is fol lowed by a discovered
check.
Mar6czy-Betbeder
Hamburg 1 930
202
White to play
1 a6 ® a 1 + 2 dlbs ®b2 + 3 00c6
®f6 + 4 d:>c7 ! . . .
White sacrifices a pawn, but after
4 . . . ® x a6 5 ®d7 + and 6
® d6 + he exchanges the queens
thanks to a double attack and wins
the end -game.
4 . . . ®c3 + 5 ®c6 ®e3 6 00c8!
and Black resigned. If Black plays
6 . . . 00f7 , White has no trouble
deciding the issue by 7 a7! ® x a7 8
®d7 + .
The double attack is also the leit­
motif of the following famous old
study.
1 08
L. van Vfiet, 1 888
203
White to play and win
After 1 ®b4 ! Black's queen can
make the following moves on the
di agonal a8-h 1 :
a) 1 . . . ®d5 (or 1 . . . ®f3) 2
®a4 + 00b6 3 ®b3 + ! ® x b3 4
b8 (®) + ;
b) 1 . . . ®g2 2 ®a3 + 00b6 3
cl'J b2 + ! cl't X b2 4 b 8 ( @) + ;
c) 1 . . ®h 1 2 ®a3 + 00b6 3
.
®b2 + 00c7 (3 . . . 00a6 4 ®a2 +
00b6 5 cl'J b 1 + !) 4 cl'J h2 + ! @ X h2
5 b8 (®) + .
In conclusion an end-game in
which a rook is fighting against a
bishop. In such positions a victory is
the exception rather than the rule.
But if a victory is to be achieved it is
only by means of a double attack.
]. Kling and B. Horwitz, 1 8 51 after which the rook penetrates into
the 8th rank.
204
5 l:le2 � d6 6 B e 8 + � f8 7 B d s
and mate on the next move.
You have seen for yourself that the
ro le of the double attack in the end­
game is not less important than in
the middle game, even if it is often
focused not on direct attacks but on
various threats of which queening is
the main one.
Black to play

White threatens a double attack by Studies

1 l:l d7 in which the bishop is at­ of double attacks
tacked and at the same time a mate
is threatened on d 8 . If the bishop The double attack in the broadest
moves to b6 to counter the threats, sense of the word occurs most fre­
White mounts the decisive double quently in studies. At a time when
attack 2 l:l b7. After 2 . . . � c5 3 chess theoreticians still regarded
l:l b8 + � f8 the rook makes a wait­ this device as one of many elements
ing move on the 8 th rank and Black of tactics, thereby underestimating
loses the game. its versatility and significance in tac­
Hence, the only continuation is tical operations, master problemists
1 . . . � g 1 ! In this manner Black had long been aware of the major
tries to elude the double attack by role played by the double attack
finding his bishop behind the White and paid due tribute to it in their
king. Now White must try to s moke works.
the bishop out of his hideout. This On the pages of this book you have
can be done as follows : examined numerous examples of
2 l:lf1 �h2 3 l:lf2 �g3 4 rlg2 ! games that were more or less con­
nected with double attacks . Now
�e5
we should like to acquaint you with
The bishop h a s t o leave its safe posi­ some outstanding works of art that
tion, because if 4 . . . � h4 then 5 are entirely in the realm of the
d/hs + (again a double attack) . On double attack. These studies also
the other hand the decisive answer reflect the aesthetic side of this truly
to 4 . . . � d6 is 5 l:ld2 � e7 6 bl c2 , universal device.

1 09
A. Troitzky, 1 913 2 . . El bS is decisively cou ntered
.

by 3 � e6 + and 4 �d6 + .
205
2 . . . rlds
Black's rook threatens to flee to d3.
3 �e6 + cbe4 4 cbg4 ! hS + 5 \l)h4
Black is in Zugzwang. Any move
with the king or with the rook (ex­
cept El d 1) leads to a fork. The most
obstinate move is 5 . . El d l, but
.

then the fork after 6 �gS + cbf4 7

White to play and win
At first sight White's chances of
winning appear dim, for to win he
has to capture the rook. But at this
stage the idea of capturing the rook
seems to be quite far-fetched . And
yet let us see what happens when
White plays
�h3 + decides the game.
The following two studies show the
queen succumbing to a double at­
tack, but each game takes an en­
tirely different course .
H. R inck, 1 92 7
206

1 �c4 + . . .
After this move the rook has every
reason to feel uncomfortable. This
is because the king has only one
square to which it can move without
i mmediate grave consequences, any
other move exposing it to a fork
from the knight.
1 . . . cbf4
White to play and win
The rook can now move almost to
any square on the 5th rank. But the The mechanism of double attack
next attack shows how limited its has already been prepared here and
freedom of movement is. all White has to do is set it in mo­
tion, although for the time being
2 �g7 ! . . .
Black's queen is beyond the knight's
At this point we realize that the "le­ reach. Despite this White's task is to
bensraum" of Black's rook is re­ prepare a strike against the queen.
stricted to the square dS, for if it Let us try driving the queen to the
moves to f8 , f6 or cS White immedi­ knight especially to prevent 1 . . .
ately forks it and 2 . . El f7 or
. ® f1 + , which wou ld result in an

1 10
immediate draw as a res ult of per­
petual check.
I �g2 . . .
The queen has not got much of a
choice of squares it can move to. It
is lost if it moves to any square on
Lhe c8-h3 diagonal . 1 . . . ®b3 and
I . . ® d 3 are worthless for the
.
same reason. No better is 1 . . .
®a3 or 1 . . . ® c3 because of 2
� x d4 + 00a7 3 �b5 +. Hence,
out of 1 2 squares the queen has only
one it can move to without danger.
I . . ®e3
.
I t seems Black can breathe a sigh of
relief. But this is not so.
2 � x d4 + 00a7 3 � g 1 � . . .
"What a nuisance these confounded
bishops are�" the queen might cry
out in despair. It managed to elude
one bishop, but there is no escape
from the other one.
It is actually impossible to find a
�afe square for the queen on an al­
most empty board. In any case all
Black squares are taboo for the
11 ueen, the transgression of which
results in an immediate loss. If it
moves to d 3 , then White plays 4
-9c6 + with an unavoidable fork.
I [ence out of 1 9 squares the queen
has only one to move to:
.� . . . ®e8
l�u t there, too, it falls prey t o a
double attack:
•I�b5 + dlbs 5 � h2 + 00c8 6
-9d6 + .
L. Kubbel, 1 93 5
207
White to play and win
It is not easy to predict that this is
the prelude to a double attack. Let
us see how the situation develops.
1 ® a 1 + dlh7 2 ®b 1 + . . .
The unfortunate posting of Black's
bishop and king enables White to
improve his queen's placing in a
series of stair-like steps, although it
is not very clear at first what White
is up to .
2 . . . d:>h8 3 ®b2 + dlh7 4 ®c2 +
00 h 8 5 ® c3 + d:>h7 6 ® d3 + 00h8
7 ® h3+ �h7 (7 . . . ®h7 8
® c3 + and mate on next move) 8
® c3 + digS 9 ® c8 � ...
Now we see what White's plan was :
His queen can now threaten a
dou ble attack. His opponent's
queen may not step on the black
squares, if it moves to f7, White
plays 1 0 � h6 + with mate. But
Black's king can avoid the danger­
ous confrontation by
' 9 . . . d:>f7
But at this point it becomes appar-
111
ent that the move 9 ® c8 contained double attack : 2 fg(®) + 00 x g8 3
a second threat, namely that of cap­ � e7 + .
turing the queen. White plays 10
�cS ! , after which the queen is
trapped .
V. Korolkov, 1 951
White to play and win
In Korolkov's study, the entire
thrilling struggle based on double
attacks is centered on the use of the
pawn .
1 f7 . . .
Already White's first move is a
double attack, threatening to queen
and in reply to 1 . . . 11f6 he has the
threat of playing 2 � b2 concealed
up his sleeve. But Black replies in
kind.
1 . . . 11a6 + !
This, too, is a double attack of a
special kind. Black has two threats
in store, i . e. if 2 00b2 then 2
11f6 or if 2 00b 1 then 2
� x fs + is possible.
2 � a3 ! 11 x a3 + 3 d!b2 . . .
Agai n we see a double attack bei ng
mou nted threatening 4 f8 (®) +
and 4 00 x a3. Black meets the sec­
ond threat with a reply which is also
based on double attack.
3 . . . El a2 + !
If 4 00 x a2, then 4 . . . � e6 + fol­
lowed by 5 . . . � x f7 . White's
king tries to hide. But where?
4 d!c l ! . . .
If White had played 4 dlc3? Black's
saving move wou ld have been 4 . . .
11 c2 + ! leaving no escape for the
king.
4 . . . rla 1 + ! (4 . . . 11c2 + would
have led to a quick defeat, for the
pawn cannot be stopped) 5 dld2
11a2 + 6 00e3 El a3 + 7 dlf4
11a4 + 8 00g5 . . .
The game seems to be over, but
Black launches a fresh cou nter-of­
fensive, which is also based on
double attack.
8 . . . El g4 + ! 9 d/h6 (9 d/ X g4
� x fS + with a draw) 9 . . . 11g8
Through a titanic effort Black ma­
nages to mobilize his rook in the
struggle with the pawn . But its plac­
ing is too weak to be effective, and
White wastes no time in utilizi ng
the situation to his advantage.
1 0 � e7! � e6 !

The other defence by 1 . . . 11g 8 i s . Black puts up a desperate defence

doomed to failure because of the and parries the threat of 1 1 fg(®)
mate.

1 12
1 1 fg(®) + � x g8 1 2 �g6 mate .
The series of double attacks ex­
changed between the two sides has
produced an unexpected and un­
ique mating position.
Although perhaps not quite so· ex­
citing, the following study leaves a
l asting impression. Here the
struggle flares up around a double
attack planned by White.
M. Have!, 1926
209
White to play a nd win
The start of the contest seems to
herald no great events. To prevent
Black from moving his bishop from
f8 and giving check, White plays:
1 � f2 . . .
S ince the bishop at g4 is attacked usu ally both sides use the double at­
White must be prevented from play­
ing 2 EI gS. Black has no choice but
to play
1 . . . �e6
Now the bishop is fair game for the
rook.
2 ld g6 �ds �
Black has to be very cautious in
choosing a safe square for the
bishop. Bad is 2 . . . � f7 or 2 . . .
� a2 because of the immediate
double attack. 2 . . . � c4 is just as
unsatisfactory in view of the double
attack by 3 ldc6, while 2 . . . �b3
leads to the main version .
3 ldgS � c4 4 .El g4 �b3 5 .El g3
� a2
At last the bishop has found "safe"
refuge at a2, having threaded a nar­
row path to escape h arass ment by
the rook . . But this has created favo u­
rable conditions for a double attack
on the a-file, and White immedi­
ate ly exploits the opportunity.
6 tif3 ! . . .
Now the second bishop is forced to
follow no less narrow path .
6 . . . � e7 7 .E!e3 � d6 8 .E!d3 �b4
9 &ld4 , and White has succeeded
after all in forcing his opponent into
a position which leaves him open to
a double attack.
In accurate combination studies,
tack and the one who "laughs last"
wins, i . e. the one who executes the
last double attack.
Here are some characteristic exam­
ples.
113
A. Troitzky, 1934 2 . . . El h2 + 3 00g4! El X h 1

210
Black has carried out his double at­
tack and achieved a balance of for­
ces which promises to end in a
draw. But in the course of the game,
White himself has set the mech:J.n ­
ism of do uble attack (in the present
case it is the discovered check) in
motion. He uses it in order to pre­
pare another form of double attack,
which is som etimes referred to as
discovered attack.
Wh ite t o play and win
4 � g l +! 00c7 5 Elb l 00d7

The game starts with a position in Black's king may move only on
which White has to ward off a white squares, otherwise the rook is
double attack, with the bishop lost when the bishop moves. With
hanging and 1 . . . El h2 + being his next move, White forces the
threatened . If he tries to bring the king to move to the queen's side
bishop to safety by going 1 � d S , and drives it towards the edge.
Black presses h o m e the attack b y 6 Be l ! \t>c6 7 Eld l \t>bs 8 Bel
playing 1 . . . �d6 2 EleS �c4 ! d/a4 9 Elb 1 dla3 1 0 � cS + or
and the game ends in a draw. This 9 . . . d!aS 10 �b6 + and wins.
leaves White no choice but to give
check.
A. Zeletski, 1931
1 Elbs + . . .
21 1
Black's king may not step aside be­
cause 1 . . . dla7 2 El aS + d/b8 3
El a8 mate, but Black finds a brilli­
ant solution.
1 . . . �b6 !
With a view to continuing with
2 . . . d/c8 after 2 El X b6 + after
which the loss of a bishop is un­
avoidable.
White to play and win
2 � X b6 . . .
Black gets a breather and is appar­ In this explosive situation White has
ently able to take the bishop with already made preparations for a
impunity. double attack (his opponent's

1 14
bishop and knight are hanging) , but We shall soon see that he has an­
I he knight fork at d2 prevents the other double attack up his sleeve.
rook taking the bishop. This threat
7 ld. x g6 + dlh?
. � till needs some preparation.
N o w it i s White w h o faces a prob­
I d/g2 � d 3 2 E:!d 1 . . .
lem. How can he find a way out of
Hlack's situatio n is critical. 2 the situation which has arisen? The
.\.�c2 is followed by 3 ld. d8 and if best way is by a double attack .
� . . . �a7 then 4 dlf3 �d2 + 5
8 ld. e6 ! . . .
.

11 x d2 � x g6 6 ld. g2 + with a de­

risive double attack. Black must de­ We shall soon see that this is an ex­
fend himself against the threatened tremely important move . This is
4 cbf3 by 3 . . . � f4 , but this is fol­ precisely the point from which the
lowed by 4 E:!d 5 + cbg4 ( 4 . . . rook can fu lfil its last mission.
\bh4 5 dlf3 again with a double at­
8 . . . � g3 + 9 dle2 ! dl x h8
tack) , and a-surprising mate by 5
�h5 + dlh4 6 �g6. But Black has
:t very effective reply, which is also 212

based on double attack.

2 . . �f2 ! 3 dl x f2 � a7 + ! 4
.

\"be 1 ! (as we shall presently see, e 1

is the only square left for the king) .
4 . . . � X g6
The first skirmish ended in Black's
favour. But the next double attack,
which threatens the next, brings
Black to the brink of defeat.
At last Black breathes a sigh of re­
5 ld. d7 � b 8 lief, for a draw seems to be in sight.
But White's next move restores or­
Here the posting of White's king
der.
plays a role. If he stood at e2, Black
could save himself by playing 5 1 0 dlf3 ! and the issue is decided,
� h5 + . for the only squares the bishop can
move to, i. e. h2, h4, c7 and b8, are
6 E:!g7 dlh6 !
all open to double attacks leading to
But Black's courage is not broken . the loss of the bishop.

1 15
G. Kasparyan, 1 9 50 s . . . � d 7 ! 6 E1c7 .E!ds
213

214

Draw

Black has managed to save two of

White is in a precarious situation.
His kn ight is threatened and cannot
move anywhere. If he moves to d8,
Black makes a double attack by the
bishop from e7 ; if it moves to d6 a
double attack on cS follows. The at­
tempt to convert the passed pawn is
also doomed to failure : 1 dlf6
bl X b7 2 E1 X b7 � X b7 3 e7
� b2 +! 4 dle6 � c6 5 dlf7 � d5 +
6 dlf8 �g7 + 7 dle8 �e6 8 dld 8
� f6, and Black wins.
So what can White do? He tries a
double attack :
1 � d6 � cS 2 e7 ! E1 X e7 3 � f5 !
� x fs
Black is two b ishops up, but White's Black pieces can go on moving back
next move conj ures up a double at­ and forth perpetually, the game is
tack.
4 E1c6! E! eS S dlf4 . . .
Now it is Black who is in trouble.
Three of his pieces are threatened,
and if he goes S . .E! dS, then
. .
his threatened pieces, but White
wastes no time before mounting the
next double attack.
7 dle4! � b6 8 .E! b7 rl d6 9 dieS !
� c5 1 0 .E!c7 �b4 1 1 .E! b7 �a3 1 2
El a7 rld3 (or 1 2 . . . .E!e6 + 1 3
d/ds .E!d6 + 1 4 dies etc.) 1 3 dle4
� cS
Black's attempt to lau nch a double
attack of his own fails: 1 3 . . .
.E! x e3 + 1 4 dld4! draw.
1 4 bl c7 �b6 1 5 .E!b7 .E!d6 ( 1 5
.E! x e3 + 1 6 dlf4 !) 1 6 dieS! and
everything starts all over again .
Since the throng o f White and
considered drawn.
This study is truly an "ode to the
double attack" . Somewhat later
Korolkov took Kasparyan's idea a
little further and created a study in
which the stronger side succeeds,

Wh ite rep lies with 6 e4 and ends the

game with an immediate draw.
Thus, his only hope is a counterat­
tack.
u nder similar circumstances, in
breaking the double attack dead­
lock.

1 16
V Korolkov, 1 953/54 double attack or at least the threat
of one occurs almost in every game.
215 We have picked eight games in
which the double attack either de­
cided the game or was its leitmotif,
in order to illustrate how situations
connected with double attacks de­
velop.
We shall try to analyse the motives
of double attacks with a view to
fathoming those mysterious forces
which engender them.
White to play and win
Queen's Gambit
White's bishop and rook are u nder R ubinstein-Snosko-Borovski
attack, and if he loses one of these Petersburg 1 909
pieces his prospects are bleak. Only 1 d2-d4 d7-d5
a counterattack can save him. 2 c2-c4 e7-e6
1 �b4 Elb3 3 � bl-c3 � g8 -f6
4 � c l -gS � f8 -e7
As in the preceding study, the rook 5 e2-e3 � b 8-d7
moves aside keeping an eye on the 6 � g l -f3 0-0
two bishops. 7 ® dl-c2
2 Eld4 eS Rubinstein's favourite move at this
Worse is 2 . . . cS 3 � aS+ dld7 4 point. He still has the cho ice of cast­
� c2 El a3 5 El a 4, and retains his ling either short or long.
piece advantage . 7 ... b7-b6
3 � c2 Elb2 4 Elc4 d5 5 � a3 E! a2 6 According to modern theory, the
l:Xc3 d4 7 � b l Ela l 8 Elb3 most energetic way of disrupting
As you can see, White h as suc­ White's build-up is by going 7 . . .
ceeded in asserting his m aterial ad­ c5 , disregarding the disadvantage
vantage. Victory is only a question of having a Black pawn at d5. After
of time. the text move, White realizes a plan
which is connected with long cast­
ling and an attack on the king's
side.
Sample g ames
8 c4 x dS e6 x d5
9 � fl-d3 � c 8 -b7
We have convinced ourselves of the
importance of double attacks. It is 1 0 0-0-0
no exaggeration to say that a In a game between Alekhine and

1 17
Yates played in Hamburg a year la- 1 4 � c3 x e4 f5 x e4
ter, an even stronger move 10 h4
was played, preventing the relief
move 10 . . � e4. .

10 . . . � f6-e4
1 1 h2-h4 f7-f5
1 2 dlct -b 1 c7-c5 ?

21 6

1 5 � d 3 x e4 !
Let's get down to brass tacks. By
sacrificing the bishop, White
creates the basis for a double attack
along the opening diagonal. As you
will soon realize, all this required
Already Black's 1 1 th move seems
extensive and exact calculation.
rather dubious (better is the quiet
move 1 1 . . . Ele8 ) , but h is 1 2th 15 . . . d5 X e4
move is an outright mistake, which 1 6 ® c2-b3 + cbg8-h8
in the given position creates condi­ 1 7 ® b 3 X b7 e4 x f3
tions for White to launch a double
Black pins all his hopes on this
attack. Rubinstein makes masterly
pawn.
use of the opportunity created by
these conditions, namely the open­ 1 8 Eld1 x d7
ing of the diagonals .
Another double attack, this time on
1 3 d4 X c5 b6 X c5 the bishop and the queen.
At this point we realize the strength 18 ... ® d8-e8
of White's 1 2th move. If the White 19 Eld7 x e7 ®e8-g6 +
king were still at c l , then Black 20 cbb 1 -a 1 Ela8-b8
could play 1 3 . . . � d x c5, because 21 ® b7-e4 ®g6 x e4
14 � x d5 is worthless, because of 22 Ele7 x e4 f3 x g2
14 . . . � x d 3 + . But now the reply 23 Eih 1 -g 1 Eif8 X f2
to 1 3 . . . � d x c5 is 1 4 � x d5!
The first impression is that by his
� x d5 15 � c4 and White wins.
sacrifice Black is able to stage a
The ability to foresee such nice
dangerous counterattack, because
points is an indispensable attribute
his pawn at g2 looks very threaten­
of the good chess-player.
ing and on top of that pawn b2 is

118
hanging. But White has calculated short on the 8th move, the knight
everything down to the last detail, move to aS being made after 9 h3.
and his next move darifies the situa­
9 �b 3-c2 c7-c5
tion.
1 0 d2-d4 ® d 8 -c7
24 Jde4-f4 ! 1 1 � b 1 -d2 � a5-c6
1 2 � d2-fl
This is how easily White fends off
the threats, for if 25 . . . }db x b2 This move, which was earlier used
(or 25 . . . }df x b2), then 26 Jdf8 + by Lasker, was very popular in
}d X f8 27 d/ x b2 with an easily those days. Today White usually
won end-game. plays 1 2 de (Rauser) or 12 d S . As
shown by the game played by Geller
24 . . . hlf2-c2
and Mecking (lnterzonal Tourna­
25 b2-b3 h7-h6
ment, Palma de Mallorca 1 970),
26 � gS-e7 l:t b8-e8
B lack gets a difficult game also in
27 00a 1 -b 1 !
the latter case.
The last fine point, which had to be
12 . . . c5 X d4
anticipated well in advance.
1 3 c3 X d4 �c8-g4
27 . . . hlc2-e2
Black's position is by no means bad
2 8 � e7 x cS l:t e8-d8
even if he plays 1 3 . . . ed.
29 � c5-d4 l:t d 8-c8
3 0 Jdf4-g4 14 d4-d5 � c6-d4
1 S �c2-d3 0-0
This final double attack ends the
fight. Black resigned . 1 5 . . . � h5 ! 1 6 � e3 � x f3 + 1 7
g f �d7 would have been more in
Spanish Opening keeping with the spirit of this game.
Capablanca-Dus-Khotim irski
1 6 �c l -e3 Jda8 -c8
Petersburg 1 91 3
And here it would have been better
1 e2-e4 e7-e5
to move the other rook to this
2 � g 1 -f3 � b8 -c6
square.
3 � ft -bs a7-a6
4 � b5-a4 � g8-f6 1 7 � e 3 x d4 eS X d4
5 0- 0 � f8 -e7 1 8 a2-a4 !
6 Jdft - e 1 b7-b5
White immediately attacks his op­
7 � a4-b3 d7-d6
ponent's pawns.
It is interesting to note that the var­
18 . . . ® c7-b6
iant played here is in keeping with
1 9 a4 X b5 a6 x bs
the latest modern theory.
20 h2-h3 � g4 X f3
8 c2-c3 � c6 --'a5
White now gets a permanent advan-
Nowadays Black usually castles

1 19
tage in view of the weak pawns at
bS and d4. How should Black have
continued? Bad was of course
20 . . . � h5 because of 21 g4, but
by 20 . . . � d7! he wou ld have kept
his two bishops and would have had
good prospects of countering, be­
cause if 21 �g3 with the threat 22
� f l , then 21 . . . � e8 22 � fl � f6
23 �hs � e s .
2 1 ® d 1 x f3 �f6-d7
Black transposes his knight to c5
and clears square f6 for the bishop.
2 2 rle t -c l ! �d7-c5
23 b2-b4 �c5-a4?
He suspects no evil designs . He
should have exchanged at d3 and
then played his bishop to f6 .
24 rlct x c8 rlf8 x c8
218
Please take a close look at this posi­
tion. The Black rook is not covered,
and this circumstance enables
White to mount a tactical operation
in the form of a double attack.
g7-g6
Although Black has parried the
threatened 26 ® fs , the pawn takes
1 20
over the role of blockbuster and de­
stroys the Black king's fortress.
26 e5-e6 rlc 8-f8
If 2 6 . . . fe then 2 7 eg4 .
2 7 �fl-g3!
Capablanca's attack is very forceful
indeed. To 27 . . . fe he intended to
reply with 28 �g4 e5 29 � x g6 !
27 . . . ® b6-b7
219
28 �g3-f5 !
The sacrifice can not be accepted
because of the inevitable mate after
29 ® x fs . With the aid of the
double attack White also throws his
knight into the attack. What can
Black do? He is threatened with 29
� h6 + and 30 � x f7 and with loss
of material.
28 . . . f7 x e6
A natural move to prevent the pawn
from being threatened. But this
move enables White quickly to de­
cide the battle in his favour by open­
ing a file. The most tenacious move
would have been 28 . . . dih8 in or­
der to deprive White of the oppor­
tunity of giving check.
29 d5 x e6! 4 �f3 x d 4 �g8 -f6
5 �b 1 -c3 d7-d6
Another surprise in the form of a
6 � f1 -c4
double attack ! To 29 . . . -&1 x f3
White replies with 30 gxe7 + , White's bishop has obvious inten­
and White captures a piece . tions regard ing squ are f7 . For a
long time it was believed that Black
29 . . . ®b7-c7
could easily neutralize it by e7-e6 .
But in the fifties it was discovered
220 that White can maintain the
bishop's attacking momentum by
the thrust f2-f4-f5 , especially if
this plan is accompanied by, long
castling.
6 . .
. e7-e6
7 0-0
Today, one usually pl ays 7 �e3 , 8
® e2 , followed by long castling.

If Black had played 29 . . . ® a7 , 7 . . . a7-a6

then things would have been easy 8 � c l -e3 ® d 8 -c7
for White : 30 �h6 + cbg7 3 1 �f7 , 9 � c4-b3 � f8-e7
threatening capture at b5 and 32 10 f2-f4 �c6-a5
'&1f4 with a telling attack. 11 ® d l -f3 b7-b5

30 ® f3 -c6 ! This is how-until this game-this

variant was frequently played,
Still another double attack, the last White's next move usually being 1 2
one in this game.
fS , which achieved nothing. For this
30 . . . ®c7-d8 reason the entire system was re­
31 �f5 x e7 + ® d 8 x e7 garded as harmless for Black.
32 � d 3 x b5 � a4-c3
1 2 e4-e5!
33 ® c6-d7 ®e7 x d 7
34 � b5 x d 7 , a n d White won In those days this was a novel idea :
quickly. White offers a pawn sacrifice .

12 . . . � c 8-b7
Sicilian Defence 13 ® f3 -g3 d6 X eS
A verbakh- Taimanov 1 4 f4 X eS �f6-h5
Zurich 1 953 15 ®g3-h3
1 e2-e4 c7-c5 White certainly did not sacrifice the
2 �g 1 -f3 gb8 -c6 p awn in order to win a piece by 1 5
3 d2-d4 c5 x d4 ® f2 0-0! 1 6 g4, because in this

121
case it is possible to play 1 6 . . . 222

� X b3 1 7 ab 'C!t:J X eS 1 8 gh '<!tJ X h 5 ,
a n d Black has a strong attack going.
15 . . . ®e7 x e5

221

An instructive position of a double

attack has developed . Two things
are threatened by Wh ite : first, to
recapture the piece by 1 8 ® x hS +
® x h5 1 9 � x g7 + , and secondly,
if the knight dodges to f6 , to cap­
ture the queen by 1 8 � f4. If Black
The first and at the same time deci­
covers the knight by 1 7 . . . g6, then
sive mistake . Analyses have proved
1 8 � d 4 decides the issue. There is
that it is correct to play 1 5 . . .
no satisfactory defence in sight for
� x b3 1 6 � x b3 and only then
Black.
1 6 . . . ® x e5. If White plays 1 7
� aS , Black can afford to get in­ 17 ... �b7-c8
volved in complications which are 18 ®h3 X h5 + ®e5 x h5
not devoid of advantage for him, 19 � e6 x g7 + dle8-d7
i . e. 1 7 . . . b4 ! 1 8 � c4 ® c7! 1 9 20 � g7 x h5
'C!t:J X h S 0-0.
The two-pawn superiority is of
1 6 � b 3 x e6! course entirely s'u fficient for a win,
but Black cannot make up his mind
The bishop, which was exposed for
to give up and continues the game
such a long time to the danger of
to the bitter end .
being taken, now offers itself to be
taken. Black has to take it, because 20 . . . � a5-c4
if 1 6 . . . 0-0 then 1 7 EHs, a very 2 1 �e3-d4 rih 8-g8
unpleasant prospect, especially 2 2 � c3-d5 rlg8-g5
since 1 7 . . . � f4 is doomed to fai­ 2 3 � h5-f6 + �e7 x f6
lure due to loss of a piece after 1 8 24 � ds x f6 + dld7-c6
ri X e5 � X h 3 + 1 9 � X h 3 . 25 � f6 X h 7 rlg5-g6
2 6 ria 1 -e l b5-b4
16 . . . f7 x e6
2 7 b2-b3 � c4-a3
1 7 � d4 x e6
2 8 r!e l -e5 � a3 -b5
29 � d4-e3 � b5-c3
30 � h7-f8

1 22
White's pieces have taken up active Euwe mistakenly played 10 b4 and
positions. The final offensive be­ after 1 0 . . . d4! found himself in
gms . difficulties, because after 1 1 � a4
came the effective 1 1 . . . de ! 1 2
30 ... E! g6 -g7
'i!f1 d 8 ( 1 2 � X c5? ef + 1 3 d/e2
X
31 E! f 1 -f6 + dlc6-c7
� g4 + ) 1 2 . . . ef + 1 3 c!Je2
32 �e3-h6 E! g7-g4
�g4 + . Black wins because White's
33 E! e5-e7 + dlc7-d 8
king has to move to the d-filc fol­
34 E! e7 -'-h7 E! a8 -b 8
lowed by check and capture of the
35 � f8 -e6 + �c8 x e6
queen.
36 E! f6-f8 mate.
10 . . . � b 8-c6
1 1 0-0 � c8-e6
Nimzowitch Indian Defence 1 2 � c3-e2 � c5-d6
Uhlmann-Averbakh 1 3 � e2-d4
Dresden 1956 Wh ite does not try to exploit the
1 d 2 -d4 � g8 -f6 weakness of the pawn at d5 and his
2 c2-c4 e7-e6 play is somewhat weaker. To 1 3
3 � b 1 -c3 � f8-b4 �d2, Black could have replied
4 e2-e3 0-0 1 3 . . . � e4! and chances would
5 � g 1 -e2 have been nearly equal .

This is hardly better than playing 13 · · · � c6 X d4

the knight to f3, but as a rule it leads 14 e3 d4 X E! f8-e8
to variations that have not been so The attempt to go over to the offen-
well explored. sive without delay does not pay :
5 . . . d7-d5
!: � g4 J; h3
· · . �X f2 1 6
.M.,
Id x f2 ! ( 1 6 w x f2 gh4 1 7 gf3 fS !
6 a2-a3 �b4-e7
7 c4 x d 5 e6 x d5 with a strong attack) 1 6 . . . � x g3
1 7 � x h7 + \t> x h7 1 8 ® d3 +
After 7 . . . � x d5 White can con­ with a double attack on h7 and g3.
tinue by going 8 g3, after which his
1 5 � g3-f5 � e6 x fs
opponent will have a hard time
1 6 �d3 f5 X ® d 8 -b6 !
proving that his position is n ot
worse. Black has overtaken his opponent in
development and now plans to play
8 � e2-g3 c7-c5!
on both Hanks.
This move, which occurred already 1 7 b2-b3 g7-g6
in the 2 5 th game between Euwe and 1 8 � f5-h3
Alekhine in 1 93 7 , is the best way to
White plans to play g2-g3 in order
restore equilibrium.
to bring his bishop to g2 . In this
9 d4 cS
X �e7 c5
X manner he protects the K-side and
1 0 � f 1 -d3 trains his sights on d 5 .

1 23
18 0 0 0 � f6-e4 ® x d5 and 32 . . . � x d 1 ) 30
1 9 ® d l -d3 � d 6-e7 E1 e2 ! and Black wins .
The bishop takes up and influential 29 . . . El e7-e2
position . 30 Ei d l -f l El e8-e3
20 g2-g3 �e7-f6
2 l � c l -e3 El e 8-e7 223

2 2 El a l -c 1 El a8-e8
2 3 Ei f l -d 1
Black has the initiative, his forces
are concentrated in the centre.
23 0 0 0 ® b6-d6
24 � h3-g2 !
White is cautious. If 24 b4 then
24 . . . � x g3 ! 25 hg E1 x e3 26 fe
® x g3 + 27 �g2 � g5 , with a very
dangerous attack for the sacrificed Black sacrifices a rook. White ac­
rook. cepts the sacrifice assuming that this
will lead to a draw. But the strongest
24 h7-h5
reply would have been 3 1 El c3 !
0 0 0

Black intensifies pressu re on the K­ b6 32 ® x a6 E1 x c3 33 ® x e2

side. If 24 . . . ® x a3 then White � X h3 + 34 � X h3 El X h3 35
would have reacted with 25 � x e4 ! ® e5 , and White has a fighting
de (25 . . . El x e4 26 El c5 El d8 27 chance .
E1 X d5 El X d5 2 8 "@ X e4 "@ X b3
3 1 f2 x e3 Ei e2 x g2 + !
29 El b 1 ® c4 30 El x b7) 2 6 ® c4
® d6 27 d5 , with very promising By no means 3 1 . . . ®g3 because
play. of 3 2 ® x d5 .
Very weak would also have been
25 ® d3-b5 a7-a6
31 . . . � X h3 + 32 d:>h 1 E1 X g2
26 ®b5-a5 h5 -h4
27 g3 -g4 because of 3 3 El c8 + d:>g7 34
Ei X f7 + ! d;> X f7 (34 . . . d:>h6 35
After 27 �f4 ®e6 2 8 � e5 h3 Black et> x g2 ®g3 + 3 6 d:>f1 , and White
wou ld have had an attacking posi- wins) 35 ® c7 + ® x c7 36
tlOn . Ei X c7 + and 37 et> x g2 .
27 � f6-g5
32 d:>g 1 X g2 YJi:td6-g3 +
0 0 0

28 � e3 X g5 �e4 x g5 "@g3 X h3 +
33 00g2- h 1
29 h2-h3 34 00h 1 -g 1 ®h3 x e3 + !
If 29 ® x ds then 29 . . . � h3 + 30
An important fine point. Although
dlfl (30 dlh 1 � x f2 + 31 d/g 1
it sounds un likely, Black manages

1 24
to capture both of his opponent's
rooks. D. Byrne-Fischer
New York 1 956
35 d/g 1 -h 1
White cannot save the situation
even if he goes 35 El f2 because of
35 . . . � h3 + 36 d)h 1 � X f2 +
3 7 d/g2 h3 + 38 cj)f 1 � d3 with an
irresistible mating attack.
35 . . . ®e3-h3+
3 6 e,bh 1 -g 1 ® h3 -g3+
3 7 d/g 1 -h 1 � g5 -f3 !
Black now threatens to play 38
® h 3 m ate. White has to give up a
rook for a knight and is then faced
with the inevitable loss of his second Indian Defence. Norm aliy this posi­
rook, being unable to prevent a
double attack. If for instance 3 8
bl c8 + e,bg7 3 9 bl x f3 then 39 . . .
® X f3+ 40 d/h2 ®g3+ 4 1 d/h 1
® h3+ 42 d/ g 1 ® x g4+ and
43 . . . ® x c 8 .
3 8 ® a5-d8 + d/g8-g7
39 bl f1 X f3 ®g3 X f3+
40 d/h l -h2 ®f3-f4+
41 d/h2-g2 ®f4 X g4+
White resigned . Black retains his su­
periority after 42 d/f2 ® f4 + 43
dle2 ® x c l 4 4 ® x ds ®c2 + 45
dle3 ® c6 46 ®eS + ® f6.
Griinfeld Indian Defence
1 � g 1 -f3 � g 8-f6
2 c2-c4 g7-g6
3 � b 1 -c3 � f8 -g7
4 d2-d4 0-0
5 � c l -f4 d7-d5
As we shall soon see, this approach
to the opening game is characteris­
tic of modern chess. For a certain
length of time the players conceal
their true intentions. Only at this
point can it be said that they have
decided in favour of the Grlinfeld
tion occurs after 1 d4 � f6 2 c4 g6 3
� c3 ds 4 � f3 � g7 5 � f4 o-o.
6 ® d 1 -b3
In this way White covers the pawn
and strengthens the pressure on the
centre. But as you know a queen
that is sent into action too early in
the game is very likely to become
the target of attack itself.
6 . . . d5 X c4
7 ®b3 X c4 c7-c6
8 e2-e4 � b 8 -d7
This quiet development move is not
in keeping with the spirit of the de­
fence chosen by Black. Black had to
try to exploit the b ad posting of the
enemy queen in order to take over
the initiative by 8 . .. b5 9 ®b3
® a S and by the threat 1 0 . . . b4.
9 bl a l -d 1 � d7-b6
1 0 ® c4-c5 � c8 -g4
1 1 �f4-g5

1 25
 224
White would like to prevent
� f6-d7. But the text move is not
only b ad because it means loss of
time which could be used more
profitably for development, but also
because it leads to an unsound dis­
position of pieces . Master Byrne of
course did not consider all the con­
sequences of his sortie, and he
could hardly have expected that the
1 3 -year-old you ngster sitting oppo-
site would so elegantly exploit the
obliges White in his desire to mount
a double attack of his own.
14 �gS X e7 ®d8-b6 �
A cold-blooded reply. If White ac­
cepts the exchange sacrifice by 1 5
� x f8 he faces a very vigorous at­
tack after 1 5 . . . � x f8 1 6 ®b3
fl x c3 17 bl d 3 ( 1 7 'l!tt x c3 ?? �b4)
1 7 . . . ® x b3 1 8 ab � b4 .
1 5 � f 1 -c4
Byrne j ust cannot face the fact that
his young opponent has outplayed
him so masterfully. He tries to com­
plicate the situation and soon finds
himself in a worse mess than before.
15 . . . � e4 x c3 !
The boy shows no fear and calcu­
l ates the complications with great
· precision. If 1 6 t!ti x c3 E1 fe8 1 7
� x f7 + ( 1 7 ®e3 'l!ttc 7) 1 7 . . .

weak move by the bishop. dl x f7 1 8 � gS + , then 1 8 . . .

d/ X e7 1 9 0-0 � X d 1 20 bl X d 1
11 � b6-a4! !
.Mv.,b 5 , an d Black f01·1 s th e attac k .
· · ·

A thunderbolt ! This shows that the

1 6 � e7-c5 bl f8 -e8 +
future wor Id ch amp10n· a1 rea d y pos-
1 7 \t>e l -f l
sesses the incisive vision it takes to
detect weaknesses in the opponent's
position. With this telling move he 225

put his opponent in a critical situ a­

tion. For the sake of a double attack
he temporarily sacrifices a piece,
which he recovers after 1 2 � x a 4
� x e4 1 3 ® c l ( 1 3 'l!ttb 4 � x f3 1 4
gf � x gS) 1 3 . . . ® a s + emerging
with a material advantage.
1 2 ® c5-a3 � a4 x c3
1 3 b2 X c3 � f6 x e4
Having carefully considered the White apparently awaits the
consequences Black seemingly knight's retreat with satisfaction,

1 26
hoping for 1 7 . . . � b5 1 8 28 � h 1 -e l � e8 X e 1
� X f7 + ! cj) X f7 1 9 ®b3 + � e6 29 ® b6-d8 + � g7-f8
20 � g5 + . But he is in for a sur- 3 0 � f3xe 1 � c4-d5
pqse. 3 1 � el-f3 � f2-e4
32 ® d 8-b8 b7-b5
17 . . . � g4-e6 !!
33 h3-h4 h7-h5
Black sacrifices his queen. In later 34 � f3-e5 d;g8-g7
years experts would point out that 35 d;h2-g 1 � f8 -c5 +
one of Fischer' s tactics was to frus­ 36 d;g 1 -f1 � e4-g3 +
trate his opponent's plans by u n ­ 3 7 d:>f1-e 1 � c5-b4 +
predictable moves. A s you s e e here 38 d;et-d 1 � d5-b3+
the American already possessed this 39 dld 1 -c l � g3-e2+
valuable gift very early in life. 4 0 d;c l -b 1 � e2-c3+
4 1 d;b 1 -c1 � a2-c2 mate.
18 � c5 X b6
Noth ing better than this. If 1 8 Queen's Gambit
� x e6 then 1 8 . . . ®b5 + ! 1 9
Averbakh-Estrin
cbg t � e2 + 2 0 ill f 1 � g3 + 2 1
Moscow 1 964
00g1 ®f1 + followed by 2 2 . . .
� e2 mate. To 1 8 ® x c3 Black had 1 c2-c4 � g8-f6
planned to reply 1 8 . . . ® X cS! 1 9 2 � b 1 -c3 e7-e6
de � x c3. 3 � g 1 -f3 d7-d5
4 d2-d4 � f8-b4
18 . . . �e6xc4 + 5 � c l -g5 d5 X c4
1 9 d;f l -g l � c3-e2 +
Black chooses the aggressive
An i nstructive demonstration of the Vienna variant which was in vogue
effective use of discovered check. in the thirties.
20 cbg 1 -f l � e2 xd4 + 6 e2-e4 c7-c5
2 1 d;f1 -g 1 � d4-e2 + 7 � flx c4
22 d;g l -f1 � e2-c3 + At this point one usually first played
23 d) f l - g l a7 x b6 7 e5 cd 8 ®a4 + � c6 9 0-0-0
2 4 ®a3-b4 � a8-a4 � d7 with a wild game ensuing, un­
25 @b4 X b6 � c3 xd l til Simagin discovered that due to
the double attack on g5 and c3 after
'rhe knight has done a thorough
7 � x c4 cd 8 � x d4 ®aS it is ad­
job : for his queen Black got one
visable to play 9 � x f6 ! � x c3 +
rook and two pieces. This superior­
1 0 be ® x c3 + 1 1 d)f1 ! ® x c4 +
ity is sufficient for a win in this posi­
1 2 d;g 1 , because the threat 1 3 � c 1
tion. The quickest road to victory is
leaves Black no time to take the
by a piece attack on the king.
piece at f6, while after 1 2 . . . � d7
26 h2-h3 � a4 xa2 1 3 � c l ®a6 1 4 � x g7 � g8 1 5 a4!
27 d;g 1 -h2 � d l X f2 White has good attacking chances.

1 27
 7 ... cS X d4 ing White's queen and square f2 .
8 � f3xd4 ®d8 -c7 But since there is no mating threat,
White simply disregards the threat.
Until this game this move was re­
garded entirely satisfactory. After 9 1 2 ® c3xg7! E1h8-f8
®a 4 + � c6 1 0 � x c6 � x c3 + 1 3 � g5-h6
1 1 be �d7 Black indeed has a good
He continues playing in this spirit.
game .
13 . . . ® c S X f2+
9 ® dl-b3 !
1 4 dle 1 -d 1 � b 8-d7
Apparently White has overlooked 1 5 Eth l -e l � e4-f6
the threatened double attack.
After 1 5 . . . � ecS White can play
9 ... �b4x c3 + either 1 6 � c7 +, followed by 1 7
� x a8, or 1 6 � e3 ® h4 1 7 �gS
Relatively better is 9 . . . ® cs , al­
® g4+ 1 8 �e2 ® a4 + 1 9 b3,cap­
though in this case,too,White gets
turing the queen.
a certain advantage by 1 0 � X f6 gf
1 1 0-0. Black pins all his hopes on 1 6 � c4 x e6 ®f2 xb2
the impending double attack. 17 'lda 1 -c l !
1 0 ®b3 X c3 � f6 X e4 Black resigns. Nothing can save
him. 1 7 . . . ® x bS is followed by
226 the decisive 1 8 � c4 + , or 1 8
� X d7 + d/ X d7 1 9 E1 c7 + d/d8
20 ® X f8 + and mate on next
move.

Tarrasch Defence
Averbakh-Platonov
Alma Ata 1 969
1 c2-c4
This pawn move is seen frequently
White's queen and bishop are under
at modern tournaments . But it hap­
fire. The first impression is that
pens frequently that as soon as the
White will not be able to extricate
competitors have revealed their
himself without losing material. But
plans we see patterns emerging on
his momentous lead in development
the board which are other than the
with the weak points in his oppo­
English opening. That is how it was
nent's camp enable him to mount a
in this game,too.
counter double attack.
1 ... c7-c5
1 1 � d4-b5 ! . ® c7-c5
2 � g 1 -f3 � b8 -c6
Yet another double attack threaten- 3 e2-e3

1 28
White prefers the prosaic Queen's
Gambit to the "romantic" Reti
Opening. Black has no choice but to
accept the invitation, because if he
plays 3 . . . g6 White replies 4 d4
�g7, and his prospects are better
thanks to the thrust 5 d 5 .
3 . . . e7-e6
4 � b 1 -c3 � g8 - f6
5 d2-d4 d7-d5
6 a2-a3 � f6-e4
Regarded as a sound answer to
White's pedestrian move. But Black
has a n-t.�mber of promising ways of 1 5 � d 3 x c4
continu ing the game : 6 . . . �e7
6 . . . a6 and finally 6 . . . cd.
7 � f 1 -d3 � e4 x c3
8 b2 x c3 � f8-e7
9 0-0 0-0
1 0 � c l -b2 b7 -b6 ?
What looks like a natural move puts
Black in difficulties. The chances of
the two sides would h ave been
about even had he chosen the cor­
rect reply, namely 1 0 . . . de 1 1
� x c4 b6.
1 1 c4 X d5 e6 X d5
If Black had taken the pawn with
his queen, White would have won a
clear advantage after 1 2 c4 and 1 3
d5.
1 2 d4 X c5 � e7 x c5
13 ® d 1 -c2 !
This is the punch-line. If White had
immediately played 1 3 c4 then
Black would have exchanged
queens after 1 3 . . . de and 1 4
� x c4. Now White can play
c3-c4, but keeping his queen and
getting good attacking chances.
13 . . . h7-h6
Bl ack is forced to weaken his posi­
tion. Even less satisfactory would
have been 1 3 . . . g6.
14 c3-c4 d5 x c4
Black has no alternative but to take
the pawn at c4, for 1 4 . . . � e6 is
followed by 1 5 &! fd 1 , an unpleas­
ant prospect, whilst 1 4 . . . d4 loses
a pawn by 1 5 � e4 .
00g8-h8
You will agree that Black's mistake
on the 1 Oth move produced serious
consequences . White has forced
open the diagonals for his mighty
bishops and is poised for an attack
on the king. He is already threaten­
ing to play 16 ®g6 with a double
attack on g7 which cannot be par­
ried. Weak would have been for in­
stance 1 5 . . . � e7, which would
have been followed by 16 ®c3 � f5
1 7 e4 and the loss of a piece. To
protect himself against 1 6 ®g6
Black is forced to lose valuable
tempo enabling White to increase
pressure.
16 ®c2-e4
White's attack takes its own course.
Winning a tempo, the queen
changes over to the K-side.
16 . . . � c8-b7
Prob4bly the decisive mistake.
Black could have put up a stout de­
fence by 16 . . . �d7, or better still
1 6 . . . ®c7. But now his situation
1 29
is becoming hopeless because his � d4 he loses a piece by 22 ed !
white-square bishop cannot take � x f3 2 3 ® g 3 ! At this point
part in the defence of the king's Black's defence is based on threat­
fl ank. ening square g2. For instance 22
El x e7 ? would be weak because
17 El f1 -d 1 ® d 8 -c8
2 2 . . . � x e7 and mate is threat­
If 17 . . . ®e7 then 1 8 ® g4 f6 ened at g2.
( 1 8 . . . fS 1 9 ®g6 and the threat of
22 Jd. d7 X b7 !
20 ® :h6 + which cannot be par­
ried) 1 9 �h4 ! and Black already White sacrifices his queen and for­
loses material. ces victory. He could, of course,
1 8 � c4-d3 ! have won using a different, quite
deliberate and what is more a less
White forces Black to open his complicated approach, n amely by
king's defences . 22 ® g3 ! ® x g3 2 3 hg El ab 8 24
18 . . . f7-f5 g es ! g x e S 2 5 � x eS with a
1 9 ® e4-h4 double attack on b8 and e 7 . Despite
the fact that Black can save a piece
Threats follow in short order. Black by 25 . . . � f6 ! 26 � x b8 �c6,
has to defend himself against 20 White wins at least a pawn by 2 7
® X h6 + . � x a7 o r 27 El x a7 . White cannot
19 . . . ®c8 -e6 be reproached for sacri ficing his
2 0 � d 3-c4 ®e6-g6 queen and achieving his obj ective in
2 1 Etd 1 -d7 ! � c5-e7 this dramatic manner, but this ap­
proach calls for accu rate and far­
s ighted calcu lation.
227

22 . . . � e 7 x h4
23 g f3 X h4 ® g6-g5

The main variant 23 . . . ®h7 24

� dS ! Etac8 (24 . . g a s 25
.

White's rook has speedily pene­
trated into the seventh rank, and
Black is unable to cover square g7
in any other way. If he tries to or­
ganize a counterattack by 2 1
Et X g7 ® X g7 26 � X g7 +
00 X g7 27 � X a8 Et X a8 28
g x fs + , and wins) 2 5 � e6 ! Etce 8
26 � X f5 ! Et X fS 27 Et X g7 ! l:t b5 !
is much more interesting. Black
tries to evade the fatal discovered
check, but White gains a decisive
superiority in material after 2 8
g g6 + ® X g 6 2 9 El X g 6 Et X b2
30 Et X h6 + 00g7 31 Et X c6 .
24 f2-f4 ® g5-g4

1 30
If Black had im mediately played
23 . . . ® g4 then 24 f3 ®g5 25 f4
(and then continued as shown
above) he would have won .
25 � c4-e2!
Black resigned.
228
. rhe bishop is unassailable because
if 25 . . . ® x e2 then 26 � g6 +
and 27 El x g7 mate. On the other
hand if 25 . . 't!t:! x h4 then 26
.
� x g7 + 00g8 2 7 � c 4 + El f7 2 8
)d. x f7 Black cannot avoid sustain­
ing substantial losses in material,
c . g. 28 . . . b5 29 � a2 � aS 30
.� e6 ! El e 8 31 El e7 + cbh7 32
� f6 + El x e7 33 � x f5 + ! cbg8
.H � x h4 and White wins.
· !"he double attack is the leitmotif of
White's operations in all of these in­
tricate variants . lt still happens oc­
casion ally that even the world's best
g randmasters fail to anticipate a
d o uble attack early enough. The
following game is a good example.
King 's Indian Opening
f>o rtisch-Karpov
Moscow 1 977
1 � g l -f3 � g8-f6
2 g2-g3 b7-b6
Nowadays the world champion
likes to fianchetto the queen's
bishop.
3 � f l -g2 � c 8 -b7
4 0-0 e7-e6
5 d2-d3
Up till now Wh ite has concealed his
opening plans. Only at this point
does it become apparent that he in­
t �� ds to take up a King's Indian po­
smon.
5 . . . d7-d5
6 � b 1 -d2 � b 8 -d7
An interesting idea. If Black had
chosen the stock reply 6 . . . �e7
White could have played 7 e4 ! de 8
de. Taking the pawn by 8 . . .
� x e4 puts Black at a disadvan­
tage : 9 � e5 ! � d6 10 � x b7
� x b7 1 1 ® f3 ! with a double at­
tack on b7 and f7 (Kotchiyev-lva­
nov, M i nsk 1 976) . This is why the
world champion first deve loped the
queen's knight.
7 E H 1 -e 1
White i s planning t o make a thrust
with the pawn e2-e4. He could
have made this move first, but if 7
e4 then 7 . . . de 8 � g5 � c5 9
� g x e4 � f x e4 and it is plain sail­
ing for Black.
7 . . . � f8 -c5 !
8 c2-c4
White is disconcerted by the
bishop's move to cS and he drops
his planned thrust e2-e4, although
he could still have stuck to i t : if 8 e4
then for instance 8 . . . de 9 � g5 e3
131
l 0 fe � X g2 1 1 d/ X g2 � d5 1 2
� de4 ! and a satisfactory game for
White. But at this point the rook
move to e 1 proves pointless.
8 . . . 0-0
9 c4 X d5 e6 Xd5
1 0 � d 2-b3
The bishop at cS is apparently get­
ting on White's nerves and he de­
cides to drive it off, although b3 is
not exactly an ideal posting for the
knight.
10 . . . � c5-b4 !
This move brings out the weakness
of the rook's placing at el and wins
valu able time. Worse would have
d5-d4. But this way Portisch aban­
dons square e4 leaving it u nder the
controi of the world champion,
who does not hesitate to take ad­
vantage of the opportunity.
15 . . . � f6-e4 !
1 6 d4xc5 ?
It is not easy for White to find a
way out. He wants to weaken
Black's central pawns and forgets
for a moment his endangered
knight. This is enough to bring him
to the brink of ruin. His answer
should h ave been 1 6 a3 to clear up
the situation in the centre.
229

been 10 . . . � d 6 1 1 � bd4 bl e 8 1 2
� bs .
1 1 � c l -d2 a7-a5
Black's prospects wou ld also have
been good if he had exchanged the
bishops, but Karpov wants to com­
plicate the situ ation.
1 2 � b3-d4 E1 f8 -e8
13 E1 a 1 -c l c7-c5
16 . . . � e4 x d2
1 4 � d4-f5
1 7 � f3 X d2 tm d 8 -g5 !
White's knight crosses the demarca­
And here is the double attack !
tion line and loses contact with the
White's two knights are in danger.
remaining White forces, and White
They can protect each other by 1 8
now has to take great care that this
� e3 , but after 1 8 . . . E1 x e3 ! 1 9
unprotected knight does not give
a 3 � x d2 2 0 t!tt x d2 Black has
Black an opportunity to mount var­
more leverage in view of 20
ious tactical attacks .
.E1 X g3 !
14 . . . � d7-f8
1 8 � f5-d6
1 5 d3 -d4
He h as no choice but to concede
One can understand White's rea­
the exchange. And indeed, as you
soning : the enemy threatened to
will recall, the best defence against
restrict his position by playing

132
the double attack is another double
att ack. But of course not in this po­
. .
Sition .
18 . . . � b4 X d2
230
Portisch considered a number of
variants in quest for a way out of
this predicament, but to no avail.
I ,et u s also consider White's possib­
ilities :
(a) 1 9 � xe8 � xc 1 20 � d6 � d2
2 1 � X b7 � X e 1 ;
(b) 1 9 h4 ® h6 20 � x e8 (20 � fs
®f6) 20 . . . b:t x e8 2 1 c6 � a6 .
Karpov says that the variant 20
c;t) x b7 is stronger -because after
20 . . . � X e 1 2 1 &J X e l bl X e2 2 2
'f1t1 x e2 ® x c l + 2 3 d/h2 ! be 24
� x dS . .White has-a chance to
launch a counterattack. But Black,
1 oo, has a better reply in this case,
i . e. 20 . . . �xc1 ! 2 1 ®xc l
rl x e2 etc.
( c) 1 9 c6 � a6 2 0 f4 ®e7 2 1 � xe 8
&.t e 3 + ! and wins . In the end Por-
t isch struggled through to the text
1 1 1ove, which is not more enticing
1 han the other variant considered by
1 1 s . There is no way of making good
I he m aterial losses brought about by
1 he double attacks.
19 � d6 X b7 � d2 xe l
20 ® d l x e 1 b:t e 8 xe2 !
Another double attack.
2 1 ® e l x e2 ® g5 x c l +
2 2 ® e2-fl ® c l -d2 !
The final coup ! The queen covers
the pawn at d5 and controls squ are
c l , which the Black rook threatens
to occupy.
23 c5 x b6
If 23 c6 then 23 . . &j c8 24 ®b5
.
® c l + and 25 . . . ® x c6 wins .
23 . . . b:t a8-c8
In view of 24 . . &j cl White ad­
.
mitted defeat, for if 2 4 ®b5 then
24 . . . &j c2 ! and the threatened
m ate forces White's queen to turn
b ack.
Summarizing
remarks
on th e double attack
When direct contact has been made
with the rival forces and the pieces
of the two sides are attacked or
threaten to attack each other, this
indicates that the tactical situ ation
on the board h as come to a head.
Threats and attacks are the basic
elements of tactics. Just as a house is
made of bricks, so all tactical opera­
tions are made of these elements .
L e t us assume that an offensive op­
eration h as been carried out and
one of our pieces attacks one of the
opponent's pieces. Certain condi­
tions have to be satisfied in order
1 33
that the operation may prove suc­
cessful and the enemy piece be cap­
tured . First, this piece must have no
way of evad ing the attack and sec­
ondly, the opponent's pieces and
pawns must be unable to ru sh to its
aid. There must not be a single piece
or pawn that can support it or pro­
tect it agai nst the attack or at least
to block the line of fire of the at­
tacking piece. In practice, such situ­
ations occur only in exceptional
cases when there are many pieces
on the board . Usu ally there are
pieces and pawns around, which
can provide assistance in time, i . e.
on the next move. For this reason
we can j ustifiably say that a simple
sin gle attack is not effective
enough. In normal tactical condi ­
tions there arc ample means of de­
fence a \·ailable. But what attack is
effective the n ?
Apart from castling one can move
only one piece or one pawn at a
time. This means that it is extremely
difficult to protect two or more
pieces or to get them out of danger
if they are attacked. I believe we are
j ustified in s aying that an attack i n
which o n e o r several pieces attack
two of the opponent's pieces (this is
usually referred to as double attack)
is a more effective offensive device
than an isolated single attack. This
first conclusion is very important
for us.
In the theory of the middle game ,
some cases of the two-fold attack.
e. g. the fork, the discovered check,
the simultaneous attack of two
pieces on two enemy pieces etc., are
seen as different tactical elements .
1 34
As a ru le, they are considered separ­
ately. But for us the exact opposite
is relevant, for in essence all these
operati ons are one and the same
th ing, namely a simultaneous attack
on two enemy pieces.
But here, in the case of the two-fold
attack, we should include another
method of attack, namely the simul­
taneous attack by two pieces on one
piece or on some important square
in the enemy camp. In the theory of
the middle game there is also a spe­
cial case of such an attack that is re­
gard ed as one of the tactical ele­
ments , namely the double check, in
which the attack by two pieces is di­
rected against the enemy king. It is
well-known from experience that
this device is extremely effective. It
can also be used with good resu lts
in attacking any other piece or in
threatening important squares in
the enemy camp. The important
feature of this kind of two-fold at­
tack is that usually the only way the
piece being attacked can be saved is
by flight.
It is not necessary to demonstrate
that such strong threats as mating
or queening threats are no less
dangerous than attacks on pieces.
This means that by combining any
one of these threats with an attack
one can get an attack that is as ef­
fective as the two-fold attack. This
conclusion is also very significant
for us. If we pursue this thought still
further we inevitably arrive at the
conclusion that by combining these
two threats we obtain an attacking
device that is just as effective. Thi s
is t h e third i mportant conclusion. It
 is only logical to go one step further
; t nd to include all threats, namely
1 he major and the minor ones, e. g.
1 hreats to immobilize an enemy
piece, to occupy a square in the ad­
versary's camp, to isolate the oppo­
nent's king, etc. To put it briefly, we
hold that a combination of any two
threats is an effective offensive in­
strument. This is our fourth conclu­
Sion.
Thus, we proceed from the assump­
tion that two-fold attacks, attacks
and threats and any double threats
are one and the same thing, so far as
their mechanism is concerned,
namely a combination of two ele­
ments of attack, which in its most
general form we refer to as double
attack.
We are of the opinion that the
double attack is a particu larly effec­
tive and important means of of­
fence. We have made a thorough
examination of the double attack
on the preceding pages . If we con­
sider the numerous positions in
which the double attack occurred,
we inevitably arrive at the conclu­
sion that this device is tru ly univer­
sal and a multi-faceted tactical
method not only of attack but also
of defence . We have seen that the
best defence against a double attack
is a double attack. We h ave also ob­
served that defensive moves leading
up to a double attack have a more
lasting effect th an ordinary moves,
which prompted us to refer to them
as moves with a particularly lasting
effect.
We have also examined a number of
complex situations in which triple
attacks, two-fold double attacks
and variou s other operations oc­
curred. The mechanism was always
the same, for their common feature
was the double attack. And lastly
we should like to stress that the
double attack in its most general
form can have different objectives,
winning material being only one of
them.
A. Anderssen, 1 842
231
Wh ite to p lay
This is how White elegantly avails
himself of the double attack : 1
® e6 ! It consists of two mating
threats (one being direct, the other
covert) . The first threat of attack by
two pieces on square f7 cannot be
parried by taking the queen because
this leads to mate in two moves :
1 . . . fe 2 � x e6 + Cbfs 3 rlh8
mate. But neither can Black defend
himself by 1 . . . .EI x b 3 , because
this is where the second mating
threat comes into operation : 2
rl h8 + ! CD X h8 3 ® h6 + Cbg8 4
® x g7 mate .
1 35
Engels-Mar6czy
Dresden 1 936
232
White to play
In this position White mounts a
double attack whose objective is the
promotion of the pawn : 1 b;l x b2 !
't!tJ X b2 2 't!tJ X c8 + ! � X c8 3 d7 \1lg2 � e2 and is saved by a double
and the pawn's advance cannot be
checked.
Even a draw by stalemate can be the right?
objective of a double attack.
A. Troitzky, 1 895
233
Draw
What can White do ? Since he can­
not keep the pawn from queening
he has to look for a feasible solu­
tion. Again it is our universal
1 36
method, the double attack, that
helps White in his plight.
1 B fs + . . .
The king has two ways to move .
First, let us see what happens if it is
played to the left :
1 . . . 00 e 7 2 B e S + 00d6 3 B e 1 !
� X e 1 4 \1lh3 ! . . .
And already we have the first
double attack. White attacks the
pawn, and if it reaches the queening
square and queens (or is promoted
to a rook) there is a stalemate. This
means that Black has to convert the
pawn into a knight, but in this case
after 4 . . g 1 ( � ) + . White plays 5
.
attack 6 OOf l .
But what if the king goes to the
,-6--..
1 . . . wg7 2w
,-6--.. h 3 ! . . .
Again a double attack : the threat
against the pawn is coupled with a
concealed threat of a stalemate.
2 . . . g l ( ® ) 3 El. gS + ! (the double
attack against the king and the
quee n forces Black to take the
rook) 3 . . ® x gS (or 3 . . .
.
� x gS) stalemate !
There are, of course, numerous dif­
ferent kinds of draw that can be the
objective of a double attack. We
s h all consider only one case in
which a draw is achieved by block­
ing the opponent's king.
V. Chekhover, 1 954 fence we shall conclude the first
part of this book by giving you
234 some exercises to enable you to
practise on your own .

Exercises
for inde p endent
study
In each of the following positions
one of the two sides can, unless
stated otherwise, emerge victorious.
White to play You should try to find the solution,
i. e . to seek out, on your own, to
To prevent the pawn from queening track down the double attack that
White is forced to sacrifice his rook. leads to victory or to winning mate­
1 Ei b l cb (®) 2 � x bl. . . rial. The caption below each dia­
gram states whose move it is. Please
But this puts him i n a critical situa­ note that the exercises are arranged
tion, because Black has a double at­ in order of increasing difficulty.
tack lined up consisting of an attack That is why it is advisable to work
and a queening threat. through the exercises in the order in
2 . e3 ! which they appear. We believe that
by solving these problems on you r
. .

There seems to be no way out of own you will acquire a sense for the
this dilemma. And yet there is a de­ double attack and that this wil l help
fence, n amely with the aid of the you perfect your tactical profi­
battle-tested double attack ! Ciency .
.� � x fs ! e2 4 � g4 ! ! . . . Exercise 1
' I 'his
is an effective retaliation.
235
White attacks the pawn and threat­
ens, after 4 . . . e 1 (®) to lock
Bl ack's king in a cage and throw
away the key by 5 h 3 . But if the
bishop is taken then the king can
s top the pawn without assistance by
�1 f3 + d)h4 6 d)f2. To enable you
t o acquire full mastery of the art of

d ouble attack and to acquaint you

iu detail with this important and ef-
fective meth od of attack and de- Black to play

1 37
E x e rc i s e 2 Exercise 5

236 239

Bbck to p l ay W h i te to play

Exercise 3 Exercise 6

237 240

Black to play B l ack to play

Exercise 4 E x e rc i se 7

238 241

White to play White to play

1 38
l ·: xercise 8' Exercise 1 1

242 245

White to play White to play

l�xercise 9 Exercise 1 2

243 246

Black to play White to play

Exercise 1 0 Exercise 1 3

244 247

Black to play White to p l ay

1 39
Exercise 1 4 Exercise 1 7

251

White to play Black to play

Exercise 1 8
Exercise 1 5
252
249

White to play. May he embark on the

White to play
double attack 1 bl g4 ?

Exercise 1 6 Exercise 1 9

250 253

White to play White to play

1 40
hcrcise 20 Exercise 2 3

254 257

White to play Black to play

J·:xercise 2 1 Exercise 24
255
258

White to play. He played 1 � gS . Was

White to play
the move correct?

Exercise 22 Exercise 2 5

256 259

White to play White to play

141
Exercise 26 Exercise 29

260

B lack t o play White w play

Exercise 27 Exercise 30

261 264

B lack t o play White t o play

Exercise 2 8 Exercise 3 1

262 265

White t o play White to p lay

1 42
1 :. x c rcise 32 Exercise 35

266 269

W h ite t o play White t o play

l·� xercise 3 3 Exercise 36

267 270

White t o play White to play

I ·:xercise 34 Exercise 37

268 271

White t o play White to play

1 43
Exercise 3 8 Exercise 4 1

272 275

Black to play White to play

Exercise 39 Exercise 42

273

White to play White to play

Exercise 40 Exercise 43

274 277

White to play White to play

1 44
h c rcise 4 4 Exercise 47

278 281

l l l ack t o play Black to play

l ·: xercise 4 5 Exercise 48

279

White to play White to play

Exercise 46 Exercise 49

280 283

White to play Black to play

1 45
Exercise 50 Exercise 5 3

284 287

Blac k to play White to play

Exe rcis e 5 1 Exercise 54

288
285

White to play· Black to play. May he play 1 . �d3 2

. .

&:1 x d3 � x f4 with a double attack?

Exercise 5 2 Exercise 5 5

286 289

White to play White to play

1 46
Exercise 56 Exercise 59

290 293

White to play. How is the game going to White to play

end ?

Exercise 6 0
Exercise 5 7
294
291

Black to play
White to play

Exercise 58 Exercise 6 1

292 295

White to play White to play

1 47
Exercise 62 Exercise 65

296

Wh ite to play
Wh i te to play

Exercise 66
Exercise 6 3 300

Black to play. M ay he play l . . .

� c X dS 2 � c X d S � X d S 3 � X g7
W h i t e t o p l ay ® X g7 4 � X dS ® d 4 + and capt u re a
paw n ?

Exercise 64
Exercise 6 7
298 301

Wh ite to play Black to play

1 48
Exercise 6 8 Exercise 71

305

White to play. What are his means of de-

Black to play. He played 1 . �eS !
fence against the threat 1 h3?
. .

How is White t o continue ?

· · ·

Exercise 6 9
Exercise 7 2
303

306

Black to play. May he play 1 . � e3 i n

. .

order t o w i n the knight d2 after 2 f e � X White to play

e3 + 3 1:l f2, by means of 3 . . . � d S ?
Exercise 7 3
Exercise 70

304 307

White to pl ay White to play

1 49
Exercise 74 Exercise 77
31 1
308

White to play. How is the game going to

White to play
end ?

Exercise 75
Exercise 78
309
31 2

Black to play
White to play
Exercise 76
310 Exercise 79

313

White to play. He may not take on bS

because of 1 .E1 x bS, followed by
. . .

2 . ®b6 + . What will he opt for?

. . Black t o play

1 50
Solutions
No. 1 Black wins : 1 . . . El. e4 +
(Marski-Yudovich, Minsk 1 937).
No. 2 Black wins : 1 . . . d4 2
� x d4 El. e2 + (the simple line
2 . . . � X d4 3 t!f1 X d4 *g2 + fol­
lowed by 4 . . . ® x h 1 is also possi­
ble) 3 � x e2, � e4 + (Rauser-Il­
yin-Genevski, Tbilisi 1 93 7) .
No. 3 Black wins : 1 . . . d3 ! 2 cd (2
hg de 3 ® x e2 ® d4 + ) 2 . . .
� x e2 3 ® x e2 ®d4 + (Taube­
von Henning, Bad Niendorf 1 934). ®g6 3 � x b6 � x b6 4 ® c5 (Ryu­
No. 4 White wins : 1 � d5 ed 2
� X f6 (Uhlmann-Schwartz, Stral­ No. 16 White wins : 1 � f6 + gf 2 ef
sund 1 97 5 ) .
N o . 5 White wins : 1 El. x d 6 ® x d6
2 e5 ! (Ahues-Ktirptihn, 1 93 5 ) .
N o . 6 Black wins a pawn : 1 . . .
g X h3 + 2 t!f1 X h3 � X c3 and if 3 cd (2 t!f1 X d3 �g5 + 3 rl d2
El. x c3 then 3 . . . � e2 + (Htib­
ner-Tal, Biel 1 976) .
No. 7 White wins : 1 � c6 !
(L. Steiner-Purdy, Sydney 1 937).
No. 8 White achieves a substantial
advantage : 1 d6 ! ® x d6 (1 . . .
� x d6 2 c5) 2 � x f5 ! (Sosin­
Chistyakov, 1 937).
No. 9 Black wins : 1 . . . � e7 ! (Kly-
atskin-Yudovich, Moscow 1 937).
No. 1 0 Black wins : 1 . . .
* X h2 + ! 2 d) X h2 � f3 + 3 00h 1
� g3 mate.
No. 1 1 White wins : 1 rl x e6
EI x e6 2 b6 + ! dl x b6 3 EI h6 !
(B. Horwitz, 1 87 3 ) .
No. 1 2 White wins : 1 � f7 + !
El. x f7 2 � eS. In the game played
by Gilg v. Alekhine, 1 9 32, White
missed this opportunity.
No. 1 3 White wins : 1 �b2 rl f8 2 2
Etc7 + digS 3 El. g7 + d/hs 4 00a2 !
or 1 . . . Et h6 2 rl g3 + 00h7 3
Etg7 + 00h8 4 ebb 1 ! (R. Bian­
chetti, 1 925).
No. 14 White wins a pawn : 1 �e7 !
� x e7 2 El. x eS (Krogius-Gauf­
fin, Helsinki 1 937) .
No. 1 5 White wins : 1 ®c3 Et c8 2 f6
min-Makogonov, 1 934).
and there is no defence against the
two threats 3 ® x f8 + 00 x f8 4
rl d8 mate and 3 ®g4 + (V. Vu­
kovic, 1 95 1 ) .
No. 1 7 Black wins : 1 . . . bl x d3 ! 2
® x b2 + 4 d/ d t ® a t + followed
by 5 . . . ® x h 1 ) 2 . . . �g5 !
(Shamayev-Ufimtsev, Leningrad
1 949) .
No. 1 8 By 1 rl g4 White loses the
exchange on account of 1 . . .
� e2 + ! 2 s:':1. x e2 ® x e3 + 3 d/ht
fg (Rubtsova-Belova, Moscow
1 945).
No. 19 White wins : 1 g7 �d5
( 1 . . . � h7 2 � d3) 2 c4 ! � x c4 3
� f3 ! � e6 4 �g4 (H. Rinck, 1 907).
N o . 20 White wins a pawn : 1
EI x g6 ! hg 2 � x eS, now the rook
is being attacked and 3 ® x g6 ! is
threatened (Sakharov-Rovner,
Lvov 1 9 5 1 ) .
151
No. 2 1 1 � g5 was followed by
1 . . . � X g3 ! 2 � X d5 � X e2 + 3
00g2 � x d5 + 4 f3 � x cl with a
decisive advantage (Monostori­
Florian, Budapest 1 9 50) . Although
2 e4 � e2 + 3 00 h 1 ®e5 4 ®e3
� f4 would have been somewhat
better.
No. 22 White wins : 1 � h5 ! ® d 7 2
®g3. In the game played by Tshi­
gorin-Mason, Hanover 1 902, the
following line was taken : 1 .El fh 5
-t!tJ X f6 2 ,El X h7 + � X h7 3
.El x h7 + digS and Black won .
N o . 2 3 Black wins : 1 . . . .El b 1 2
� X d6 f3 + ! 3 d) X f3 .El X b3 +
followed by 4 . . . ed (Rossetto­
Sherwin, Portoroz 1 9S S ) .
N o . 24 White wins : 1 � aS ! a 1 (®)
2 .El a6 + OObS 3 � c6 + followed
by 4 .El x a1 (conclusion of a study
by Platov brothers, 1 929).
No. 2 5 White wins : 1 ®d3 ! ®cl 2
® a 3 + 00c4 3 b3 + or 1 . . . ® a 1 2
®c3 + 00a4 3 b3 + (conclusion of
a study by L. Kubbel, 1 929) .
No. 26 Black wins : 1 . . . .E1 g2 ! 2
® x d5 .El x h2 + followed by
3 . . . ed (Bellon-S. Garcia, Cuba
1 976).
No. 2 7 Black wins : 1 . . . � h4 2
®e2 (2 ®e4 ® d 1 + 3 ® e 1
®f3 + ) 2 . . . ® b 7 + 3 OO g t .El d2 !
4 -t!tJ X d 2 � f3 + 5 00f2 � X d 2 6
.El c7 ® f3 + 7 00e 1 ®e3 + 8 00d 1
� c4 (Reefleger v. Wirthensohn,
Hanover 1 976) .
No. 28 White wins a pawn : 1
.El X d 7 ! � X d7 ( 1 . . 'dfJ X d7 2
.
1 52
� x e5) 2 � gS ®bs (2 . . . � c6 3
� x c6 ® x c6 4 � ce4 !) 3 � x aS
® x a S 4 � x h7 ! (Alekh ine
v. Euwe, 1 9 37).
No . 29 White wins a piece : l ®f3 !
® x f3 ( 1 . . . � d 5 2 � x e7
� X e7 3 � e4 or 1 . . . ®d7 2
®h3 !) 2 gf � d 5 3 � x e7 � x e7 4
.E1 d 7 . In the game between Ahues
and List, 1 93 2 , White missed the
opportunity and played 1 ® f 1 ?
No. 30 White wins : 1 � X c6 be 2
� X d5 d/h8 (2 . . . cd 3 � X dS +
followed by 4 � x a8) 3 � x e7
® x e7 4 ®b4 _.E1 f6 5 .El x c6
(Pogrebysski v. Kortschmar, Kiev
1 93 7 ) .
N o . 3 1 White wins : 1 � g6 ! In the
game between Tartakover v. Sultan
Khan in 1 932, White did not hit
upon this move .
No. 32 White wins : 1 ®b2 ! 00h3
( 1 . . . g4 2 d4 ®g5 3 ® h2 mate) 2
d4 ® d6 3 ®g2 + 00h4 4 � fS mate
(A. Troitzky, 1 897) .
No. 33 White wins : 1 .El X b6 � b4 !
2 � f7 ! � e S 3 00 X b 4 � X f7 4
OOcS dla7 5 dlc6 � d 5 + 6 00c7
(conclusion of a study by M. Libu r­
kin, 1 9 3 1 ) .
No. 34 White wins : 1 .El x h7 + !
00 x h7 2 ® e7 + 00g6 3 E1 g8 +
OOfs 4 .E1 x gs + oo x g5 s ®g7 +
00f5 6 ®d7 + (Duras v. Cohn,
Karlsbad 1 9 1 1 ) .
No. 3 5 White wins : 1 � x d 4 cd 2
'dfJ X b7 + E1 X b7 3 ,El X e S ! d) X eS
4 � x c6 + .E1 d 7 5 00e2 00e7 6
� x d7 00 x d7 7 00d3. In the game
played by Ragozin v. Alatortsev,
Tbilisi 1 9 37, White missed this op­
portu nity.
No. 36 White wins : 1 ® x g6 + �
� X g6 2 � X g6 00 X g6 3 g4
Et h2 + 4 cbg3 EI d2 s � x fs +
\"bf6 6 � x d7 (Spielmann v. Hen­
linger, 1 937) .
No. 37 White wins : 1 b6 � x e4 +
2 00h4 ! � d 6 3 � X d6 + 00 X d6 4
\!lgs ! � e7 5 b7 cbc7 6 d6 + ! (con­
clusion of a study by L. Kubbel ,
1 929) .
No . 38 Black wins : 1 . . . . ®b 1 + 2
EI c 1 ® x c 1 + ! 3 � x c 1 EI x c6 4
� d 3 E! c l + 5 � X c l � X a8 (Mi­
ken as v. Aronin, Moscow 1 950).
No . 39 White wins : 1 EI x f7 ! de 2
E! g7 + 00h8 3 EI X d7 00g8 4
EI g7 + 00h8 5 EI e7 + and mate i n
two (Schiffers v . Steinitz, 1 8 96).
No. 40 White wins : 1 ® x bS �
® x b5 2 d5 + E! g7 3 ef h6 4 � d4
etc. (Capablanca v. Corso, Havana
1 900).
No. 4 1 White wins : 1 EI x g7
� x c2 2 EI g6 + 00h7 3 � x c2
� c4 4 E! g7 + 00h8 5 EI h7 + 00g8
6 E! h 8 mate (P. Frydman v. Vu­
kovic, Podjebrad 1 9 36) .
No. 4 2 White wins : 1 ® h6 ® f8 2
'<!f:t X h7 + 00 X h7 3 hg + 00 X g6 4
�e4 mate (Fischer v. Mj agma­
suren, Sousse 1 967) .
No. 43 White win s : 1 � g7 + !
00 X g7 2 � e8 + ! <1Jh6 3 ®f4 + gS
4 ® f6 + <11 h s s � g7 + cbh4 6
®f2 mate (Mabs v. Alexander,
London 1 96 1 ) .
No. 4 4 Black wins : 1 . . . ®e7 2
®hs ®e3 + � (not 2 . . . ® x e2 be-
cause of 3 EI x f8 + 00 x f8 4
EI fl + 00g8 5 ®f7 + and mate in
two) 3 cbh 1 ® x e2 (Mattison
v. Vukovic, Debrecen 1 925).
No. 45 White wins : 1 � e4 ®e6 2
� fS g6 3 EI x c8 EI x c8 4 � e7 +
(Eiiskases v. Muffang, 1 9 35).
No. 46 White wins a pawn : I
� x f7 + , because 1 . . . EI x f7 is
useless because of 2 ® d 8 + EI f8 3
EI x g7 + dJ x g7 4 ® x aS (Lo­
wenfisch v. Ilyin-Genevski, Tbilisi
1 9 37) .
N o . 47 Black wins : 1 . . . � g 3 � and
if 2 cb x g3, then 2 . . . � x e 1 +
followed by 3 . . . EI x b 1 (Dzagu­
rov v. Simagin, Moscow 1 936) .
No. 48 White gains superiority in
material : 1 � x eS bS (1 . . .
� X h3 + 2 dJ X h3 'diJ X e5 3
EI x c6 ! ®hs 4 00g2 be 5 ®d4 + ) 2
'diJ a l ! � X h3 + 3 dJ X h3 '<!f:t X e5 4
EI x c6 ® hS + 5 00g2 EI x c6 6
b4 + � 00g8 7 ba (Hubner v. Rogoff,
Biel 1 976).
No. 49 Black wins : 1 . . . ®d1 !! 2
� x b6 E! c l (Arnold v. Duras,
Prague 1 920). 2 � eS e2 + 3 cbh 1
® x e 1 + � 4 ® x e 1 � f2 � would
not have helped either.
No. 50 Black wins : 1 . . . ®h7 + 2
E! h3 .El d t + 3 dlh2 Ei h 1 + � 4
00 X h 1 (4 \t>g3 .E1 X h3 + 5 gh
®h4 mate) 4 . . . '<1!1 x h3 + 5 ddg 1
® x g2 mate.
No . 5 1 White wins : 1 Ei b 3 � (Kla­
man v. Lisitsin, Leningrad 1 9 37).
1 53
No. 52 White gains advantage in
material : 1 .Id x eS ®d4 + 2 d/h 1
.Id d8 3 .Id x d S ! (Sakharov v. Pav­
lenko, Lvov 1 96 1 ) . Better is 1
.Id d 8 .
No. 5 3 White wins : 1 n X e 6
� X d4 + 2 '® X d4 ! r:l X d4 3
,Id X g6 + hg 4 r:l X d4 (Botvinnik
v. Szab6, Groningen 1 946) .
No. 54 1 . . . � d 3 2 '® X d3 � X f4
is weak because of 3 � f7 ! � x e3
(3 . . . 00 x f7 4 0-0 or 3 . . .
-® x f7 4 .Id f 1 ) 4 � x h 8 (Taimanov
v. Sakhodyakin, Moscow 1 945).
No. 5 5 White wins : 1 n X c7 � d 8
2 � X g 6 + ! d/ X g 6 (2 . . . d/g7 3
.Id x f7 + dl x g6 4 � e S + OOhS 5
g4 + 00h4 6 cl'lb3 etc.) 3 cl'lb 1 +
OOg7 4 .Id b7 � f6 5 ®f5 � gs 6
� x gS, and Black resigned (Duras
v. Nenarokov, Petersburg 1 909) .
No. 56 After 1 n X b 7 ! e4 ! 2
-® X a4 � X h2 + the game ended
in a draw (Simagin v. Aronin, Mos­
cow 1 947) .
No. 57 White wins : 1 � x h7 + !
00 X h7 2 ®h5 + d/g8 3 ,Id X f7 !
® x f7 4 g6 (Kasparyan v. Che­
khover, match 1 936).
No. 5 8 White wins : 1 � c7 ! ! n X cl
2 cl'lb7 + rl x b7 3 rl x cS mate or No. 6 6 Black loses his queen after
1 . . . -® x c7 2 EI x c5 + ® x cS 3
® b7 + 00 X aS 4 EI a 1 m ate (Tar­
rasch v . Allies, Naples 1 9 1 4) .
N o . 59 White wins : 1 ® b 7 + !
� X b7 2 � X b7 + 00a5 3 �b4 +
d!bs 4 � c3 + dies s � x d4 +
,Id X d4 6 .!d X d4 d/ X d4 7 � X c8
(Lyublinski v. Nechiporovich,
Tsheboksary 1 950).
1 54
No. 60 Black wins : 1 . . . e4 ! 2
� x e4 (2 � e2 gS) 2 . . � x e4 3
.
® x e4 gS ! 4 � f3 � f5 (Napolitano
v. Batik, correspondence chess
game 1 95 8 ) .
No. 6 1 White wins : 1 r:l x g7 ! , and
if 1 . . . dl x g7 then 2 ®g3 + dif8
3 � g6 + (Marshall v. Kupchik,
Chicago 1 926) .
No. 62 White wins : 1 �b7 ! � x b7
2 a 8 ( -® ) � x a8 3 f8 (®), e. g. 3 . . .
� b7 4 ® b 8 ; 3 . . . � c6 (or 3 . . .
� dS) 4 cl'l d6 ! ; 3 . . . � e4 4 cl'l f4 !
(E. Pogosyams, 1 97 6) .
No. 63 White wins a pawn : 1
� X f7 + !, and 1 . . . ® x f7 cann ot
be played on accou nt of 2 EI d8 +
dlg7 3 cl'lc3 ® f6 4 ®c7 + etc. (Yu ­
dovich v. Chekhover, Leningrad
1 934) .
No. 6 4 White wins : 1 ®h 1 + ! dlc7
2 ® c l + did8 3 ®gs + dlc7 4
® x b5 did8 5 ®gs + dlc7 6
® a S + dlc6 7 ®c3 + (E. Pogosy­
ants , 1 976) .
No. 65 White wins a pawn : 1
� x c2 ! � x c2 2 � a6 ! � x a 1 3
� X c8 cl'J a7 4 � X d7 + d!J X d7 5
E1 x a 1 (Flohr v. Stahlberg, Zurich
1 934) .
1 . . . � c X d5 2 � c X dS � X d5 3
� X g7 'l1f:t X g7 4 � X d5 d!J d4 + 5
d/h 1 ® x dS 6 � c4 ! (Flohr
v. Bronstein , Parnu 1 947) .
No. 67 Black wins : 1 . . . h 5 ! 2 ®e2
c2 ! 3 ® x c2 -® h 1 + 4 dig3 h4 + 5
dif2 ® h 2 + followed by 6 . . .
® x c2 (Snosko- Borovski v. Ale­
khine, Petersburg 1 9 1 3) .
No. 68 The only defence White has No. 74 White wins : 1 .b! g2 � c l 2
is 1.b! d4 ! .b! es (1 . . . h3 2 ® d s) 2 El g 1 � f4 (2 . . . � h6 3 � x d6 +
®a6 ! hg 3 hg El h5 4 El h4 (Reti \1l X d6 4 ,b1 g6 + ) 3 � X d6 +
v. Mar6czy, New York 1 924) . d;l x d6 4 \1le4 � h 2 5 E1 g2 �eS 6
No. 6 9 1 . . . � e3 i s a bad move. El g6 + (J. Gunst, 1 927).
Bl ack found himself hemmed in af­ No. 7 5 Black wins : 1 . . . � f4 ! 2
ter 2 fe � x e3 + 3 � f2 ®ds 4 ,b1 X f4 � X f4 3 (!iJ X f4 (!i1 b 1 + 4
cbf1 � x d2 5 ®c2 and lost a piece ® c l (!iJ X c 1 + 5 � X c l ,b1 a8 !
because 6 e3 followed by 7 \1le2 is (Vikman v. JovCic, correspondence
inavoidable (Flohr v. Makogonov, chess game 1 95 5 ) .
Parnu 1 947).
N o . 76 The correct reply is 1 b4 !
No. 70 White wins : 1 .b! e7 ! and White wins after 1 . . . � X b4 2
then e. g. 1 . . . � x d4 2 .b1 x f7 ! � x bs .b1 x b s 3 � x bs ®b6 + 4
� X f2 + 3 \1lh 1 El X d 1 + 4 \1lh 1 ® X bS 5 tl b 1 ! ·e. g. 5 . . .
E1 x d 1 ® x c4 5 .b1 x g7 + and � a6 (5 . . . ® c4 6 ® a4) 6 (!i1d6 !
mate in two. If 1 . . . �e6 then 2 (Alekhine v. Chajes, Karlsbad
� X e6 ! El X d 1 + 3 El X d 1 fe 4 191 1).
E1 d8 + etc. (Sosin v. Nekrasov,
Moscow 1 93 1 ) . No. 7 7 White won after 1 � e4 !
r:1 X e7 2 � X f6 + \1l h 8 (2 . . .
No. 7 1 The winning move i s 2 d/f8 3 � X h7 + \tlg8 4 � f6 +
®d6 !, e. g. 2 . . . � f2 + 3 \1lh4 \1lf8 5 � x e7 gf 6 � X c6 ®e8 7
gS + 4 \1lhs, and the attack is re­ ®b4 aS 8 (!i1 c3) 3 � x e7 ® x e7 4
pulsed. In the game played by ®e4 ! ® x e4 5 � x e4 (Alekhine
Smyslov v. Petrosyan in a candi­ v. Kimura, Tokyo 1 9 33). If 1 . . .
dates' tournament in 1 95 3 , events � x e4 then 2 ® x e4 ! (not 2
took the following course : 2 E1 x d7 because of 2 . . . ® x d7 3
® x d3 cd 3 d 8 (®) draw. r:1 X d7 � X c3 ! 4 � h6 + \tlh8 5
No. 72 White wins : 1 cbb4 + \1ld5 � x f7 + with perpetual check)
2 e4 + cbd4 3 ® a2 ! \1ld3 (3 . . . 2 . . . � X f5 3 E1 X d8 r:1 e X d8 4
et> x e4 4 d3 + or 3 . . . (!iJeS ®c4 ®e2, and White has the advantage.
mate) 4 ® c4 + cb x d2 5 ®a2 + .
No. 78 White wins : 1 ®g3 ! !
The same thing h appens if 1 . . .
(!iJ X h6 + 2 ® h3 ® d6 3 \1lh 1 \1lg8
� c3 2 ® x c3 + \1lds 3 d 3 ! ® x e2
4 E1 x e8 + \1lf7 5 El h 8 (Duras
4 ® c4 + \1le5 5 d4 + (L. Kubbel,
·v. Spielmann, Piestany 1 9 1 2) .
1 93 4 ) .
N o . 79 Black wins : 1 . E1 X b6 ! 2
N o . 7 3 White wins : 1 ® a2 + \1lb4
. .

ab ® cs + 3 \1lh 1 � c6 ! 4 ® as ef ! 5
2 ®b2 + \1lc4 3 ® c2 + \1lb4 4
� X f3 (!iJ X c2 6 'f!tt X a6 � X f3 7 gf
cbb2 ! ® d s 5 ® a4 + ! ! d;l x a4 6
� hS etc. (Lj ubojevic v. Karpov,
� c3 + \1lb4 7 � X dS + ebbS 8
Manila 1 976) .
� c7 + \1lc6 9 � X a8 (L. Kubbel,
1 924).

1 55

Part 2
The combination
Co-ordination
of piece moves
The simplest example of the co-or­
dinated movement of pieces is the
two-fold attack. The important
thing is not whether one or two of
the enemy's pieces are attacked, but
that the piece or pieces are attacked
by two pieces simultaneously. The
co-ordination expresses itself in this
simultaneity. But we shall not go
into this question here because we
have already dealt with the two­
fold attack in Part 1 of the book in
detail. But the two-fold attack is not
the only possible case of co-ordi­
nated mo"tl'ement of pieces. Let us
consider some other possibilities .
Chistyakov-Simagin
Moscow 1 93 5
314
White t o play
By sacrificing the exchange by 1
.El x c6 .El x c6 2 � x d5 White en­
tangled his opponent in a double at­
tack. Black's only satisfactory
answer is 2 . . . .El c 8 . This is fol ­
lowed b y a second double attack by
3 �e6 . The only reply is 3 . . .
.El c6. The double attack has not
ach ieved the objective yet because
B l ack can move the rook away and
keep the knight covered . But White
has an extraordinari ly strong move :
4 � d7.
315
Black t o play
On the one hand this is a simple at­
tack, the white-square bishop at­
tacking the rook and depriving it of
squares c8 and e6 . On the other
hand, the bishop is supported by its
black-square comrade. Although it
does not end anger the rook i tself, it
does cut it off from squares c7 and
d6. The rook has no place left to
hide. B lack has to reply by 4 . . .
.El d6, and after 5 � X d6 White has
a won end-game. Here we have
seen the case of the co-ordinated
movement of pieces in which one
piece attacks an enemy piece, while
another restricts its sphere of ac­
tion. A situation arises in which a
1 57
simple attack proves irrefutable.
Strictly speaking, the pawn at d4
(or the pawn at b4) which deprives
the rook of the square c5 also takes
part in the attack on the rook. But
the Black pawns at b6 and f6, which
restrict the rook's movement on the
sixth rank, and the knight at c4,
which blocks the c-file, also play an
important role. S uch cases of co-or­
dinated play in which one piece at­
tacks and another one supports it
are particularly frequent in end­
games.
Here is one of the oldest end-games
A. Troitzky, 1 923
on record.
Sairab, 9th century
316
White to play and win
White solves the problem of captur­ this can be done.
ing the knight in the following man­
ner : 1 bl e3 ! Black has no choice but
to play 1 . . . � g l . The knight is
immobilized now, but it still has to
force the king to move to g4 in or­
der to cut it off from squares f3 and
h3 in order to attack it with the
rook from the first rank. If he tries
to do this by a direct approach by 2 6 dle7, and the f-pawn is the first to
dlf4 he fails to achieve co-ordi-
158
n ated play of the king and the rook
after 2 . . . d:>d4 ! e. g. 3 &l e t
� h3 + 4 d:>f3 � g5 + 5 \t>f4
� h3 + with a draw. Only 2 d:>f5 !
did 4 3 <ilf4 dlc4 4 d/g4 and 5 bl e 1
leads to victory. This example
shows that as early as a thousand
years ago chess players recognized
the significance of the co-ordinated
play of pieces .
The co-ordinated action of pieces is
essential in chess. In the present
b ook we try to show this i n the most
general form.
31 7
White to play and win
From this position White m anages
to capture Black's bishop, although
one does not realize right away how
1 f6 ! . . .
White blocks the bishop's escape
across f6.
1 . . . gf
If 1 . . . g6 then 2 d:>b7 � d 8 3 � d 4
CQd2 4 \t> c 8 \t>d3 5 CQ X d 8 CQ X d4
go to the queen .
 ) \t>b7 � d 8 3 ct>c8 � e7 4 cbd7
.<.:Z.f8
· I 'he bishop wants to flee across h6. erations so well that they do not
Time is ripe for White's second
piece to go into action.
�'i �e3 + dld 1 6 dieS �g7 7
\�) X f7 � h8 8 dlg8
l 'he king has caught up with the
·
fleet-footed bishop after all. In this
game the main burden rested on the
king's shoulders , but White's bishop
also contributed its share by not al­
lowing its opposite number to move
to h6.
The next example shows another
case of the co-ordinated operation
of pieces.
E. Pogosyants, 1 961
318
White to play a n d win
How can White win here ? The
pawn at d6 is hopelessly weak, and
if White covers it by 1 .bi d5 then
1 . . . dle6 2 .bi d 1 � h6 and it is lost W hite to play and win
after 3 . . . � f7 . The road to vic­
tory is very instructive.
l bl h 8 ! dlg7 2 digS ! et> X h8 (if
2 . . . � f6 then 3 bl d8 � e4 + 4
dJf4 � X d6 5 ,bi X d7 + ) 3 dJg6
We see here that the king and the
pawn have co-ordinated their op­
permit Black's pieces to make a
single step. The knight is lost, and
White wins easily after 3 . . . � h6 4
dJ X h6 r;tlg8 5 dJg6.
Neither White's king nor his pawn
directly,attacks the opponent's
pieces. But they help create a situa­
tion in which Black is in Zugzwang
and is forced to make weak moves.
Co-ordinating the play of pieces
with a view to forcing one's oppo­
nent into Zugzwang is one of the
standard devices of the end-game.
In the middle game on the other
hand, this happens very seldom in­
deed.
The creation of a Zugzwang posi­
tion is also the theme of the follow­
ing problem, although White has to
try very hard to achieve success.
G. Zachodyakin, 1 93 1
319
In end-games s uch as this one can
win only by conversion. But in this
situation this does not seem feasible.
Perhaps one should try to hunt
down the knight?
1 59
1 dieS � c7 2 did6 � e 8 + 3 die7 !
� g7
The only possible reply. I f 3
� c7 then 4 dif7 � d 5 5 g6 and 6 g7
mate. But even now the king and
the knight are deprived of their
Here, Zugzwang was achieved
through the co-ordinated action of
all forces at White's disposal.
V Fedo rov, 1 94 7
320

freedom of actio n . It seems that

White can win easi ly by tightening
his grip on the knight and then cap­
turing it by 4 dif7 � h5 5 g6 � g7 6
� e4 � hs 7 � f3 � g7 8 � g4 . Only
8 . . . � h5 ! destroys all i ll usions,
because Blac k is stalemated after 9
� x h5.
The struggle for victory i s very
hard . After 4 dif7 � h5 one can try
Wh ite to p l ay and win
to make a delaying move with the
bishop, e. g. 5 �e4 in order to reply
to 5 . . . fl g7 by 6 � g6, but then Judging from the balance of forces
Black has the excuse 6 . . . � f5 ! th is looks like a drawn garrie. White
Th is means that White's task is to can win only if he manages to cap­
capture the knight, fending off ture his opponent' s k night. Let us
stalemating threats in the process. If try to solve this problem.
for instance he transfers his bishop
1 fl e6 ! . . .
to f7 , the danger of a stalemate is
averted, because Black is free to By immobilizing Black's king White
move to h7 and h 8 . C onsequently, forces the hostile kn ight to move to
the right approach is 4 � g6 digS 5 a weak position .
�f7 + . Now if 5 . . . cbh8 then 6
1 . . . fl a6
d/f6 00h7 7 00e5 00h8 8 OOf4 00h7
9 cbg4 00h8 10 g6, and we have Other replies are out of the ques­
reached the Z ugzwang position we tion. If 1 . . . � a2 then White
wanted . mou nts a decisive double attack 2
But Black cou ld have chosen a fl c7 + , if 1 . . . � c2 (or 1 .
. .

stronger move : 5 . . . 00h7 ! in order � d 3) then White makes a double

to frustrate White's plan after 6
d/f6 00h8 7 00e5 00h7 8 dlf4 ct> h 8
9 d/g 4 et> h 7 . B u t even this obstacle
can be overcome. Instead of 8 OOf4
the right move is 8 ct>e4 ! and after
8 . . . 00h 8 9 ct>f4 ct>h7 1 0 00g4
d/h8 1 1 g6 B lack is in Zugzwang .
attack by 2 � h7 ! (attacking the
knight and threatening mate by
� g6) .
2 � h 7 ctlf7 3 � d3 � b4 (3 . . .
� b8 4 � c5 followed by 5 ct>c7) 4
� c4 ! . . .

1 60
By threatening with a double at­
l ack, White prevents 4 . . . � c2 .
, , . . . dlf6 5 � d4 �
Black has no defence against 6
\ bc5 , by which he captures the
k n ight.
The next example shows a co-ordi­
nated attack by three against two
p 1eces.
W Neustadt, 1 92 9
321
White to play and win
White has to capture both Black
pieces to win the game. This seems
possible by 1 � d6 + dlf7 2 � cS ,
b u t after 2 . dlg6 3 <11 x g8 ct:>g5 �
. .
Black drives the bishop g4 away j ust played 4 . . . �d7 and lost the ex­
in time and frees his knight.
The fine move by which White wins
is 2 � f4 !, by which he forestalls thetile pieces it is particularly import­
counterplay mentioned above. The
forced reply 2 . ct:>f8 is followed
. .
by an elegant manoeuvre which de­
cides matters : 3 � h6 + ct:>f7 4
�e3 � ct:>fB 5 � cS + ct:>f7 6 � x g 1
dlf8 7 � c5 + dlf7 8 � b 4 .
Similar cases of the co-ordinated
use of forces in attacking enemy
pieces also occur frequently in the
middle game.
ALekhine-Euwe
Contest 1 93 5
322
Wh ite t o play
If White takes the pawn by playing
1 .Et x b7 then Black apparently re­
covers it immediately by 1 . . . � c 8
2 .Et b 1 � x d4 .
B u t Alekhine foresaw the possibility
of co-ordinated play by the bishop
and the rook against the hostile
rook at a8 . He replied 3 � x d4
-&:! x d4 4 � f3 ! and Black's own
bishop finds itself blocking the
rook's retreat on the 8th rank. The
rook is being attacked and square
b8 is controlled by rook b 1 . Black
change.
If a piece ventures deep among hos­
ant accurately to assess beforehand
if it will find itself in a dangerous
position and if it will be exposed to
a combined attack by hostile pieces.
Let us consider some examples in
which the queen has strayed into
the enemy camp and found itself
under fire.
161
A correspondence game played in knight to move to f4 and the queen
1 93 5 is trapped. No wonder that White
played 1 3 � e2 !, after which Black
323
suffers a substantial loss in material,
e. g. 1 3 . . . � x e4 14 � f4 � x f3
1 5 � X h5 � X d1 16 ® X d 1 , or
1 3 . . . h6 1 4 � f4 hg 15 � x hs .

Velimirovii-Nunn
Moscow 1 977
325

Black to play

This position occurred after 1 e4 eS

2 � f3 � c6 3 � b s d6 4 d4 ed 5
® X d4 � d 7 6 � X c6 � X c6 7
� c3 � f6 8 0-0 � e7 9 �gS 0-0
10 .b1 ad 1 ®c8 1 1 .b1 fe 1 .
The queen recklessly hurls itself o n
the enemy position : 1 1 . . . ®g4 1 2
h3 �hS. I t i s n o t too late to turn
back yet.
324
White to p lay
Black had pursued a long-term plan
to reach this position apparently as­
suming that he could move his
queen to the K-side and make
dangerous threats there. But let us
see what actually happened.
1 El d 1 ®e4 2 � d 3 ! (not 2 ® x e4
� x e4 3 1:1 x aS ? because of 3
� c2) 2 . . . ®g4 3 h3 ®hs
326

But Black apparently thought that

the queen's posting at hS was pro­
mising. Let us take a closer look at
the resulting position . Black's queen
has but one free square, namely g6.
This circumstance alone is a danger
sign, for it is enough for White's

1 62
 A l though Black's queen does not His queen's freedom of movement
is somewhat restricted, but one
h ave a single square to move to, it is
does not realize at a glance how this
1 1 0 t easy to attack it, because if

White plays 4 �e2 Black has an can be exploited.

;td equate counter : 4 . . . �e4 5 Alekhine continued by 2 ®e2 ! ap­
tA't d2 � x cS. But White found a pearing to prepare for 3 � x f7.
\trong reply, which his adversary Black defended himself against this
temporary threat by 2 . . . � e 8 ,
l t :\d apparently overlooked : 4 � gS !
I low can Black defend himself forgetting for a moment the precar­
now? If 4 . . . h6 then 5 �e2 ®h4 6 ious placing of his queen. No won­
r1 d 4 ! and White may gain a deci­der that this move proved to be his
' ive material advantage. But if first mistake, because now his
Black plays 4 . . . ®h4 then the knight's retreat is blocked, which
.� implest solution is 5 g3 ® h6 6 only underlines the gravity of the
-9 X aS � X c5 7 � X b7 � X b7 8 queen's situation. In an effort to
-i:l x e6 ® x h3 9 � x f8 , because capture the queen Alekhine played
1)• • • 3 g4. Black cannot take the pawn at
� g4 is followed by 1 0 � e 4 .
Seeing no satisfactory way out, h3 because of 4 � d3 ®h4 5 dlg2
Black conceded victory to his oppo­ followed by 6 � h 1 . Euwe suddenly
nent. found himself confronted with the
difficult question of how to save his
queen. He continued with 3 . . .
Alekhine-Euwe � c6, but after 4 dlg2 ·� x eS
Competition 1 93 6 (otherwise White plays 5 � f3) 5 de
is able to free his queen at the ex­
327
pense of a piece, the knight at f6
having nowhere to move. That is
why Black should have played 3 . . .
h6 ! 4 d/g2 � h7 !, giving the queen
enough time to retreat. In the mean­
time Botvinnik has pointed out that
Alekhine's move 3 g4 was not the
strongest. The correct continuation
was 3 ® f l ! In this case Black would
not have had enough time to play
White t o play 3 . . . h6 on account of 4 g3 ® hS 5
g4 ®h4 6 � f3 . Even 3 . . . � hS
In this position White attacked the would not have helped Black be­
queen by 1 �c 1 . Instead of evading cause of 4 g4. The conclusion is ob­
to fS , which would have led to vious. If the freedom of movement
nearly the same game, Black played of one of our pieces is wholly or
l . . . ® h4. partly restricted, then the danger of
its being attacked grows in leaps

1 63
and bounds. One has to be particu­ nated action of pieces in attacking
larly careful in such cases to avoid any target occurs not only when the
running the risk of an unpleasant pieces attack the target jointly.
surpnse. They can play different parts and
still operate in a co-ordinated fash­
The following situation is a con­
ion by acting collectively in harmo­
structive example.
mous concert.
A verbakh-Boudy
Polanica Zdr6j
T h e mec h anism
328 of th e mating attack

The harmonious team-work of for­

ces can be seen with particular clar­
ity in the mating attack. We pro­
pose to examine how such co-ordi­
n ation comes about and what ef­
fects it has on the attack on the
king. We shall investigate the gen­
esis of the mating mechanism and
White to play its function. We hope the reader
will not mind if we start off with the
The position is double-edged. most elementary of things, because
White can capture a pawn, but he as our approach is quite novel its
risks being attacked on the b-fi le, details have not been sufficiently
which would then �e open. He explored yet. But in this manner we
played 1 dd a 1 ! acting as though he shall be better equipped to gain a
were getting set to take the pawn better and deeper understanding of
and moving the king from the dang­ the truly complex question of the
erous file. Black naturally covered co-ordination of forces.
the pawn by 1 . . . aS. But White
329
had something entirely different in
m i n d . His plan was b ased on the
fact that Black's queen placed in the
centre hardly had any squares to
move to. What followed was 2 f4 !
t!t:t x e4 3 � e 1 , and Black's queen
was trapped. After 3 . . e6 4
.

� x f6 Black was forced to resign.

We have examined e nough exam­
ples to be able to draw some conclu­
sions. We have seen that co-ordi- Mate in two

1 64
· 1 'o mate the king we have to de-
1 1 1·ive it of five squares while attack­
' ' ' g it on the sixth. Hence, to
. 1 c hieve the objective we have to
n Hnbine two things : control of the
\quares around the king and the di­
rect attack o n the king. In the given
position the king naturally takes
1 JVer the function of controlling the
\qu ares, while the queen does the
: lltacking.
By going 1 00g6 White immediately
d eprives the opponent of squares f7 ,
g7 and h7 and threatens mate by 2
M1 a 8 . Black can do nothing to avert
d isaster. This means that White has
.� et the mating mechanism in motion
hy 1 00g6 , for his king commands
1 he necessary squares, while his
queen prepares to deliver the deci­
sive blow.
Another kind of attack is also con­
ceivable.
330
Mate in seven
Given this relation of forces the
king can be m ated only in the
corner that is controlled by the
bishop. To do this White has to de­
prive the king of three squares and
attack it on the fourth. In this case
too the job of controlling the
squares is p_erformed mainly by the
king, while the light pieces carry
out the attack. By playing 1 00b6
Black's king is cut off from the 7th
rank. Black's king seeks safety in
flight by 1 . . . 00b 8 , hut after 2
� g4 the king's freedom of move­
ment is curtailed still further and it
is forced to go 2 . . . 00a8 . There
are only two squares left for it to
move on. By united efforts the king
and bishop have cornered the king.
Now the knight moves in for the
kill : 3 � c3 OObB 4 � b5 OOaB.
331
Let us stay with this position a little
while. White's king cuts off squares
a7 and b7, his light pieces have
taken up threatening positions. The
m ating mechanism has been set in
motion. The final assault begins by
5 � c7 + 00b8 6 � a6 + . The king
is pushed into the corner and
mated : 6 . . . 00a8 7 � f3 mate.
B asically the mating mechanism is
the same as in the preceding case -
the king controls the squares and
the rem aining forces attack the en­
emy king. The o nly difference is
that here mate is preceded by a
1 65
number of attacks, whereas in the
preceding position the mating at­
tack was carried out r :ight away.
But there is yet another variant of
the mating mechanism .
332
Mate in three
To preve nt stalemate White has to
m ove his rook. The correct line is to
p lay 1 ld. c7 d:>b8 , a fte r which the
rook can move back to any square
from c6 to c l . By this purely waiting
move, which seemingly does not
threaten anything, White sets the
mating mechanism in motion. Black
is in Zugzwang : the king is forced
to go in the corner 2 . \1Ja8 and
. .
White rep lies with 3 Il c8 mate. It
should be mentioned here that the
m ating m echanism based on Zug­
zwang occurs almost exclusively in
·
th e end-game.
With regard to the co-ordination of
piece movement I should like to un­
d erline the fact that the shaf"es of
th e burden each offen sive piece has
to carry are not equal. In the cases
co nsidered the king did not take di­
re ct part in the attack, its function
h aving been solely to deprive the
opponent's king of some crucial
squares. The remainin g pieces at-
1 66
tacked the king and delivered the
decisive blows. The characteristic
feature of the mating attack is that
some pieces play an active, others a
passive role.
For instance if the king does not
take part in the attack, it can be re­
placed by other pieces or even
pawns. Here is a case in which the
king's part is successfully played by
a rook.
333
Mate in five
White has succeeded, after 1 ld. d7
digs 2 � d6 \1Jh 8 3 � fs \1Jg8 in
setting up the mating mechanism il­
lustrated in diagram 3 3 1 . Here,
too, of course it operates in the
same manner : 4 � h6 + d:>h8 5
� e5 mate.
Thus, substituting pieces to play a
given part basically does not change
anything.
But what happens if we substitute
attacking pieces ? Let us substitute a
knight for the bishop hS in diagram
3 30 . To prevent the king from mov­
ing to square c8 in the same manner
as in the diagram we put the knight
on f7.
 334 335

White to play Mate in eight

Let us try setting up a mating me­

chanism by 1 <ilb6 <ilb8 2 � d6 1 dlb6 <i>bs 2 � d6 dlas 3 � d4 f3
<ila8 3 � c3 <1/bs 4 � cb 5 dla8. Up 4 � 4b5 f2 5 � cl + dlb8 6 � a6 +
to now everything went the same d>as l � b5 (or l � e 8) l . . .
way as before-with one exception : f1 (®) 8 � bcl mate.
the knight at d6 does not threaten
This device of exploiting hostile un­
anything yet. After 5 � cl + d>b8 6
its occurs very often in mating situa­
� a6 + et> as White is only one
tions, because the passive task of
move behind, which he needs for
depriving the attacked king of
regrouping. But he has no time to
squares can be performed by the o� ­
make it because the king is in a
pent's pieces and pawns. Aft�r all, It
stalemate position. It is precisely on
makes no difference to the kmg
account of this stalemate position
whether it is its own or enemy for­
that White is u nable to bring the ces that block its escape, the deci­
mating mechanism to a successful
sive thing is that its elbow room is
conclusion. The situation would be
restricted.
altogether different if Black had an­
other pawn. It could have had disas­
trous consequences for him, for 336

White's knight would have blocked

its advance up to a certain point,
then joined in the attack and-while
letting the pawn advance to the
queening rank-delivered the ?eci­
sive blow. But you know all this
yourself. Here is a characteristic ex­
ample.

White to play

1 61
In this position in which the forces 338

of the opponent are exactly equal,

White gives mate in two moves, the
first being 1 � fS + . To protect
himself against the coming mating
attack Black's king should move to
square h7, but it is occupied by a
piece of its own colour. On the
other hand, if it moves to g8 then
White follows up with 2 � h6.
Square f8 to which it could have
White to play
dodged is also occupied.
Here are another two elementary Here the rook and bishop are at­
examples of similar cases. tacking. The king is forced to go in
the corner after 1 Et g7 + , square f8
337 being occupied by his own knight.
After that any move by the rook
along the g-file (except of course
g8) leads to mate, for Black's pawn
at h7 is, figuratively speaking, a lid
to the coffin .
Through the skilful u s e o f the oppo­
nent' s forces one can make do with
a minimum of offensive pieces for
making a mating play . Let us ex­
amine a few examples in which the
White to play
mating move is executed by a single
attacking piece, the task of preven­
It would be advisable for Black to
ting the king from escaping to the
reply to 1 � f6 + by playing his
adj oining squares being performed
king to f8 . But u nfortunately this
by the opponent's chessmen.
square is already occupied by his
own bishop. He is forced to play his
king to d8, which is followed by 2
Et d7 m ate, for the square next to it
is blocked by his knight.

168
1 ,',/achodyakin, 1 93 1 N N - Pillsbury
1 'r111clusion ofa study
340

339

Black to play

W hite to play and win

In this position Black mates his op­
ponent in three moves : 1 . . .
/\ t first the move 1 h6 seems to be a
®f1 + , forcing the bishop to move
1 n istake because this pawn is lost
to g 1 -2 � g l - , followed by 2 . . .
i m mediately after 1 . . . 1:l g4 + 2
1,?) f4 1:l x h6. But by threatening the ® f3 + ! 3 � x f3 � x f3 mate.
It is not difficu lt to visualize a posi­
k night by 3 digS White puts his op­
tion in which a single pawn delivers
ponent in a critical position. As
the mating blow .
. ) . . . dlg7 costs Black a knight on
account of 4 rl d7 + , he prefers 341
playing it back to g 8 . But this limi­
u.tion of the king's freedom of
movement proves fatal : 4 rl h2 +
�11g7 s EI h7 + dlfs 6 a f7 mate.
The king's own knights prevent the
king's escape from his fate.
[ am sure you are familiar with the
smothered mate in which the king
surrounded by his own p awns and
pieces is checkmated by a solitary White to play
knight. For this reason allow me to
show you another rare case in Here White gives mate in two. The
which a single bishop forces m ate. king's free square e7 is taken by 1
a e7 + � X e7, making it possible
to give mate by 2 f7. Black's king is
surrounded from all sides by his
protectors, which does not prevent
the brave White infantry man from
outsmarting them.

1 69
The question we briefly touched
upon here is very important. We
have seen that in addition to co-or­
dinating the movement of one's
own pieces, keeping an eye open for
possibilities of using the opponent's
forces plays an important part in the
m ating mechanism. How can we
win them over to our side, how can
we force them to hamper their king
instead of helping it?
We shall continue to give this sub­
j ect the attention it deserves. But to
conclude this chapter let us con­
sider some more examples in which
the opponent's pieces are success­
fully exploited for making mating
attacks.
A. and K. Zarytchev, 1 93 0
342
White to play and win
Judging by the way the material is
b alanced this looks like a drawn
game. White can win only by forc­
ing a mate or by capturing the
knight, which is immobilized at
present.
1 � eS d/ g7 2 � d 8 ! . . .
Black's king wants to drive the
1 70
knight off square e S , but White's
bishop comes to the knight's assist­
ance.
2 . . . dlfs 3 dlf2 dies
And now the king wants to free his
knight from captivity.
4 � aS dle7 5 d:>e3 dld6 6 dld4
� c7
Black has achieved his objective,
but what a price he had to pay ! 7
� b4 mate. If the king retreats by
6 . . . dle7 , then White first wins
the pawn at e6 by 7 dieS <l>f6 8
d:>d6 dlfs 9 � d 8 and then the
knight.
A similar situation develops in the
next study.
S. Birnov, 1 95 5
Conclusion ofa study
343
White to play and win
The only line for White to take is to
try winning by m ate ; his chances of
capturing the knight are not very
good. But White starts pursuing the
knight by 1 cbe3 � h 3 2 � d 7 � gS
3 cl)f4 � h7. The knight eludes its
pursuers and arrives safe and sound
::tt the king's side. But it deprives its
king of a free square, thereby en­
abling White to finish off the game
by 4 � g6 + d:>g8 5 � e6 mate.
G. Kasparyan, 1 93 6
Conclusion ofa study
344
ing mechanism. There is no reason
why the opponent's pieces should
not be used in certain situations for
playing the role of static compo­
nents.
Individual contacts

In the starting position the white

and the black forces are separated
from each other by neutral terri­
tory. At first everything is quiet and
tranquil - there are no threats from
either side, because apart from the
knights , no piece can move out into
the open.
And now comes the first move : 1
White to play and win e2-e4. Immediately there is a radi­
cal change : Two pieces-the queen
How can White stop the pawn at and the bishop-threaten to attack
f2 ? The problem seems insoluble. the weak point f7 situated near, and
But the next few moves show that protected solely by the king.
White can profit from the pawn's Black replies with 1 . . . e7 -eS, and
conversion. again the situation changes. Now
l
1 � gS ! f (®) + Black poses the concealed threat of
attacking point f2 . In addition,
The pawn has queened and given White's pawn e4 is blocked and its
check. But White has some effective advance is checked. This serves to
counter checks. illustrate that as soon as the forces
2 � f3 + d:>h3 3 Il hS + d/g2 4 begin to be developed and to take
up forward positions concealed
Il h2 mate.
threats and attacking threats, i. e.
The newly-crowned queen prevents threats of direct contact with hostile
the king from freeing itself from a pieces, begin to emerge. But the
mating hold. Having studied the pieces of either side are not limited
various mating mechanisms we ar­ to offensive operations but also
rive at the conclusion that there are have to protect and cover each
always pieces that play an active other. In other words a variety of
dominant role as well as those that different forms of contact, from
play a passive secondary part. The weakly developed contacts to
former can be termed dynamic, the strong, direct ones evolve between
latter static components of the m at- the pieces on the board . " Invisible

171
lines of force" covering the whole ing the squares in front of them or
board are formed between the by occupying them with pawns or
pieces as the engagement of forces pieces . Thus we have defined two
progresses. These different forms of types of contact - the aggressive at­
contact that develop between the tacking and the restricting contacts .
pieces and pawns of one sid e and They occur with both pieces and
between both sides as well as pieces pawns, although there are differ­
and important squares on the board ences in the effects they produce,
can be referred to as individual con­ depending on the pieces in ques­
tacts. Let us try to define the differ­
tion.
ent types of individual contacts . There is another kind of aggressive
contact, which occurs with passed
pawns, but as a rule this happens in
1 . The attacking contact the end-game. A passed pawn is at­
tracted by the queening square as if
We shall call the contact that exists by a magnet. This gives rise to a line
between the attacked piece or pawn of force and consequently to an in­
and the attacking chessman the at­ dividual contact between the pawn
tacking contact. An attacking threat and its queening square.
can also be regarded as an attacking
contact, even though it is weaker.
Not all pieces can attack each 3. The queening threat
other, e. g. the king can attack
neither its opposite nu mber nor the It is easy to see why a queening
queen. But the king can assert itself threat is no less dangerous than an
successfully against the queen if it is attack on the queen itself. The or­
left without cover within the king's der of threat is inversely propor­
radius of action. As regards pawns, tional to the pawn's distance from
they can attack each other only on the queening square. Obviously a
adjacent files. pawn one square away from the
q ueening squ are is more dangerous
than one that still has several steps
2. The restricting contact to make.
Let us consider some examples of
The operating range of a piece is individual contacts.
restricted the moment it cannot
move to squares which would
otherwise be open to it. Tli.is can be
done simply by attacking such
squares or by occupying them. In
both cases restricting _contacts oc­
cur. The same applies to pawns.
They can be obstructed by attack-

1 72
 345
Here we see an example of a re­
stricting contact, the king being in
control of both squares which the
knight could otherwise move to. In
a situation such as this, where the
knight is concerned, the restricting
contact also acts as a threatened at­
tack, for on the next move the king
could attack the knight. This means
that in the position shown here the
king can cope with the knight
single-handed.
And here is a situation showing that
the bishop can confine the knight
but cannot capture it without assist­
ance.
346
The bishop has wholly confined the
knight on the edge of the board, but
it can do no more. Additional pieces
are needed to attack it.
347
A similar situ ation with a rook. The
rook can completely immobilize the
knight on any of the following four
squares : b2, b7, g7 or g2, but it can­
not capture it.
348
As could be expected the queen is
best able to cope with the knight. In
the diagram the knight is in the cen­
tre, but it takes the queen only a few
moves to hunt it down.
1 ®e4 � d7 (1 . . . � f7 2 ®ds
� h6 3 ® e6, and the knight falls) 2
®fs � b6 3 ® b s � c8 4 d!Jcs with
the same result.
I n this case the restricting contact
became at the same time an attack­
ing contact. The important condu­
sion to be drawn here is that a re­
stricting contact can not only de-
1 73
velop into an attack, hut that a re­
striction can itself be an attack.
4. The protective contact
It often happens that an attacked
piece can be covered or protected
by another piece giving rise to a de­ ing contact exists only between the
fensive contact between like-col­
oured pieces, which we shall call the attacked. While, vrhen a piece is
protective contact. It goes without
saying that such cover is given only tact exists between the attacking,
when the attacking piece is more
valuable than the attacked piece,
because an exchange would not be
desirable if the opposite were the
case.
The mobility of the covering piece
is impaired because it cannot move
away without endangering the
piece it covers . We shall consider
this point in detail later on in con­
nection with some examples of pro­
tective contacts, but I should l ike to Attempts to stop a pawn heading
point out in advance here that the
king can protect not only a pawn
but also its advance.
5. The interposing contact
Sometimes one can defend a piece
attacked by a hostile piece by plac­
ing a piece or pawn between it and
the attacking piece. In this case we
have a defensive constellation
which can be c alled an interposing
contact. Here, too, the defending
piece's mobility is impaired, be­
cause if it m oves aside then the at­
tacked piece is lost. But there is a
fundamental difference between
1 74
protecting a piece by covering it and
by interposing another piece. Nor­
mally, when a piece is covered the
piece giving cover is not attacked,
whilst in the case of interposing the
protected piece remains more or
less u nder threat. In other words,
when a piece is covered the attack­
attacking piece and the piece being
protected by interposing then a con­
the attacked and the interposed
piece, although in the later case the
relations are altered, i. e. the
·
shielded piece is now under threat
of attack, while the interposed piece
is actually attacked.
6. Refuting the queening
threat
for the queening square by attack­
ing it or the squares in front of it or
by blocking it lead to restricting
contacts, which we have already
discussed. But the independent indi­
vidual contact which occurs when
the king comes from afar to stop the
pawn, should be examined separ­
ately. Contact exists between the
king and the pawn approaching the
queening square . It is described by
the familiar quadrangle rule. We
call it the "queening threat" .
 349
You will recall that the king will
stop the pawn if it is in the qua­
drangle c l -c6-h6-h l . From the
periphery of the next bigger qua­
drangle, it can threaten to stop the
pawn, whilst from the rest of the
squares on the board the king can
try to make such a threat. These are
threats of different orders again. It
is very important to note that nega­
tive restricting contacts can also
happen between like-coloured
pieces. They occur when pieces 1.>r
pawns for some reasons restrict t-he
lines of action or paths of their own
pieces, thus hampering their move­
ment.
A elementary example.
350
White to play
White's king enters the pawn's qua­
drangle, but it is unable to stop it
because after 1 dlc6 hS 2 ®ds h4 3
00e4 h3 the square f3 is occupied by
the king's own pawn causing it to
lose valuable time and to arrive too
late. Since we have already consid­
ered contacts that arise in encoun­
ters between the king and the pawn
it will be instructive to discuss the
elementary and familiar end-game
" king and pawn versus king" as an
example illustrating the application
of the theory of contacts. The bone
of contention around which the
whole struggle rages is the queening
square .
351
In this position the king not only
protects the pawn but also supports
its advance because it controls the
queening square. In this way two
protective contacts are formed ­
the cover given to the pawn and the
support given to its conversion.
These protective contacts are at the
same time restricting contacts, mak­
ing it impossible for the opponent's
king to do anything to influence the
inexorable course of events.
175
 352 In this position the squares marked
"0" are those from which White's
king can establish contact with the
pawn and its queening square,
thereby ensuring the pawn's promo­
tion. In the theory of pawn end­
games these squares are known as
key squares, because the objective
of this kind of end-game , i . e .
queening, c a n b e achieved once
they are occupied by White's king.
We know that in this position
If the king is unable to get to these
White's pawn forces its way to the
squ ares then he can never support
queening squ are despite the fact
his pawn's advance to the queening
that the Black king is posted on the
square.
queening square. If it is White's
turn to play, he quickly transposes it. 354
to c7 or e7. After 1 dld6 White' s
king threatens t o occupy o n e of
these important squares from which
it can support the pawn's advance.
Black's king tries to restrict it in or­
der to keep it off these squares . But
Zugzwang frustrates Black's plan ;
his king has to step aside and let his
opposite number pass : 1 . . . dieS 2
dle7 or 1 . . dieS 2 dlc7. If it is
.

Black's move then 1 . . . dlc8 is fol­ Black to play

lowed by 2 d6 dldS 3 d7. The p awn
h inders Black's king by contesting
square c8 . It is forced to move back
3 . dle7 and after 4 dlc7 queen­
. .
ing cannot be prevented.
353
Black must obviously keep White's
king off c7 and e7. This can only be
done by 1 . . . dld s . After 2 dle6
dieS or 2 dlc6 dieS Black's king
prevents his opposite number from
penetrating to the key squares. But
if White throws his pawn into ac­
tion by 3 d7 + 00d8 it becomes
clear after 4 00d6 that White's king
and pawn have deprived Black's
king of all free squares and this is
tantamount to stalemate.

1 76
 355 White is unable to overcome
Black's restricting contacts, and the
result is the same regardless of
whose move is next. If it is White's
move then Black gives stalemate by
1 00a8 00b6 2 a7 00c7. If it is
Black's move then 1 . . d:>c8 2.

00b6 00b8 3 a7 + 00a8 4 d:>a6 and

Black is stalemated. The strategy of
this game is of elementary simpli­
city : to win, White's king has to get
to b7 or b8 thereby establishing a
Here, too , White seeks to gain con­
protective contact with the pawn's
trol of the key squares and Black
queenmg sq4are.
tries to hinder his opponent and
prevent him from achieving his ob­ 357
jective. This shows that the out­
come of such elementary games is
determined by the possibility of es­
tablishing protective or restricting
contacts. If White manages to esta­
blish contact between the king and
the pawn's queening square then he
can support its advance to the
queening square and there is no­
thing to prevent its pawn from get­
ting there. But if he fails to break White to play
through his opponent's restricting
Black's king is in the pawn's qua­
contacts then the game ends in a
drangle and it would be pointless
draw. The a- and h-file pawns have
for it to advance. But White's king
special parts to play in such end­
succeeds in reaching the key square
games .
g7. This is done as follows :

356
1 00g4 ! OOeS 2 digS . . .
While White's king presses for­
ward , it prevents its adversary from
moving to squares f6, f7 or f8 .
2 . . . 00e6 3 d:>g6 00e7 4 00g7
The pawn inexorably marches on to
the queening square.
The term "contacts" not only helps
to describe the nature of the

1 77
struggle in this end-game but also
i ts special features and its essence.
From this it follows that the theory
of contacts put forward here is fully
in accord with the theory of key
squares. But the latter covers only
the struggle of kings in end-games
with pawns, whilst the theory of
contacts also applies to other, much
more COIJiplicated end-games and
middle game positions. We shall
presently have an opportunity of
convincing ourselves of this.
The co- ordinated attack
In the preceding chapter you have
been acquainted with the new term
of " contact" . This term enables us
to reflect special features of situa­
tions that have developed on the
board ; it can also be used for mea­
suring the degree of co-operation
of forces on the two sides . We have
defined six different kind 5 of ele­
mentary contact :
1 . the attacking contact
2. the restricting or confining con­
tact
3. the conversion contact
4. the protective contact
5 . the interposing contact
6. the refutation of the conversion
threat
With certain qualifications all these
contacts can be broken down into
three groups, i. e. actively attack­
ing, restricting and defending con­
tacts. Active contacts include at­
tacking threats, attacks and queen­
ing threats. Restricting contacts in­
clude confining threats, restrictions
1 78
and the warding off of conversion
threats . Defensive contacts include
defensive possibilities, support pos­
sibilities of interposition and the in­
terposition itself. But this break­
down does not apply to all cases,
for the function of a certain contact
depends, as we shall soon see, on
the concrete situation .
In individual contacts, pawns and
pieces are u nited by almost imper­
ceptible bonds (except, of course,
the queening threat, where the
pawn is linked to the queening
square) , giving rise to diverse cross­
links between them.
Let us now examine co-ordinated
piece cross-links , which are charac­
terized by certain contacts.
We have already exhaustively stud­
ied the mechanism of the mating at­
tack (cf. pp. 1 64- 1 7 1 ) . We have
found out that in this mechanism
there are always pieces (one's own
or the opponent's) , which hem in
the hostile king ; in addition there
has to be a piece that delivers the
actual death blow to the king.
Thus, every mating attack involves
an attacking threat in the form of an
attack on the king and some re­
stricting contacts. This means that
any mating position-regardless of
the forces taking part-can be de­
scribed in terms of combinations of
contacts, namely combinations of
restricting contacts and attacks. Let
us consider as an example two
cases, in which the king exposed to
a mating attack is either in the
corner or on the edge of the board .
In the diagrams the squares on
which the king's freedom of move-
ment has to be curtailed are marked (b) Mating with the rook
with an x .
360

1. The cornered king

358

Here restricting contacts have to be

created with two squares . The task
of restricting the king can be per­
formed, e. g. by the king or a rook.
If the cornered king is to be check­ A finish in which only White pieces
mated it has to be cut off from three attack is conceivable : A knight
squares and attacked on the fourth . hems in the opponent's king and
The mating pattern varies accord­ covers the mating rook.
ing to the attacking piece.
361

(a) Mating with the queen

359

And here are a few positions with

Black's pieces taking part in confin­
ing the king.
In this case the restricting contact is
required o nly for one square. It is
created by covering square a7 with
any piece. It would also suffice if it
were occupied by a black pawn or
rook.

1 79
 362 (c) Mating with a bishop
Two squares have to be restricted in
order to give mate with a bishop.

365

363

In the chapter entitled "The me­

chanism of the mating attack" we
have analysed in diagram 3 3 0 the
attack with a knight and a bishop
on the lone king. But a position in
which a pawn plays the part of the
king is also conceivable as a mating
finale.

364 366

A similar ending results when the

In diagram 362 a black rook can be
king is threatened by two bishops.
just as fatal as the black pawn . In
the next diagram the knight can be
replaced by a pawn. And finally, di­
agram 364 shows the king being
checkmated on the 8th rank, which
is quite common.

1 80
 367 370

Here are two more examples in

which the bishop is assisted by the 371

king and a pawn.

368

The second position shows the case

of mate with a single bishop, which
occurs seldom in practice.

369 (d) Mating with a knight

The king has to be deprived of three
squares in order to give mate with a
knight.

372

Positions in which restricting con­

tacts are made by black pieces or
pawns merit special attention . Here
are two such positions.

181
First a position in which only white
pieces are attacking.
373
And here are two examples in which
the king's own pieces and pawns
restrict the king's freedom of move­
ment.
374
375
Diagram 375 illustrates the well­
known "smothered m ate" position.
1 82
(e) Mating with a pawn
Like a bishop, a pawn needs two
restricting contacts, but unlike the
former, it has to be supported by
one of its pieces . The simplest mat­
ing position is in which White's
king is on c8 or cl and square a7 is
controlled by a knight on c6 or a
bishop on cS .
2. The k(ng on the edge
If the king is being attacked on the
edge of the board then the mate
ending does not differ essentially
from that in which the king is corn­
ered. The only difference is that i n
t h e former case t h e king h a s t o b e
c u t off from five squares a n d at­
tacked on the sixth, which means
that it is only a matter of the num­
ber of restricting contacts, whilst
the number of attacking contacts
remains the same, namely one.
376
(a) Mating with the queen
If the queen attacks horizontally,
then two restricting contacts are
needed.
 377
The same applies if it attacks along
a file.
378
Th e queen is the only piece that can
attack six squares if it is supported
by one of its pieces or pawns. Here
is an example of such an attack.
379
As we shall see l ater, mating attacks
in which the queen is covered by a
piece or pawn are quite frequent in
practice.
(b) Mating with a rook
To mate the king on the edge of the
chessboard a rook needs three re­
stricting contacts. Perhaps the most
typical position is the one shown in
diagram 364, which we have al­
ready discussed. The fact that we
have shifted it three files to the right
makes no difference.
( c ) Mating with a bishop
381
The problem of creating four re­
stricting contacts can be solved in a
n umber of ways. In the example
shown in diagram 382 this job is
performed by the queen, while the
attacking is done by the bishop.
1 83
 382 385

In the following cases not only

white but also black pieces take part 386
in hemming in the king.

383

(d) Mating with a knight

Like the bishop, the knight needs

384 four restricting contacts in order to
execute a mating attack.

387

Here are two examples of such mat­

ing positions.

1 84
 388
389
(e) Mating with a pawn
In conclusion, here is a mate ending
in which a pawn delivers the fatal
blow. To give this kind of mate
there have to be four confining con­
tacts and the pawn has to have piece
cover.
390
To illustrate this kind of mate I have
picked an example in which only
White men take part in the offen­
SIVe.
391
I did not intend to draw up a com­
plete list of all possible mate end­
ings. If you wish you can extend this
list without any trouble. This would
be a profitable exercise by the way.
Mating pos itions in the centre of
the board do not differ basically
from the positions we have ex­
amined. Here too there is always a
piece that makes the mating attack,
other men restricting the king's mo­
bility-an important but still only a
supporting part. They influence the
course of the struggle but they do
not determine it.
In all these mating systems using
various piece constellations there is
always an attacking contact, the ac­
companying confining contacts
and in some instances-e. g. in dia­
grams 36 1 , 368, 373, 3 79 and
39 1 -protective contacts. All these
contacts centre on the king. This
means that any mating position can
b e described in terms of a co-ordi­
n ated combination of an active at­
tacking contact, confining contacts
185
and occasionally protective con­ Here is an example of a stalemate
tacts . The number of confining con­ ending.
tacts varies depending on whether
Nikolayevski- Taimanov
the king is posted in the corner, at
Tallinn 1 966
the edge or in the centre of the
board, but the active attacking con­ 392

tact determines the mate ending .

Situations in which the object of at­
tack (in this case the king) is sub­
jected to an attack and at the same
time its freedom of movement is
restricted will henceforth be called
combined attacks. I should like to
repeat that a combined attack al­
ways involves an attacking cont�ct,
a confining contact and occasion­ Black to play
ally a protective contact. In this
connection I should like to empha­ At first sight Black's position seems
size that the co-ordinated attack on utterly hopeless because he is about
other pieces, which we discussed at to lose his pawn at cS. But after
the beginning of Part 2 , is basically 1 . . . 00b6 2 OOdS et> as ! we realize
the same as a m ate ending. Here too that Black's king has an ideal hiding
we always find an active contact place at a S . If Wh ite takes the pawn
along with confining contacts at cS he deprives Black's king of b6,
aimed at the piece being attacked. its only free squ are, and the game
That is why such a co-ordinated ends with a stalemate .
combination of contacts can also be
But what conditions ensue if perpet­
called a combined attack.
ual check is given ? Let us examine
E nd-games leading to a stalemate
the finale of a study by Rinck.
are characterized by positions in
which the king is subj ected exclu­ H. Rinck, 1 904
sively to confining contacts, but in
393
which there are no attacking con­
tacts . In stalemate positions the
king has to b e cut off from three
squares if it is in the corner, from
five squares if it is on the edge of the
board and from eight squares if it is
in the centre. Thus, a stalemate is al­
ways a co-ordinated combination
of confi ning contacts, the only dif­
ference being the number of such
contacts. Draw

1 86
No matter where Black's king at­ By continuing with 1 . . . ®h4 + 2
tempts to flee he cannot escape per­ ® h 7 ® d 8 + Black allows his oppo­
petual check : nent to get a second queen by 3
g 8 ( ® ) . But after 3 . . . ®f6 + 4
1 � d 3 + ddg4 2 �e2 + ddh3 3
® hg7 ® h4 + 5 ® 8h 7 ®d8 +
� f l + ddg4 4 �e2 + ddf5 5
Black gives perpetual check.
� d 3 + dde6 6 � c4 + ddd7 7
In the final position we see a co-or­
� b S + ddc8 8 � a6 + .
dinated combination of an attack­
White's bishop operating on the ing contact by the white queen and
white squ ares is responsible for the two confining contacts by the two
attacking contact, while his partner white queens. If i nstead of the white
covering the black squares makes queen at g7, for instance, there
the confining contact. Black's stood a bishop and instead of the
queen, bishop and the two pawns queen at h7 a pawn, Black could
play an equally important role in even checkmate the king. Although
the perpetual check mechanism. for a brief moment the two queens
They confine their own king's can establish an i nterposing con­
movement to the white diagonal. tact, a fresh line of attack is opened
Just as in the mate ending we have a immediately enabling Black's queen
co-ordinated combination of an at­ to give check again and resulting in
tacking contact and confining con ­ a perpetual attack.
tacts. But these confining contacts Manoeuvres leading to a draw,
were only good for giving perpetual which are based on the building of a
check, they did not suffice to check­ stronghold or on a blockade, are
mate the king in the centre of the also characterized by certain co-or­
board . dinated combinations of contacts. It
happens frequently that the
T h e combination o f contacts can
stronger side is simply u nable to
also be of a different nature in per­
penetrate the opponent's fortifica­
petu al check situations.
tions .
G. Lolli, 1 763 Here is an amusing example de­
394 voted to this theme.

Black to play and force a draw

1 87
E. Rudoiph, 1 91 2
395
Draw
Black has an overwhelming mate­
rial superiority, but White takes ad­
vantage of the opportunity to give
away one of his bishops and to bar­
ricade all e ntrances end exits to the
fortress.
1 � a4 + ! d) x a4 2 b3 + \t>bs 3
c4 + d)c6 4 d5 + \t>d7 5 e6 +
d) x d8 6 fs .
As if with the m agic wand, a "magic
fence" made up of white and black
pawns suddenly materializes on the
board. This fence has produced a
combination of confining contacts,
which cannot be penetrated by a
single black piece . The contacts are and combinations
directed against all black pieces,
completely paralysing their activity.
Unable to find a target, Black sim­
ply does not know which way to di­
rect his forces.
1 88
A. Troitzky, 1 91 0
396
Draw
Here we see an example on the sub­
j ect of blockading the king. Black's
monarch sits on h4 in a trap and
cannot flee because of the combina­
tion of confining contacts directed
against it. The queen can do no­
thing on its own .
We have seen that draws by stale­
mate, perpetual check, stronghold
and blockade are also characterized
by a co-ordinated combination of
contacts .
Contacts
in the double attack
I should like to remind the reader
that the concept of double attack
has been given a broader scope here
than has been the case in theory so
far. We regard the three following
combinations of pieces as instances
of double attack : 1 . A piece or a
pawn attacks or threatens to attack
two objects ; 2. Two pieces attack or
threaten to attack a certain object; Here is a characteristic example.
3. Two pieces attack two objects. 397
These objects can be either pieces
or squares. If pieces are attacked
then two active contacts ensue in
the double attack. But generally
speaking, a combination of any two
of the six elementary contacts can
prove to be a double attack.
This important idea is one of the
pill'ars of the author's Theory of
Contacts. This extends the range of
the concept of double attack used in This double attack is harmless be­
modern theory still further and en­ cause the rooks provide adequate
ables us to consider a wider array of mutual cover. But let us make a
positions from this point of view slight change in this position by in­
and makes it easier to understand terposing a black pawn.
the motive forces behind the events 398
taking place on the board. The ex­
tension of the concept of double at­
tack is justified and methodically
expedient, for it enables us to give
due credit to this very important
manifestation. Thus, the double at­
tack is always a co-ordinated com­
bination of contacts, which in the
first and third cases are directed
against two, and in the second
against one 9bject. We shall now This immediately enhances the ef­
examine the various manifestations fectiveness of the double attack be­
of the double attack in proper or­ cause it deprives the rooks of their
der. mutual cover. The loss of one of
them is inevitable. Thus, in the first
1. One piece attacks example there was an equilibrium of
or threatens to attack contacts-both attacks were refuted
two objects through the mutual cover of the
rooks. The pawn introduced be­
This kind of double attack covers tween the rooks has upset the equi­
all variations of the "fork" and "pin­ librium by severing the contact be­
ning" . In an attack on two pieces its tween the rooks thereby,substan­
effectiveness depends on the con­ tially enhancing the effectiveness of
tact existing between the two the double attack.
threatened pieces.

1 89
An interesting double attack with
active contacts is shown in the next
diagram .
399
On the one hand the pawn at c7
threatens to queen (this is a first ac­
tive contact), while on the other
hand it is attacking the knight at d 8
(this i s the second active contact).
To refute this double attack one has
to gain control of squares c8 and d 8
with some other piece. The knight
itself is not in a position to do this.
As pointed out earlier, a combina­
tion of any two of the six elemen­
tary contacts that I have mentioned
previously can lead to a double at­
tack.
Here are some examples of this.
400
White to play
1 90
The queening threat is met by the
bishop at bS. At the same time the
pawn g4 is outside the white king's
sphere of influence. But by 1 dieS
White launches a double attack
threatening to take Black's bishop
and to enter the quadrant of the
pawn at g4. After 1 . . �d7 2 d/d4
.
dlb6 3 dle4 <!lc6 4 d/f4 dld6 5
e 8 ( ® ) the game ends with a draw.
The double attack was based on
two contacts, namely an active one
(the attack) and a passive one (the
threat of stopping the opponent's
pawn) .
401
White to play
I � this position 1 dlc6 also l�ads to
a double attack, namely to a combi­
nation of an attack and a threat to
hem in Black's bishop, or in other
words, to limit its sphere of action.
The pawn can be advanced both af­
ter 1 . � e 3 2 � c5 and after
. .
1 . . � d 8 2 � c7 with awin for
.
White. Here we have an active and
a confining contact.
 402 the concealed threat of attacking
the pawn at a7. But there is an'o ther
contact, for through the text-move
White gains control of square d4
thereby hemming in Black's king,
which cannot move there. In this
manner White prevents the Black
king from reaching square c7,
which consequently cannot prevent
White's king (after capturing the
pawn at a7) from asserting control
White to play
of the key squares b7 and b 8 . After
1 ct>b4 2 cbc6 cbas 3 cbb7 ct>bs
At first sight White's position seems
. . .

4 cb x a7 cbc6 5 ct>b8 White ad­

completely hopeless. His pawn is
under fire, and his king is far from vances the pawn to the queening
the black pawn's quadrant. But square. Thus, we see how an active
contact can be combined with a
White saves the game by continuing
confining contact in a double at­
with 1 cbb7. This move is a typical
tack.
double attack, combining the threat
of supporting his pawn with that of It is very important to understand
stopping his opponent's pawn. The the conclusion of the following
issue is settled after 1 . . cb x aS 2
. study in order to grasp the meaning
cbc6 , establishing two defensive of the double attack.
contacts.
The next diagram shows an inter- G. Zakhodyakin
esting example of the double attack.
404
403

White to play
White to play
By playing 1 cbds White threatens
In this position White confronts his to stop the pawn at h4. That is why
opponent with a double attack by Black proceeds with 1 . h3 . This
. .

going 1 cbds ! The first contact is is followed by 2 cbc4, which con-

191
fines Black's king. This move in it­ E. B. Cook, 1 864
self would not be so dangerous, but
with the assistance of the bishop 405

and the pawn White threatens to

give mate in two moves. And the
game actually ends in a win for
White, for if 2 . . . h2, then 3 � b 4
h 1 (®) 4 b3 mate. Thus, strictly
speaking, the move 1 OOdS is a
double attack, for in addition to
threatening to intercept the pawn,
this move contained a mating threat
of the third order. Despite the fact
Black to play, White force a draw
that the king itself cannot give
mate, in co-operation with the pawn After 1 . . c2 2 � b7 + 00c8 White
and the bishop it poses a combined
.

apparently cannot save himself. But

mating threat. This factor is very
important, for it shows that even a
mating threat-despite the fact that
it is a complex combination of a
threat and confining contacts-can
be represented as a single contact of
any piece if it brings about a mating
combination . By playing the king to
dS White has created a combinatio n
o f pieces consisting o f the white
king, the black pawn (refuting the
queening threat) as well as the
black king and the white pawn and
bishop (mating contact). Thanks to
the co-ordinated combination of
contacts, the double attack has
fused all these pieces into single
u nit. This can also be expressed in a
different way : It is precisely this
combination of pieces that has con­
j ured up the double attack.
Threatened stalemate can also form
part of a double attack.
he finds a way out, which is con­
nected with a double attack : 3
� bS ! This move contains two
threats . The first is to stop the pawn
by 4 .Et cS( + ), leaving Black no alt­
ernative but to continue with 3 . . .
c l ( ® ) . But then the second threat 4
� cS + ! ® x cS stalemate material­
izes. Thus, this double attack con­
sists of the threat of stopping the
pawn and of the concealed threat of
a stalemate by sacrificing a rook.
The fact that Black was forced to
take the rook because of the double
attack on the king and the queen is
also important here. The double at­
tack was staged in such a way that
B lack was forced to accept the sac­
rifice .

1 92
A. Troitzky, 18 9 7 pieces, whilst the squares around it
Conclusion ofa study · are subjected to two-fold attacks.
In studying mating attack contacts
406
we have established (cf. diagram
379) that for instance the queen,
supported by any one of its men,
can deprive the king posted on the
edge of the board of all squares. By
our definition, these cases are
double attacks.
Let us take a well-known elemen­
tary position in which the queen
and a pawn lau nch the double at­
White to play
tack as an example to illustrate this
point.
By 1 \1lh3 ! White attacks the pawn
and threatens stalemate. If Black
407
goes l . . . g l ('®) (or 1 . . . g 1 ( Ei )
then White replies 2 b7 + dl x b7 3
c8(®) + dl x c8 stalemate ! On the
other hand, 1 . . . gl ( � ) + is fol­
lowed by 2 d/g2 and after 2 . . .
� e2 3 \ll f l with a double attack on
the bishop and knight.
There were actually three connec­
tions in this position : an attack, a
threatened stalemate and a con-
cealed threat of a two-fold attack. White to play
An important conclusion can be
drawn from all this. In addition to White stages a double attack
the single connections examined by against g7 by 1 f6. To ward off mate
us, a double attack can also comprise Black plays 1 . . . g6 thereby sever­
such complex connections as mate, ing the queen's contact and repell­
stalemate or a two-fold attack. ing the attack. But then the queen
steps aside by 2 ®h6 and restores
2. Two pieces attack contact with g7, after which mate is
imminent. By the way, the pawn at
or threaten f7 plays a negative part in such situ­
to attack an object ations, preventing both the king's
flight and the cover of the fatal
First and foremost this category in­ square g7 from the rank.
cludes all positions in which attacks But we have already examined mat­
on the king are carried out with two ing attacks in which two pieces are

193
involved. The attack with a rook
and a knight, for instance, which
leads to a double attack, is shown in gives mate : 1 � c6 + dieS 2 Eld 8
diagrams 337 and 36 1 .
Double check i s a special case of the Note the special feature of this var­
double attack. It is a terrible wea­
pon in a mating attack.
First of all two examples illustrating
situations in which double check
leads to mate. In the first case the
rook that gives mate is supported by
a bishop.
408
Mate in two moves
After 1 �gS + the king has to step
aside by going 1 . . . dle8, this is
followed by 2 ht d8 mate.
409
Mate in two moves
1 94
This is a similar position in which
the rook, supported by a knight,
mate .
iant of the double attack : At first
the two pieces attack one point,
then one of them occupies that
point and gives mate, whilst the
other piece supports it. Conse­
quently two contacts occur in the fi­
nal phase of this double attack,
namely an attacking contact and a
protective contact. Confining con­
tacts by the opponent's pieces are
also usually involved .
Here is another instance of such a
double attack.
41 0
White to play
Two white pieces are attacking the
pawn at h7. It cannot be covered,
but Black's king is ready to flee : 1
® x h7 + d/fs 2 ® h 8 + dle7. But
as you know, the strong move to
make in such positions is 1 ®h6 ! In
this way White cuts off the king's
escape route and threatens to
change the line of fire, or more pre­
cisely, to transpose the double at­
tack: 2 � x h7 + d/h8 3 �g6 +
00g8 4 ®h7 + 00f8 5 ® x J7 mate. Kotov-Kortchmar
This device is used qu ite often in at­ Tu/a 1 95 1
tacking the ki ng.
41 2

3. Two pieces
attack two objects
The simplest case of such a double
attack is when two pieces attack
two hostile pieces . This is shown in
diagram 4 1 1 .

Kan- Yudovitch
White to play
Game �·.1 riant
Tbilisi 1 93 7
White seems to be in trouble here,
and this would be true if Black's g­
file pawn was posted on g6. But as
things are White has a chance to ex­
tricate himself from a difficult situa­
tion by means of an effective sham
sacrifice : 1 ® x h7 + ! After 1 . . .

00 X h7 2 � d 3 + (the saving double

attack !) 2 . 00g8 3 .Et x f2 White
. .

has better prospects.

White to play Popiei-Marco

Monte Carlo 1 902
White establishes two active con­
tacts by 1 � f5 !, the bishop attack­ 41 3
i ng the queen and rook e 1 attacking
t he rook at e7. After 1 . . . ® x f5 2
r:I x e7 White's double attack gives
h i m an advantage in material .
This kind o f double attack i s parti­
cul arly effective when one of the at­
t :.1.cked pieces is the king.

Black to play

195
 This famo us p osition keeps crop­ sition the bishop establishes the at­
ping up in chess textbooks as a cu­ tacking contact, whilst the rook
riosit y. The bisho p at d4 is attacked makes the confining contact.
by three piece s, and there seem s to
be no way of givin g it addit ional The following diagram shows a fas­
cover. That is why M arco decid ed cinating case of defence by double
to resign , altho ugh he could have attack.
not only defen ded h imsel f but even
�on the game ! By playing 1 . . . S. Birnov
.,.. g 1 � Black cou ld have made a mat­ Excerptfrom a study
ing threa t on h2 and atttacked the
414
queen at d 3 . As yo u see, the doub le
attac k in this · cas e consi sts of an at­
tacki ng conta ct an d a matin g
threa t, which in turn consi sts of an
attack ing conta ct (by the queen )
and a prote ctive co ntact (by the
bishop ) .
Whe n we were exam ining the mat­
ing mech anism, on sever al occa­
sions we came across doubl e attack s
in which one piece attacked the White t o play
king, while other p ieces cut it off
fro� adjoi ning sq u ares there by en­ Black's pawns are pressing forward
abling the attacking piece to check ­ inexorably. How can they be
mate it. A_ typic al exam ple is the stopped? There is a way out. White
mate endm g shown in diagram 3 3 8 . plays 1 Ei g l ! thereby establishing
I t merit s speci al attentio n . First of contact with the first rank (confin­
all Whit e launc h es a doub le attack ing contact) and being ready to at­
�:m g7 by 1 EI g7 + driving the king tack the king with the bishop. The
mto the corne r. Th is doub le attack threats of 1 . . . a1 (®) and 1 . . .
consi sts of an attac king conta ct (by e 1 (®) are suddenly refuted , be­
the rook) and a p ro tective contact cause if either move is made, White
(by the bisho p) . Th is is follow ed by plays 2 �g2 + and the new-made
anoth er doub le a tta ck in which the queen is forfeited .
rook move s back al ong the file
And now a few examples in which
keep ing watc h over squar e g8
the double attack is accompanied
while the bisho p ch eckm ates th e
by complex contacts.
corne red king. Note that in this
m ating mech anis m consi sting of
two doub le attac ks the tasks of the
piece s and conse qu ently also of the
conta cts altern ate : I n the final po-

1 96
Fine- Thomas
Hastings 1 93 6
41 5
White t o play
By going 1 ®g4 ! White organizes a
double attack on g7 and poses a
mating threat. Two contacts came
into being here : an attacking con­
tact (by the queen) and a protective
contact (by the knight). At the s ame
time White has set up another
threat, i. e. a second double attack
by 2 � h6 + , threatening the king
and attacking the u nprotected
queen at d7 with his queen.
Here too we have two contacts : the
th �eat of harassing the king with the
kmght and the threat of attacking
the queen with the queen. Thus we
have seen that White's move 1 ® g4
creates an involved system of con­
tacts consisting of two double at­
tacks . Being unable to defend him­
self, Black resigned.
Alekhine-Rubinstein
Karlsbad 1 923 · .
41 6
White to play
The weakened position of Black's
king and the menacing stance of the
knight at gS seem to favour the idea
of transposing White's queen to the
K-side. White would indeed tri­
umph immediately if his bishop
stood on g2 by lau nching a double
attack by 1 ® f3 , attacking f7 and
threa�e ? ing ® hs + . But the bishop
at e4 ts m the queen's way. If one
co � ld only move it out of the way
qutckly without giving Black time
to play 1 . . . g6.
�lekhine solved this problem by go­
mg 1 �g6 ! This move is a double
attack on f7 by the bishop and the
knight. At the s ame time White pre­
pared a concealed threat of a sec­
ond double attack consisting of an
attack by the queen on h7 and of
the supporting posting of the
knight. If for instance Black plays
1 . . . fg then 2 ® e4 (2 ®g2 is also
good) 2 . . . � x b4 3 ® x g6 r3Jg8
4 ® h 7 + r3Jf8 . Now the decisive
o nslaught begins : 5 t!tJ h8 + r3Je7 6
® x g7 + dJe8 7 ®g8 + � f8 8
1 97
®g6 + dle7 9 'dt1 X e6 mate. That is The last finesse resulting in a remar­
why Rubinstein replied with 1 . . . kable drawn position.
® e s , but after 2 � x f7 + .61 x f7 3
� x f7 ® fS 4 El fd l ! White got a 418

qu ick win.

G. Zakhodyakin, 1 93 0
Conclusion ofa study

41 7

White's pieces are linked by invisi­

White to play
White' s position looks catastrophic,
. the knight confines the king and
b o th pieces being under fire and
1 . . . f 1 (®)_is threatened. In princi ­
pl e this is a three-fold attack, but
White finds an astonishing solution .
1 � cS ! . . .
'This compelling move is a double
co unterattack : The pawn is under
. bishop and the p awn ; the pawn
fire and White threatens perpetual
check with his knight.
1 . . . f l (®) 2 � h6 + . . .
'Thi � � ove also has its strong points,
fo r 1t IS a combination of defence
(the knight escapes from the threat)
and attack (the king is pushed into
_th e corner from which there is no
es cape) .
2 . . . dlh8 3 � d6 !
ble defensive contacts, which frus­
trate all attempts by Black to bring
his overwhelming material superi­
ority to bear. If the knight g7
moves, White forces it to return to
its original square by giving check
with the bishop at eS. In this posi­
tion the bishop does the attacking,
pr�tects the pawn and the bishop,
while the pawn supports the knight
and deprives the queen of square f6.
This position comprises one attack­
ing contact (the bishop threatens to
attack the king) , two protective
contacts (the knight protects the
covers the knight) and three confin­
ing contacts (the bishop hems in
Black's knight ; the pawn prevents
the queen from moving to square
f6 ; the knight pins down the king) .
All the white king has to do is not to
interfere in its forces' work to­
wards a draw.

1 98
Contacts with their confining contacts. The
ratio of contacts in relation to the
and combinations
contested square is 2 :2. A shaky
with passed pawns equilibrium prevails for the time be­
ing. But White has a knight that he
In this chapter we shall try to work can use to upset it in his favour. He
out the characteristic features of does this by playing 1 � c7 . The po­
combinations of contacts that occur sition is typical of a double attack.
in the struggle between passed White's knight attacks his black op­
pawns and different pieces. We al­ posite number and at the same time
ready know that in end-games with sets up the concealed threat of
pawns it is necessary to have, in ad­ queening if Black takes his knight.
dition to a threatening (active) con­ Black has no choice but to reply
tact between the pawn and its 1 . . � b 8 , which is followed by
.

queening square, a protective and a another double attack 2 � bS + ,

confining contact with this square this time a decisive one. White at­
in order to prevent the opponent's tacks the enemy king, whilst his
king from stopping the pawn. But own king is about to capture the
what happens in other kinds of end­ knight. Black's king has to retreat
game ? and his knight is lost.
The following position is character­ If White had a white-square bishop
istic of end-games with knights. i nstead of a knight, his road to vic­
tory would also be uncomplicated.
41 9

420

White to play and win

Here we have the active contact be­
tween the pawn and the queening
square as well as the protective and
confining contact between the king
and this square. But this still is not
enough for a win. Black's king and
knight prevent the pawn from
queening by acting on square b 8
White to play and win
By 1 � c4 White drives the knight
to b8 and by 2 � bS ! he creates con­
fining contacts round the hapless
knight, which has no square it can
move to and the game is over. As
you see, a combined attack (which

1 99
we are already familiar with) devel ­
ops in which an attacking contact
(the king attacks the knight) and
various confining contacts (the
bishop is fully in control of the
knight) occur.
In the struggle between a knight
and a pawn on the one hand and a
bishop on the other it is very im­
portant that the knight should
establish a confining contact before
the king can come to the bishop's
assistance .
421
Black to play, White wins
White wins without trouble in this
elementary position. He threatens
to play the knight to c7, thereby
severing the contact between the
bishop and square b 8 . That is why
Black has to play 1 . . �b8 him­
. can see for yourself that all final po­
self. But even after this move 2 � c7
wins. White completely immobilizes
the bishop by this confining contact
and gets an immediate win.
In such a position Black can refute
the confining contact only if his
king can assert itself on square c6 or
b6.
And here is the ending of an e le­
mentary end-game with rooks.
200
422
White to play and win
White retains the upper hand by 1
.E1 a8 + cbg7 2 e7 00f7 3 El f8 + .
Even after playing 2 e7 White can
· already threaten two contacts : the
queening threat and the protective
contact of the castle. The best Black
can do is make a confining contact
with his rook. Thus, by playing
2 . . 00f7 Black restores the equi­
.
librium of the contacts with regard
to square e8 for the time being. But
after 3 El f8 + , by pushing aside the
king, White establishes a confining
contact, at the same time restoring
the 2 : 1 ratio of contacts at square
e 8 . Here too we see a combined at­
tack In the final position consisting
of the active contact at square e8
and the contact in which the white
rook hems in the black king. You
sitions in which passed pawns reach
the queening rank are characterized
by combinations of contacts,
namely either by a combined attack
and the associated contacts or by a
double attack.
Up till now we have examined com­
binations of contacts at a certain
point in time, statically as it were.
But now we shall study the dynam-
ics of this question on the basis of
certain types of position. We shall
see how contacts and their combi­
nations come into being in the
course of a game and what bearing
they have on the course of events
and on the final result. We shatl an­
alyse positions in which a rook has
to contend with a far-advanced
pawn, but the rook is so badly
placed that it cannot stop the pawn
directly and the rook has to look for
other ways to establish contact with
the queening square.
After A. Troitzky, 1 8 95
423
Black to play, White to win
Black finds himself in a precarious
position. The pawn threatens to
queen and his own king blocks his
rook's access to the 8 th rank. The
contacts that have come into being
are marked in the diagram. The
pawn's active contact with square
d8 is marked with an arrow ; the
square dS watched over by the king
is marked with a cross, whilst the
squ ares controlled by Black's king
are marked with dashes. I did not
mark all squares controlled by the
pieces but only those that are of
consequence in the clash that is
about to take place.
How should Black continue ? In
desperation he attacks the king.
1 . . . bl g6 + 2 �eS � . . .
The check was by no means as
h armless as it looked at first sight.
Actually it is an attack that is cou ­
pled with an attacking threat and a
concealed threat of a double attack.
If White had answered with 2 d:>fs
Black could have played 2 . . .
bl d6. But the natural 2 ctde7 would
have cost White a win, because
White's king would have entered
the zone of the hostile king' s con­
fining contacts . This could have led
to 2 . . . bl g1 � and after 3 d 8 ( ® )
bl e 1 + the king would have had to
step on the cl-file, which would have
resulted in a double attack and a
draw.
2 . . . bl gS + 3 �e4 bl g4 + 4 d!e3
bl g3 + 5 �d2 bl g2 + 6 �c3
Since the king is moving along the
c-file, it keeps an eye on the cl-file
and finally it hides on square c7.
I n the case discussed above the
rock's own king prevents the rook,
through a negative confining con­
tact, from getting at the pawn .
424
Black to play
20 1
Again the essential contacts are
marked in the diagram. The pawn is
in active contact with squ are d 8
and-thanks t o t h e fact that the
rook is posted on the e-file-it is in
confining contact with square e 8 .
Just as in the preceding example,
White's king controls the squares of
the cl-file. Here too Black finds
himself in a q-itical situation, for thepened if Black's king had stood on
pawn cannot be stopped by normal
means. But he can start harassing
the white king with checks.
1 . . . E! e6 + 2 ct>c5 ! . . .
we have succeeded in describing the
events taking place on the board
with precision.
The characteristic feature here is
that the king does not participate in
any way in the events. By moving
to d2 White's king safely evades the
squares controlled by his counter­
part. But what would have hap­
b2 i nstead of b 1 ? In this case square
c3 would have been within his
sphere of influence. Up to a certain
point the game developed as in the
previous example : 1 . . E! e6 + 2
.

Again White has to think very care­ <Jdc5 E! e5 + 3 <Jdc4 E! e4 + and ,

fully before replying. For instance if now White should play his king to
2 <Jdc7 then 2 . . . E! e7 with an im­ d3 because square c3 is controlled
mediate draw was possible. It would by the opponent's king.
also have been wrong to try attack­ Black's reply to this double attack is
ing the rook by 2 <Jdd5 because of 4 . . EI e 1 with the threat of
.

2 . . . E! e 1 , and in view of the launching a double attack of his

threatened attack the rook has ma­ own . But White defends himself
naged to establish contact with the with a second double attack by 5
pawn's queening square . ffid2 against which Black is finally
powerless. This means that the con­
2 . . . E! e5 + 3 ffic4 E! e4 + 4 <Jdc3
fining contacts of the black king at
E! e3 + 5 ffid2 !
b2 do not suffice to upset White's
This is the objective White has pur­ plan and to support the rook. The
sued. Only now does he threaten game takes an altogether different
the rook, making use of a typical turn if Black's king is posted on b3.
double attack : the pawn threatens to
queen (first active contact) and the
king attacks the rook (second active 425

contact) . This double attack is deci­

sive, for Black has no way out. We
have seen that in the position shown
in diagram 424 ; White's plan is to
prevent Black from harassing his
king with the rook. White wards off
the attack with a manoeuvre of the
king that culminates in a double at­
tack. In an alysing these positions

202
In this altered position, square c4 is
not accessible to White, and he can­
not step on d4 because of the rook
move to e 1. This means that with
the black king on b3 the game ends
with a draw through perpetual
check. Please note that in this per­
petual check position the active
contacts cancel each other as it
were : the queening threat (an active
contact of White) is cancelled by
the attack on the king by the rook
(an active contact of Black) .
What if Black's king were posted on
a4 ? It seems to be unable to influ­
ence the course of events from
where it stands . But this is not so.
After 1 . . � e6 + 2 dieS � eS + 3
.
dlc4 the following position ensues :
426
Black to play
The confining contacts emanating
from the black king prevent the
white king from fleeing to the b­
file. Taking advantage of this Black
immediately moves his rook to e 1
because he realizes that after 4
d8(®) � c l + and 5 . . � d 1 + he
.
loses his queen and the game.
These examples s how that a strong
chess-player must be able not only
to calculate the variants with preci-
sion but also plan ahead, to recog­
nize the developing contacts in time
and to assess the development of
events on the basis of these contacts
and their changes. The following
study for instance, which is known
throughout the world and which we
get by shifting the position shown in
diagram 424 one file to the left, is
characterized by a very instructive
change of contacts.
F. Saavedra, 18 9 5
427
Black to play
At first events develop as in diagram
424, even the contacts of the two
sets of position being the same, i. e.
1 . . . � d6 + 2 <t?bs � ds + 3 dlb4
� d4 + 4 dlb3 �d3 + s <t?c2 . . .
As in the preceding position White
entangles his opponent in a double
attack. Note the posting of the
kings . White's king placed at c2 has
rendered two of the three free
squares inaccessible to its opposite
number. At this point one might
well decide to play for a draw.
5 . . . bl d4 !
203
Black evades the attack and sets up B y attacking the rook White's king
a cunning trap at the same time : if 6 changes to another front, confining
c8(®) then 6 . . . rlc4 + ! ! 7 other squares than ltefore. The c­
® x c4 draw ! The queen had to file becomes free and a second ac­
take the rook because of the double tive contact is established, i. e. the
attack. But let us stay yet a little mating threat at c l .
while with the position after 5 . . . The final position entails a double
ri d4. atta.ck again, i. e . an attack coupled
428
with a mating threat.
If the black king in diagram 427
was posted on b 1 , his rival could not
reach c2 at all and the game would
end with a draw through perpetual
check. In addition, Black could use
an alternative line of play to achieve
a draw. Instead of 1 . . r1 d6 + he
.

could also play 1 . . B d2 !, al­

.

though this would enable White to

create a fresh queen by 2 c8 (® ).
What happens if we choose to take
a rook instead of a queen ? Let us try 430

it!
6 c8 ( ri ) ! . . .
There is only one line of defence
against the threatened mate on a8 .
6 . . . B a4
It is not hard now to find the move
that will decide the issue.
7 00b3

429
Black utilizes his king's control of
squares a2, b2 and c2 and gives per­
petual check, as White's king can­
not cross the c-file because of a
double attack.
A similar situation enables Black to
achieve a draw in the following
seemingly hopeless position.

Black to play

204
 431 The solution begins with a seem­
ingly senseless move 1 digS !, al­
though from the point of view of
contacts it is completely natural.
Thinking ahead, White starts mak­
ing preparations for a draw the me­
chanics of which we are already
familiar with, by vacating square h7
for his rook. After 1 . . f2 2 .El e7 +
.

<j)f3 3 .El f7 + \tlg2 4 .El g7 + 00h3

5 .El f7 00g2 6 .El g7 + \tlfl 7 .El h7
White to play
we have the same position as in the
preceding study except that the
White plays 1 a6 and tries to colours are reversed.
engineer a double attack by means
of a pawn sacrifice. Black accepts The following study proves that the
the bait by 1 . .El x a6 . After 2
. .
same perpetual check can also be
00b7 a standard situation of piece given horizontally.
combination ensues : the king is at­
�c�ngilie rook andilie pawn G. Afanasyev, 19 50
threatens to queen. But Black frus­ Conclusion ofa study
trates his opponent's plan by re­
433
covering the rook after a2 and giving
perpetual check after 3 c 8 ( ® ) .
Now that you have made the
acquaintance of this draw mechan­
ism, the following study will cer­
tainly not come as a surprise to you.

� Korolkov, 1 95 0

432
Draw

Queening cannot be prevented. Af­

ter 1 .El d4 + \tle2 2 .El e4 + \tlf2 3
.El f4 + \tlg2 the king can no longer
be molested by checks ( 4 .El g4 +
would be a b ad mistake in view of
4 . . . 00f3 since the pawn at h2 con­
trols the square g l ) . But thanks to
the fact that the king at g2 and the
Draw pawn at h3 effectively hem in the

205
newly arisen queen, White makes a
splendid move : 4 .!d. b4 ! The threat
of intercepting the pawn forces
Black to play 4 . . . h l (®). But at
this point the concealed threat of
perpetual check comes into play : 5
.!d. b l ® x h3 (5 . . . ®h2 6 .!d.b2 + )
6 .b! b2 + , and we have reached an
analogous position, even if the file
h l -h8 is only a mirror image. By
shuttling back and forth between
the squares b 1 and b2 the rook
keeps pestering the opponent's king
with interminable checks .
In this section we have examined
the tactical peculiarities of a num­
ber of positions characterized
mainly by a struggle between a
passed pawn approaching the
queening rank and a rook trying to
stop it. The concept of contacts and
their combinations I have intro­
duced were used in analysing the
struggle. With their help the essence
of the events taking place on the
board was described exactly and
impressively.
From this we can draw the import­
ant conclusion that the proposed
theory of contacts and combina­
tions helps analyse a wide variety of
tactical situations and makes for a
better understanding of the course
of the struggle.
Rules prohibit the noting down of
contacts during the game. But by
bearing in mind the contacts and
remembering how they come into
being and what combinations they
give rise to, you will find it easier to
find your way through the various
complex constellations on the
. b oard . But the theory of contacts
206
and combiQations is significant not
only because it helps explain the es­
sence of the events taking place on
the chessboard . Far from it ! I shall
soon demonstrate that as a rule con­
tacts occur even before a combina­
tion materializes and that each
combinational ending is highlighted
by a certain co-ordinated combina­
tion of contacts. One can hardly
overestimate the meaning of these
two facts. First we are able to recog­
nize the evolution of combinations
in advance by carefully noting the
occurrence of contacts. And second­
ly, once we have an idea of which
combinations of contacts are char­
acteristic of certain combinations
we can prepare such combinations
in advance or frustrate them by de­
stroing the contacts. In either case
one thing is clear : the theory of
contacts and combinations is a very
effective means of obtaining a bet­
ter u nderstanding of the course of
the struggle.
Wh at is a combination?
The word "combination" is bound
to occur in any chess book we h ap­
pen to come across. This term is
widely used in chess literature. But
when it comes to giving a precise
definition of combination we
usually notice that by no means
everybody uses this term to mean
the same thing.
The question that springs to mind is
why we need a precise definition of
combination in the first place,
whether the fact that different peo-
pie have different interp retations of
the term is so important. The ques­
tion is j ustifiable. But this problem is
not as simple as it may appear at
first sight. First of all , the important
thing here is not so much the ter­
minology as the classific ation.
There can be no u nanimity in the
classification of combinations if the
term "combinatio n" is g iven differ­
ent interpretatio ns. And at present
this is in fact the case. Browsing
through books on the theory of the
middle game we soon d iscover that
every author has his own classifica­
tion. This tends to confu se many
who want to learn the middle game
and to improve their skill. O ne of
the objectives I have set out to
achieve in this work is to introduce
some order into this rather con­
fused state of affairs. Above all I
want to help you, the re ader, to as­
cend the steps to chess mastery and
to acquire the art of combination,
which is seen as the pinnacle of
creative chess.
The word "combination", as used in
our everyday speech , is derived
from the Latin word "combi n:1.tio".
In chess, the word combinatior-.
should be applied to connections of
pieces and pawns of the two oppos­
ing sides. You have seen on numer­
ous occasions that the most diverse
and at times very remarkable con­
nections of pieces can o ccur on the
chessboard. Pioneers of chess the­
ory have tried to fathom the laws
�overning the outwardly so plain
chessmen from the very beginning
of our game's development. Some
positions i nvolving different con-
nections or combinations of pieces
and pawns led them to an important
observation. When a player carried
out a well-planned operation, it
happened occasionally that his op­
ponent's responses were forced and
he was unable to influence the in­
exorable course of events.
I n such situations the pieces and
pawns of both sides seem to hang
on invisible threads. While obey­
i ng-like marionets-the will of the
player, they circled in the whirlpool
of a strange "dance". This original
connection of pawns and pieces,
which enabled one of the contes ­
tants to regroup his pieces and
pawns in a certain manner, can be
seen as a combination. But this de­
finition is too static. It lays stress on
the original position in which the
future, often surprising, events can
be deeply hidden and difficult to
understand rather than on the on­
going course of events . It was there­
fore not surprising that gradually
the term "combination" ceased to
be used i n con nection with the start­
ing position and with the original
connection of pieces and pawns,
despite the fact that its name derives
from it. One started using the term
to describe changes in connections
between the pieces and pawns of
both s ides . But any regrouping of
pieces leads up to variants or even
up to constellations of variants.
Finally the following formulation
crystallized : " . . . if a constellation
of variants contains a positive result
that calls for attention, then the to­
tality of these variants is called a
'combination', and the player who
207

discovers such a variant and puts it
into action is said to realize the
combination" (Em. Lasker, "Man-
ual of Chess", 1 925).
/
a6 ddd5 3 a7 ddc6 4 a S( �) + etc.
But if it is Black's turn to play, then
the king catches up with the pawn :
1 . ,dde4 2 a5 ddd5 3 a6 ddc6 4 a7
. .

Somewhat later Romanovski pro-
posed a more precise formulation of
Lasker's definition : "A combination
is a variant (or a group of variants)
in the course of which both players
make forced moves and which end
in an objective advantage for the ac-
tive side" (Romanovski : The Mid-
die Game, 1 929) .
You will note that both of these de-
finitions have been completely dis-
associated from the connection of
pieces and pawns, but they have re-
tained two essential features con-
nected with the combination: the
forced moves and the winning of an
advantage by the side carrying out
the combination.
But is such a definition not too
broad, especially in view of the fact
that it encompasses situations that
can hardly be seen as combinations ?
Let us consider, for example, the
following position illustrating the
quadrangle rule.
00b7. In both cases the moves are
forced (the pawn flees, the king
pursues it) , and there is also an ob-
jective (White wants to get his pawn
queened, Black wants to stop it) .
· But for all that one can by no means
regard this elementary tactical op-
eration, which could be called pur-
suit, as a combination, because the
m ain ingredient, namely the con-
nection is missing. The only thing
we have here is a piece and a pawn
changing their position, which does
not qualify for a connection, be-
cause connection in the true sense
involves at least three objects.
435

434

White to play

The first player wins by not permit­

ting the black king to gain control
of square b7 :
1 ddb3 \t>d4 2 \t>b4 ddds 3 ddbs
ddd6 4 ddb6 \t>d7 5 \t>b7 \t>d6 6 a4
ddcs 7 aS 00b5 8 a6 etc.
The sequence of moves in this end­
game consists of two phases, both
If White is to play, his pawn reaches being characterized by forced
the queening square by 1 aS \t>e4 2 moves. In the first phase, a struggle

2 08
flares up over the control of square
b7, in the second phase the pawn
advances to the queening square. In
the end White won an objective ad­
vantage, having succeeded in
queening his pawn. Nevertheless,
this cannot be seen as a comb ina­
tion either, because the essential
feature of a combination is the con­
nection of pieces and pawns. But all
we had was the kings changing their
position in the first phase and a king
and pawn changing theirs in the
second.
I hope you will have acknowledged
the main deficiency in Lasker' s and
Romanovki's definition. It is too
broad and covers many positions
which contain no combin ations.
No wonder that this definition
prompted criticism from Botvinnik
as early as 1 9 39, who proposed a
defi nition of his own, which at first
sight looked very clear and exact. It In both positions, White check­
is generally accepted that in posi­ mates his opponent in three moves,
tions giving rise to combinations the in the first by 1 El a 8 + � b 8 (� b 8 )
values of pieces u ndergo a complete 2 El X b 8 + � X b 8 ( � X b8) 3
change. In such positions we readily ld d8 mate, in the second by 1
sacrifice material to achieve our ob­ bl a B + ld b 8 ( 1 . . . � b 8 2 ld d8
jectives. For this reason Botvi nnik mate) 2 El x b8 + � x b8 3 El d 8
used the sacrifice as the basis of his mate .
definition. "The combin ation is a According to Botvinnik, the first
forced variant with sacrifices," is position contains a combination,
how he put it. Botvinnik's sugges­ but not the second, because only an
tion was accepted by the chess com­ exchange-not a sacrifice-occurs
mu nity with applause, but later it in it. But it is not necessary to prove
was found that this defi nition is not that basically there is no difference
perfect either. Let us examine the between the two positions. Both po­
following two positions for in­ sitions end in mate, the mating me­
stance. chanism being the same in both
cases (the king is mated by a rook
s upported by a bishop) , the black
pawn making a confining contact in

209
both positions. The only difference R . Rhi, 1 922
is in the way the m ate ending is in­ Conclusion ofa study
itiated : In the first case White sacri­
438
fices an exchange in order to divert
Black's pieces from the defence of
square d8, in the second case the
knight is diverted either by an at­
tack (if Black's knight moves to b8)
or by an exchange (if the rook
moves to b 8 ) . As we shall show the
sacrifice is the strongest means of
forcing moves, but there are also
other means available in tactical
clashes, such as threats, attacks, Black to play
double attacks and exchanges.
To tell the truth, another reason that In essence this move is a typical
I don't like Botvinnik's definition is ·double attack. White threatens to
because he depreciates the end­ intercept Black' s p awn while giving
game and makes it appear less chal­ s upport to his own . BL:�.ck is power­
lenging than the middle game to less against this double attack. The
many chess enthusiasts. It is gen­ ensuing contacts are marked in the
erally known that sacrifices occur diagram, namely a protective con­
much more rarely in the end-game tact (threatening to support the
than in the middle game. But Bot­ white pawn) and a confining con­
vinnik's formulation creates the im­ tact (threatening to intercept the
pression-intentionally or not-that black pawn) . An interrelation has
combinations are rare in the end­ come into being between the king
game. And m any end-game studies . and the two pawns, and this is tan­
have been devoted to this view. But tamount to a combination on the
is it tru e ? I n an article "On the ques­ subject of double attack. As you
tion of defining the combination" know, this combination is basically
Botvinnik quotes a well-known the same thing as a common "fork",
study by Reti and maintains that it the only difference being that in the
is devoid of any combinations and case of a "fork" there are normally
that it contains merely an impres­ two attacks, here we have two
sive manoeuvre. I believe this asser­ threats.
tion to be wrong. We have dis­ Those who accept our conclusion
cussed this study of Reti's in depth will concede that Botvinnik's defini­
in the first part of the book, and at tion is too narrow. It fails to cover
this point let me only call to mind m any positions which have all the
the position that results towards the features of combinations. Botvinnik
end of its solution after d;Jf6-e5! is right only to the extent that mid­
dle game combinations are often ac-

210
companied by sacrifices. But the
sacrifice is not a determining factor,
it is only an attendant circumstance
of a combin ation. It would be incor­
rect, for instance, to say that a com­
bination is involved every time there
is a sacrifice. After all, we are famil­
iar with sacrifices made for the sake
of initiative or of development, etc.,
which can by no means be regarded
as combinations. The converse of
this-where combinations are made
there have to be sacrifices-does not
hold true either, of course.
From all this it follows that, of the
two definitions of the combination
we have, one is too broad and the
other too narrow. But the two can
easily be fused into a third which
seems to us to be the most exact
one. Before acquainting you with
the unified formulation, I should
like to point out that in seeking a
definition of the term "combina­
tion" it would be methodically un­
acceptable to start out from the ori­
ginal point of departure, i. e. from
the semantic meaning of the word­
connection, chain of moves, var­
i ant-because variants reflect
changes of position of pieces and
pawns, i. e . changes in their connec­
tions. But in combinations, the con­
nection of some pieces of both sides
leads to a forced final connection
which is characterized by the fact
that one of the two opponents gains
a n objective advantage. And that is
basically a definition, it only has to
he formulated with greater preci­
�ion. We know that the ending of a
winning combination can be mate,
1 he capture of a piece or the queen-
ing of a pawn. The ending of a
drawing combination, on the other
hand, can be a stalemate, perpetual
check, fortification, blockade, etc.
We have made a thorough analysis
of these endings o n pages 1 7 8 to
206 and fou nd that all of them are
characterized by a co-ordinated
connection of contacts. Thus we
may formulate the defi nition as fol­
lows :
A combination is a rearrangement
of the connection of pieces of both
sides, which forces a co-ordinated
connection of contacts , which is ad­
vantageous to one side."
We shall need the definition of this
term for classifying combinations.
We shall notice that despite the
large number of different combina­
tions, it is very easy to classify them
according to the final connection of
contacts .
Another question that arises in con ­
nection with the defi nition of a
combination is whether or not it
should take into acco u nt aesthetic
factors . Some authors are in favour
of this, arguing that combinations
appeal to our aesthetic feelings and
are often accompanied by effective,
beautiful and u nexpected moves.
But I am of the opinion that we can
neglect this aspect. The combina­
tion has to be objective and it
should be b ased on positional fea­
tures . The beauty of a combination
o n the other hand is a purely subjec­
tive factor, being influenced by per­
sonal taste and views as well as by
the system of assessment used. A
combination can be elementary,
common or p urely technical, but
21 1
even if it fails to fascinate one it is
still a combination. The only thing
to be said against such combina­
tions is that they win no beauty
pnzes.
Various means
of applying force
The pieces involved in playing out a
combination are not only one's
own , but also those of one's oppo­
1. The attack
By attacking a lone piece we can
force it to retreat and occupy
squ ares we want it to occupy. The
following examples s how how the
attack can be used as a means of
conj uring up a mate ending.
A. Herbstm an, 1 945
Con clusion ofa study
439

nent-and not only as objects of at­

tack. "T hese pieces can sometimes
play a negative, confining role, de­
serting, as it were, their own ranks
and coming over to fight on our
side. We have already come across
such situations in studying the co­
ordination of piece games. Now we
want to find out what means we can
use to force our opponent's pieces
to come over to our side. How can
we make them help us in achieving
our objectives or at least prevent
them from interfering? We shall be
dealing with this question i n the
present chapter. We shall examine
various means of forcing enemy
pieces to occupy those squ ares we
want them to occupy or to neglect
important defence tasks. You know
of course that attacks and threats of
all kinds, exchanges and Zu gzwang
are such means of applying force,
White to pla y and win
White can take the pawn at g6, but
by so doing he allows Black' s knight
to regain its freedom , and the in­
significant material advantage can­
not be converted. The winning
move is 1 f3 ! In this m an ner White
hems in the knight well in advance
and prepares for a m ate ending. Af­
ter the only possible reply 1 . . .
� h6 + White launches a double at­
tack whith 2 dl x g6, threatening
the knight, and if 2 . � g8 then
. .

but the most lasting effects are he threatens mate by 3 � f7 .

achieved with sacrifices.
Let us examine these means in the
order mentioned.

212
S. Birnov By playing 1 dlg2 dlg4 2 Eld6 ! and
Conclusion ofa study thereby threatening a double at­
440
tack, White prevented his opponent
from playing 2 . . . dlf4. Instead,
,Black answered with 2 . . . � h7.
This is followed by 3 dlf2 dlf4 4
El h6 ! This move is actually a triple
attack. White attacks the knight,
prepares for the double attack 4 . . .
� f8 5 Elf6 + and if 4 . . . � g5
suddenly makes the third threat by
5 Et h4 mate !

White to play and win Kasparyan's study also shows a

marvellous mate finish.
White attacks the knight by 1 dlc3
and forces it to retreat 1 . . . � a6. G. Kasp aryan, 1 961
This is followed by 2 Et b 1 + . O nce
again we see a double attack, the 442

bishop attacking the knight at a8

and the rook assaulting the king.
The only way to evade the attack
and to cover the knight is by pl ayi ng
2 . . . dla7, but then White check­
mates h is opponent's king by 3
Elb7.
As c an be s een in the following
study, occasionally mate endings
occur unexpectedly. White to play and win

A. Gurvitch, 1 958 First of all White gets a double at­

tack going by 1 ® c5 + <1/a6 2 ®c2.
441
Black defends himself by 2 . . .
.b1 bS + 3 \t>c4 � c7 . Then White,
by d ifferent manoeuvres, forces
Black's pieces to move to squares
from which they impede their king's
movements : 4 ® a4 + El aS 5
'® c6 + �b6, to deliver the irrefut­
able final blow 6 ®c8 .

White to play and win

213
H. Rinck, 1 924 Z ita-Bronstein
443
Vienn a 1 95 7
444

White to play and win

Here White forces his opponent's
pieces to take up ineffective posi­
tions by a series of rook attacks,
winning an advantage in material in
the end.
Black to play
The second player can immediately
u nleash a double attack, but it is not
sufficiently effective yet, because
White's ·reply to 1 . . .b1 f2 + is 2
.

1 .b1 a7 �fl (1 . . . � c8 2 .b1 a8 ) 2 ® d4. But Black found a way of in­

.b1a2 �h8 3 .b1a8 + ! 00g7 4 .b!a l , creasing its effectiveness. He
and the bishop f l finds no free played :
square to move to.
Black's black-square bishop moved 1 . . c6 ! 2 .b1d4 . . .
.

to h8 in order to avoid any unfore­ He has no choice ; any other reply

seen complications. White managed loses immediately.
to neutralize it and to capture the
other bishop. But Black did not 2 . . . cS ! 3 .b1 d5 .b1 f2 + , and White
have a better option. If, for in- . resigned.
stance, 2 .. . � c l , then 3 .b1 f2 with
a double att�ck (attack on fl plus Capablanca-Marshall
threatened 4 � d3 + ) . If, on the New York 1 909
other hand, 2 . . . � f6, then 3 .b1f2 445

� e7 + 4 00b3 � b s 5 � dS + 00e8
6 � c7 + . And finally, if 2 . . . � g7 ,
then White again plays 3 .b1 f2, and
after 3 . . . � a6 4 � e6 + 00g8 5
.b1 g2, capturing the bishop at g7.
Not to arouse the impression that
attack is used as a means of forcing
moves only in problems, I should
like to include two examples from
actual games. Black to play

214
The white castle's posting at a4 is
not exactly good, but the first
player is confident that he can re­
fute the Black bishop's attack by 2
�bS . But Black harassed the rook
by 1 . � d7 2 �bS �fS ! Let us
. .
try to unravel the motives behind
this move . It is easy enough to see
that the effect produced by this
move is a very lasting one indeed.
Black has lured the bishop to bS in
order to win a tempo, retreat and
attack the other rook. At the same
time he threatens to drive away the
bishop by a7 -a6 in order to move
his bishop back to d7 and to win the
exchange after all . There is no de­
fence. After 3 bl b2 a6 ! 4 � e 2 � d7,
Black obtained a decisive material
advantage thanks to his combina­
tion.
2. The threat
Experience shows that a strong
threat is j ust as dangerous as an at­ mate in one move. Apart from mov­
tack. The following delightful prob­ ing the knight at c7, any move with
lem demonstrates the power threats the knight at c6 results in mate. But
can have.
C. Kipping
446
Mate in three moves
White's pieces ominously converge
on the black king, but for the time
being there is no mate in sight. And
without the white ki ng's support
there can be no mate. For this rea­
son :
1 <i>as ! ! . . .
Unintimid ated by the pawn which is
about to queen, White's king is in­
tent on establishing a protective
contact with his pieces. Two threats
suddenly loom large after this
move : 2 � d4 + <1>a7 3 � bS mate
and 2 � e7 + <1>a7 3 � c8 mate .
But we must not forget that Black
can queen and give check at the
same ti me !
1 . . . e t (® ) + 2 \t>b6 ! . . .
A unique position ! Black has mate­
rial superiority, but White has a
whole handful of threats to com­
pensate for it. That is why Black is
unable to defend himself against
Black cannot refute all threats with
a single move.
It is interesting to note that 1 ct>bs
would not have led to mate in three
moves, because Black could have
defended himself by 1 . . . bl g8 2
®b6 bl c8 . On the other hand, 1
<1>a5 bl g8 is followed by 2 � d4 +
\t>a7 3 � bS mate .
And here is a study that reveals the
power of the threat.
215
G. Nadareishvili, 1 94 7 H Rinck, 1 92 1
448
447

White to play and win

White to play and win
White seems to be unable to gain
the upper hand because the two
sides are approximately level. But
his first move initiates a double at­
tack which immediately puts Black
in a critical situation.
1 � c3 ! ! . . .
The rook finds no free squares on
the fifth rank, because 1 . . .Id. g5 is e ach other, the rook cannot be in­
.
The forces of the two sides are bal­
anced, but thanks to the active post­
ing of his pieces, White succeeds in
creating decisive mating threats.
1 .!d.e7 + . . .
In principle this is a double attack,
b ecause one rook trains its sights at
the hostile rook and the other at the
king. Since the black rooks cover

followed by the decisive 2 � e6 + , terposed to shield the king, and the

whilst 1 . . . .Id. d5 by 2 � f5 + \t>g8 l atter has to retreat.
3 � e7 + . Hoping for better luck on
the eighth rank Black plays 1 . . .c1>h 8 2c1>h6 ! . . .
1 . . . lj, a8 2 \t>b7 lj, f8 A fresh double attack. Keeping the
rook at d8 under fire, White threat­
With considerable difficulties, the ens to give mate with the other rook
rook m anaged to shake off its assai­ at h7. Since 2 . . .Id. gf8 is out of the
lants. But by so doing it blocked a
.

question because of 3 .!d.h7 + \t>g8

square that was vital for the ki ?g, so 4 .!d.g l mate, Black has but one re­
that a familiar m ating connection p ly.
comes into being after 3 � f5 +
\t>g8 4 � h6 m ate. 2 . . . rl ge 8 3 lj,dd7 ! . . .
A threat against which there is no
defence. The following finish is
conceivable: 3 . . . \t>g8 4 .Id. g7 +
c1>h8 5 rl h7 + c1>g8 6 .!d.dg7 +
\t> f8 7 .Id. h8 mate.

216
In this position too White succeeds 7 . . . a 1 (®) 8 .El af6 mate.
in entangling the opponent's king in
Now let us see what happens if in
a mating net with the aid of threats.
the th ird move the king had moved
to c6.
A. Votava, 1 95 2
449
4 .El c8 + 00b7
If 4 . . 00b6, then 5 .El x a2 b 1 (®)
.

6 .El b8 + , w�ilst 4 . . 00d6 is fol­

.

lowed by S &:! acS ! 00e6 6 00e8 00f6

7 &:! 8 c6 + 00g7 8 .El gS + 00h7 9
00f7 with mate on the next move.
S ri bS + ! 00a7 6 00c7 ! a 1 (®) 7
.El b7 + 00a6 8 .El a8 mate.
It is fascinating to see how White
White to play and win manages to set up a mating mechan­
ism at different ends of the board
Black's pawns look impressive. But through threats .
White's attacking potential is even
greater.
3. The exchange
1 .El haS ! . . .
An immediate mating threat. The The exchange is an important
king has to flee. means of exerting force in combina­
tional engagements. In this way one
1 . . 00c6 2 00c8 (again the threat) can, for instance, liquidate hostile
.

2 . . . 00d6 3 00d8 00e6 . pieces that have defensive functions

We shall examine the possibility of by exchanges, thereby establishing
the king's return to c6 after analys­ mating connections.
ing this line of play. The following study provides an ex­
ample.
4 .E18a6 + ! 00f7 s .El fs + 00g7 6
.El gs + OOf7
If 6 . . . c;1>h7, then 7 .E1 x a2 !
b 1 ( ® ) 8 .El h2 m ate. But at this
point Black's king is caught in the
mating net.
7 .El gg6 ! . . .
The confining contacts h ave been
established. All White has to do is
deliver the decisive blow.

217
G. Kasparyan, 1 9 5 9 A. Gurvitch, 1 92 9
Conclusion ofa study Conclusion ofa study
451
450

White to play and win

White to play and win
1 � e 3 ! � c6
First of all White encages the bishop
The only defence against the bishop
by 1 El h7 and forces Black's king to
check that follows.
move off : 1 . . . d/f6. He then cuts
off its retreat by 2 d/f4. For the time 2 � cS + � b4 3 � d 3 aS
being the rook checkat h6 does not
Everything is adequately protected
look very dangerous, because Black
on Black's side. White's attack
can interpose the bishop at g6 . But
seems to have petered out. But its
how should he continue ? He tries to
successful continuation lies in the
lau nch a counterattack by 2 . . .
exchange.
El b7. But White is on guard: 3
� c4 ! A precise move. If White had 4� x b4 + ! ab 5 El h 8 rl e6
played 3 � x f7, Black would have
A different line of defence is not in
forced the king aside by 3 . . .
sight. After 5 . b2 6 El a8 + dlb3
El b4 + . But if Black goes 3 . . .
. .

7 � cS + Black's rook falls prey to a

El a7, then 4 � d6 and White wins a
piece. Hence 3 . . . El d7, but this is
followed by 4 � x f7 . The ex­
change of bishops decides the game
here : 4 . . . El x f7 5 El h6 mate.
Sometimes it is possible to build up
a mating mechanism through ex­
change operations.
fork.
6 El a8 + rl a6 7 El x a6 + ! . . .
With this outwardly inconspicuous
move White sets u p an instructive
m ating mechanism in which pawns
are at fault for all confining con­
tacts.
7 . . . ba 8 � b 2 ! aS 9 d/a 1 . . .
In the end it is Zugzwang that de­
cides the game. Black is. forced to
shut the lid of his own coffin.

218
9 . . . a4 1 0 � c4 mate. We should add that White would
have given mate in a similar manner
Occasionally the king or some
after 4 . . . dlh l : 5 <1> x f2 dl x h2 6
other enemy piece can be steered to
� g4 +! dlh 1 7 dlfl h2 8 � f2
a bad square through exchange.
mate.
G. Bagdasaryan, 1 938
452
4. Zugzwang
In the two preceding examples we
have seen how Zugzwang is used to
create a mating mechanism. And
now I should like to acquaint you
with perhaps one of the most im­
pressive examples of Zugzwang.

f. Sehwers, 1 902
Conclusion ofa study
White to play and win 453

There is no obvious way for White

to win. Black's king threatens to
move to g2, to drive off the bishop
and to advance the pawn to the
queening square. For this reason the
first moves might seem somewhat
odd, for White goads Black's king
precisely where Black himself wants
to have him.
White to play and win
1 � e3 + dlg3 2 � f1 + dlg2 3
� h2 ! � f2 Black has superiority in material,
but the whole extent of his tragi­
White has forced the knight to va­
comic situ ation comes to light after
cate the corner. But now comes the
1 � c3 + dlc l 2 �cS ! He is in
crucial manoeuvre :
Zugzwang and has no defence
against the mate by the knight at b3
or d 3 .
Let us recapitul ate. By exchanging
A n ingenious position.
the b ishop for the knight White has
pushed the opponent's king in front In the next study, Black's queen
of the pawn and now he gets set to bears the main responsibility for the
give mate by Zugzwang. mate, although it spares no effort to
protect the king.
6 � fl h2 7 � g3 mate.

219
H. Rinck, 1 91 7 ]. Halumbirek, 1 93 0
455
454

Mate in four moves

White to play and win
1 ® f4 ! Black can opt for two lines
White starts out by chasing the op­ of play :
ponent's king.
(a) 1 . . dda6 2 b4 \l;a7 3 ® f8
1 ®c7 + dda8 2 ® aS+ ddb7 3
.

d:> a6 4 ® a8 mate ;
� cS + ddb8 4 cl:J"b6 + d/c8 S (b) 1 . . b4 2 ® f l ! bs 3 ® f6 b6 4
®b7 + ddd8 6 \l;d2 ! . . .
.

Up till now Black h as only been
making individual or forced moves,
all other responses would have
b rought an early loss. And suddenly
he is given freedom of movement.
But this freedom is deceptive. Black
has to keep square d7 covered and
the only answer he h as is 6 . . .
®e7. Of all the moves this is the
one that creates a confining con­
tact, which enables White to give
mate by 7 ® b 8 .
® a l mate.
Zugzwang is a very effective instru­
ment of attack and defence in the
end-game. I should like to show
you two more examples in which it
effects a certain co-ordinated con ­
nection o f pieces a n d p awns, which
puts the side using it at an advan­
tage.
0. Duras, 1 908
456

Here is a position i n which a l l of

White's operations are aimed at
prompting his opponent to make
disadvantageous moves . In the pro­
cess, confining contacts are estab­
lished by the pawns, which literally
provoke a final coup.

White to play and win

220
Will Black's king succeed in reach­ To pin the king i n the corner for
ing the corner a8 ? This would be his good, White transposes his bishop
only chance of achieving a draw.
1 � b4 ! dlf7 2 a4 dle6 3 aS dlds 4
a6 dlc6 S �aS !
I n this manner White creates the
strange mechanism by which he
cuts off Black's king from the
squ are a8. Black is in Zugzwang.
The cl-pawn's advance is checked
by White's king. As soon as this
h appens Black's king has to step
aside and White's pawn proceeds to drawn.
the queening square. An attempt by
the king to approach along the
eighth rank would also h ave been
frustrated : 2 . . . dle8 3 aS dld8 4
�d6 dieS S a6 and Black is again in
Zugzwang.
H. Rinck, 1 948
457
Draw
The unfavourable posting of
Black's pieces enables White to
drive the opponent's king into the
corner.
1 � b4 + dla7 2 � c6 + cba8 3
�e2 ! . . .
to a6. Now Black is in Zugzwang.
The queen has to dodge to a square
from which it can control a6.
3 . . ®c8 4 dlb6 ! . . .
.
Now Black has to reckon with S
� a6. There is only one line of play
for him that prevents the bishop
from moving to a6, n amely 4 . . .
®b7 +. But after S dieS Black is
again in Zugzwang and the game is
As regards sacrifices, I propose to
deal with them in a chapter devoted
exclusively to this subject in order
to analyse all its aspects.
Kinds of s acrifice
One can force the opponent's
pieces by threats, attacks, ex­
changes or Zugzwang situations to
support one's plan. But in the mid­
dle game such attempts can usually
be refuted by the enemy if he sees
through our plans early enough.
The situation is altogether different
in the case of sacrifices, which have
a strong element of force. The main
advantage of a sacrifice is that it oc­
curs suddenly and at least the oppo­
nent is usually caught by surprise.
This is due mainly to the fact that
we tend to think in terms of more or
less fixed values of the pieces and to
forget that such values are relative.
In the heat of the battle we often fail
to bear in mind that the values of
pieces are variable.
22 1
The importance of a certain chess­ 458

man in a limited conflict at a parti­

.
cular time is determined to a great
extent by the part it plays in the op­
eration to be carried out. No s acri­
fice in material is too great when we
mount an offensive against the
king's fortifications if we are con­
vinced that we will checkmate the
king or recover our material losses
with interest. The same holds true White to play
for any other tactical operation. If a
certain combination promises to Black's king is in a quandary, but its
give us an advantage, we stop at no defence still seems to be intact, for
sacrifice, provided our calculation if White should attack by 1 b3 + the
shows that we shall r�cover every­ king can retreat to a3. But White
thing and end up with an advan­ spots a weak point in Black's de­
tage. The problem is that we have to fence system. First he sacrifices his
give something first before we can rook : 1 rl a3 + ! The purpose of this
recover it, which means that such move is to deprive the hostile king
operations always entail a risk. If of its only remaining free square.
our calculations are wrong, it is un­ H aving forced his opponent to ac­
likely that we shall make good our cept the sacrifice by 1 . . ba, White
losses. This means that any opera­
.

gives mate by 2 b3. What White did

tion in.Jdving a sacrifice has to be
calculated with great precision and
sometimes far in advance.
Sacrifices can play a part in a wide
variety of plans, but their main p ur­
pose is to disorganize enemy de­
fences or to neutralize them alto­
gether. It happens occasionally that
our forces cooperate well, but to
achieve full success it is necessary to
weaken or eliminate enemy de­
fences. This is where the sacrifice
comes in h andy. The following ex­
amples illustrate how this is done.
was force the pawn to occupy
square a3, thereby m aking it inac­
cessible for the black king. Such
s acrifices are termed decoy sacri­
fices and the operation itself is
called decoy. In the case in hand,
White used a simple mating combi­
nation based on a decoy sacrifice.
Here is another example of a decoy
s acrifice.

222
H. Otten, 18 92
Conclusion ofa study
459
White to play
White's pawn is heading for the
queening square, but Black plans to
intercept it at a6 by 1. . �e3.
.
Should White decide to go 1 cl)e4,
Black can play 1 . . . �f8. White
unexpectedly goes 1 g5 + ! In this
manner he manages to confuse
Black's defence. After 1 . . cl):g5
.
the bishop's diagonal is blocked and
the pawn can advance unmolested.
If Black chooses to take the pawn
with his bishop by 1 . � x g5,
. .
White answers with 2 00e4 ! �h4 3
cl)f3 , and again the bishop is unable
to stop the pawn.
Sometimes the decoy sacrifice oc­
curs only as a threat.
F Amelung
Conclusion ofa study
460
Black to play
What line should Black take ? Odds
seem to be stacked heavily against
his building up a defensive system.
But his situation is not as hopeless
as it looks. He launches an effective
attack by 1 � e7 + ! With this
. . .
move Black threatens to make two
double attacks : If 2 cl) x e7 then
2 . . . � c6 + , but if 2 ® X e7, then
2 � g6 + . This means that
. . .
White's king h as to retreat to e 8 ,
b u t then the bishop returns and
threatens to build up an impreg­
nable fortress by·3 . . . � g7. After 2
00e8 � f6 3 00f8 � e7 + ! the game
ends in a draw through repetition.
461
White to play
223
First of all White goads the rook to
square g8 by 1 � h6 (mating threat)
and then by s acrificing his queen by
playing 2 ® x h7 + ! he demolishes
his opponent's fortifications. The
game ends with mate after 2 . . .
dl x h7 3 l:l h2. This was an exam­
ple of a destructive sacrifice. Its pur­
pose was to eliminate the protection
of the enemy king-in this case the
pawn at h7.
462
Mate in three moves
1 � f8 + � h S 2 rJf:J X hS + ! gh 3
l:l h6 mate.
This is an example of a decoy sacri­
fice. By sacrificing the queen at h S ,
White has diverted the pawn g 6
from securing t h e square h 6 .
C. Cozio, 1 766
463
White to play
224
This antique position shows how
White manages to break through
and get a pawn queened. He does
this by making two decoy sacrifices.
The first sacrifice, 1 b6 !, serves to
steer either the a-pawn or the c­
pawn to the adjacent file. If Black
plays 1 . . . ab , White sacrifices his
second pawn by 2 c6, thereby di­
verting the b-pawn from his task of
controlling the square a6. After
2 . . . be the a-pawn gets the green
light to proceed to the queening
square.
These examples illustrate that the
defence can be eliminated or
thrown into disarray by two kinds
of sacrifice, i. e. decoy sacrifices and
disorganizing sacrifices. But sacri­
fices can be not only destructive but
also constructive and they can pro­
mote the co-ordination of forces,
which is such an essential factor.
464
White to play
Here White forces mate in two
moves : 1 l:l e8 + dl x e8 2 ®e7
m ate. The decoy s acrifice enables
White to co-ordinate the operations
of his pieces (knight and queen)
against the king and to establish a
m ating contact.
L. Kubbel, 1 92 5 The first move is very difficu lt and
Conclusion ofa study quite surprising :

465
1 00d7 ! . . .
With this move White prepares to
make the necessary connection of
pieces.
1 . . . dle4 2 El d S ! . . .
This decoy puts the dot on the i.
2 . . ell x dS 3 ® d4 mate.
.

Occasionally one's own pieces im­

White to play
pede the co-ordination of forces.
One can dispense with such pieces
of course by sacrificing them.
In this position White uses the post­ Here is an elementary case.
ing of the black pieces to his advan­
tage and forces mate by a decoy Reisch- Troyanski
sacrifice. Sofia 1 969
1 Y.!t1 a3 + ! <l/ x a3 (otherwise Black 467

lo�es his queen) 2 � c2 mate.

The use of the decoy sacrifice for
co-ordinating one's forces also oc­
curs quite frequently in problems.
This idea is very lucidly expressed
in the following old miniature .

F. Healey, 1 902

466
Black t o play

If it were not for the knight at eS,

Black could force mate by moving
the bishop to eS. He has no qualms
about parting with his knight of
course :
1 . . . � f3 + ! 2 t!t1 x f3 � eS+ , and
White resigned.
In this case the sole purpose of the
s acrifice was to vacate a square. I
Mate in three moves should point out, however, that this

225
square played a very important squ ares, ranks , files or diagonals
part, because it was from here that are known as vacating sacrifices.
the bishop attacked the king,
There is yet another ki nd of sacri­
thereby taking part in a co-ordi­
fice shown in the following exam­
nated connection of rook, pawn
ple.
and queen.
We have thus got to know a further
kind of sacrifice, namely the vacat­ Marshall-MacClure
ing sacrifice. Vacating sacrifices can New York 1923
also be used to open a blocked rank,
file or diagonal to enable a piece to 469

carry out a certain task.

The next example illustrates such a
situation . '

Manov-Khairabedyan
Bulgarien 1 962

468

White to play

Here, White saves himself surpri­

Black to play
If we remove the bishop at hS and
the two Black rooks, the second
player can mate White in two
moves by giving check with his
queen at h8 . 1t is not difficult to get
rid of the pieces mentioned.
1 . . � e2 ! 2 ® x e2 El h8 + 3
. gory of self-destructive sacrifices . It
d:>gl E1 h 1 + ! 4 d:> x h l E1 h 8 + 5
d:>gl E! h l + ! 6 d;J X hl '!'tfth8 + 7
d:>gl ®h2 mate.
Such sacrifices designed to vacate
singly by destroying his own rook
and pawn.
1 El h6 ! !j X h6 2 h8 (@) + El X h8
3 bS !
And Black is unable to avert the
stalemate set up by his opponent.
The sacrifice of a rook and a pawn
witnessed in the foregoing can be
called a self-d estructive sacrifice.
This kind of sacrifice frequently oc­
curs when a player tries to get a
stalemate. The so-called "rabid"
piece certainly belongs to the cate­
is characterized by repeated acts of
self-sacrifice .
And he re is a somewhat more com­
plex case of self-destruction.

226
Ormos-Bdt6czky
Budapest 1 9 5 1
470
Black t o play
The second player's position looks
critical-his king cannot move.
Black takes advantage of this predi­
cament to save himself through
stale mate by a series of self-destruc­
tive sacrifices.
1 . . El b l + 2 d:lh2 El h l + ! (the
.
rook commits suicide) 3 d:l x h l
� g3 + ! (so does the knight; it has
to b� taken, otherwise the queen is
lost) 4 fg ® x g2 + ! (the last s acri­
fice puts everything right) 5 d:l x g2,
and Black is stalemated !
It is time to draw some conclusions
now. We have seen that the sacrifice
can be resorted to for disorganizing
or eliminating the defence just as ef­
fectively as for co-ordinating for­
ces. There are different kinds of
sacrifice : we have examined five
main categories of sacrifice occur­
ring in combination. Let us list them
again briefly :
l. The decoy sacrifice in which an
opponent's piece (or p awn) i s
steered to a certain square ; 2 . The
decoy sacrifice in which an oppo­
nent's piece (or pawn) is diverted
from fulfilling a certain defence
task; 3. The destructive sacrifice, in
which the opponent's fortifications
are smashed by pieces or pawns ; 4 .
The vacating sacrifice, in which
squares , diagonals, ranks or files
are vacated, which are of crucial
importance for operations by pieces
of the sacrificing player; 5. The self­
destructive sacrifice, in which the
sacrificing pl ayer lets his unwanted
pieces be taken or in which he has
no choice but let them be taken.
All these kinds of sacrifice are en­
countered in practical chess . But
sometimes it happens that a single
sacrifice can have several or even all
of these functions at once.
471
White to play
White gives m ate in two moves by
sacrificing his queen : 1 ® c6 + de 2
� a6 mate. First this sacrifice steers
the pawn (or the bishop at e4) to
square c6 after which it becomes in­
accessible to the king, and secondly
it vacates square a6 for the mating
move with the knight. It is a combi­
nation of decoy sacrifice and vacat­
ing sacrifice.
227
P. Romanovski, 1 950
472
White to play
Here too White forces mate in two
moves :
1 ®f8 + ! ! lJ x f8 2 � e7 mate or
1 . . . dl X f8 2 rlh8 mate .
By sacrificing the queen White
either steers the Black rook to
squ are f8 , thereby diverting it from
covering the square e7, or he lures
the king to f8 as a result of which it
relinquishes control of square h8.
This is an example of an intimate
combination of the two kinds of de­
coy sacrifice.
Alekhine- Yates
London 1 922
473
White to play
228
The first player conjures up a com­
bination based on a sacrifice.
1 � d7 d/h8 2 � f6 ! rlgf8 3
lJ X g7 ! . . .
White smashes the opponent's de­
fence and at the same time lures the
black rook to f6 .
3 . . . bl x f6 4 dieS ! Black resigns,
because 4 . . . bl ff8 or 4 . . . bl af8
is followed by mate in two moves.
The move 3 n X g7 ! is a combination
of decoy and destructive sacrifice.
Gogolev- Varshavski
Aluksne 1 96 7
474
Black to play
Black's prospects seem gloomy, but
he manages to achieve a draw by
stalemate.
1 . . bl d3 + ! 2 ® x d3 (White has
.
to accept the s acrifice because if he
plays 2 d/g4 ® d 1 + he actually
loses) 2 . . . ®e3 + ! 3 ® x e3 stale­
m ate .
How can these two sacrifices be
classified? The first one is obviously
a self-destructive sacrifice, at the
same time it is designed to keep thl'
queen in the vicinity of square e 3
for the sake o f the second sacrifice,
i. e. it is also a decoy sacrifice. The
second sacrifice is particularly fasci­
nating. The queen sacrifice at e3
not only destroys itself, but also
steers the opponent's queen onto
this square from which the pawn gS
is pinned. All conditions have now
been met for stalemate.
This example illustrated a combina­
tion of decoy and self-destructive
sacrifice .
Involved combinations may involve
all in one s acrifice.
Kortchmar-Polyak
Kiev 1 93 7
475
White to play
The white pieces converge omin­
ously on the enemy king, whose life
hangs on a thread. The following
combination proves the truth of the
saying "A chain is only as strong as
its weakest link".
1 � b4 ! ab 2 'i!!:J X d6 ! 'i!!:Jd 7 3 'i!!:Jd S ! !
dJf8 4 i:l X g7 ! ® X d S 5 r! g8 + !
dJ x g8 6 .El e8 + with mate on next
move.
Let us work our way through this
maze. By 1 � b4 White eliminated
his knight. This is a vacating sacri­
fice which opens the cl-file for the
White queen. Move 2 ® x d6 is a
destructive sacrifice. The knight at
d6 controls the break-through
square e8 and has to be eliminated
for this reason. The queen cannot
be taken because of the mate in
three moves after 2 . . rl x d6 3
.
ll e8 + .EX fs 4 .EX x g7 + dJhs 5
rl x f8 mate. No doubt the most
beautiful move in the combination
is 3 ® d 5 ! The queen is again u nas­
sailable, but White threatens 4
.El x g7 +. If Black had continued
with 3 . . . g6, White would have
countered by 4 .El ge3 ! with an un­
avoidable penetration to e8. This is
why Black was forced to retreat
with his king to f8 . The thrust 4
.El X g7 ! is at once a destructive sac­
rifice and a diversion sacrifice, be­
cause if 4 . . .El x g7, then Black's
.
rook is pinned and White can retal­
iate by 5 ® x d7. And finally, 5
.EX g8 + is an example of a decoy
sacrifice : By l uring the king to
square g 8 , White diverts it from its
task of covering square e8.
We can thus count a total of four
different kinds of sacrifice in this
combination: a vacating sacrifice, a
destructive sacrifice, a diversion
sacrifice and a decoy sacrifice. Our
examination of sacrifices covered
m ainly those that paid out in full at
the end of the combination. Some
authors refer to such sacrifices as
dummy sacrifices.
But there are also other kinds of
s acrifice. There are those that are
229
made to wrest the initiative from the
opponent, those designed to speed
up one's development, sacrifices
made to impede mobilization of
hostile forces and others.
They are normally referred to as
"true" sacrifices, because they do
not pay straight away or at least not
very soon. The only thing they offer
are certain advantages that may
eventually, but not necessarily, lead
to a perceptible advantage . True
sacrifices are usually motivated by
strategical considerations and con­
stitute an integral part of the long­
term plan of the game, although
they can also give rise to tactical in­
volvements and combinations. True
sacrifices sometimes occur in the in­
itial stages of the game, e. g. in
King's Gambit, Evans Gambit, Si­
cilian Gambit, Muzio Gambit, Ale­
khine-Chatard Gambit and i n many
other sharp openings and systems.
Let us conclude this chapter with a
true sacrifice, which occurred in the
final stages of a qu iet and balanced
openmg.
Smeikal-Ljubojevic
Moscow 1 977
476
White to play
230
This position evolved from the
following opening moves :
1 � f3 � f6 2 c4 cS 3 � c3 e6 4 g3 b6
5 � g2 �b7 6 0-0 a6 7 b3 d6 8
� b 2 � bd7 9 ld e 1 E! b8 1 0 e4 bS 1 1
d 3 b4.
Having noticed that by taking the
initiative on the Queen's side Black
was far behind in development,
White boldly played 1 2 � ds !
The purpose of the sacrifice is to
open the e-file, although for the
present it does not herald a combi­
nation. But after 1 2 . . . ed 1 3 ed +
�e7 1 4 '®e2 it is clear that Black
will have a hard time to solve his de­
velopment problems. His pieces are
very constricted by contrast with
White, who has a manifest advan­
tage in time and fu ll freedom of
movement.
The study of true sacrifices is a sub­
ject all to itself. You might be inter­
ested to know that Ljuboj evic was
overwhelmed by the mounting diffi­
culties of defence and lost.
Classification
of combinations
The combination is one of the most
fascinating, but at the same time
most difficult elements of the game.
Like any other complex manifesta­
tion, it can be characterized by a
number of different features. One
could try, for instance, to classify
combinations according to the com­
parative strength at the beginning
of a combination, but such a char-
acterization w o u ld be too formal
and divorced from the true e"s sence
of the combination.
A classification based on those mo­
tives and circumstances that deter­
mine the genesis of a combination
in its incipient stages seems very
tempting, b u t it would be very diffi­
cult to implement such a classifica­
tion. Such thi ngs are usually more
complicated than they look on the
surface . In fact they often lie so
deep that eve n a well-trained eye is
unable to perceive them.
In classifying combinations one
usually starts o u t from the final po­
sition of a comb ination that has ful­
filled its purpose. If we classify com­
binations on the basis of the obj ec­
tive achieved , it is easy to break
them down into two main groups :
comb inations in which one side
wins and combinations in which
one side manages to achieve a draw.
They are commonly referred to as
winning com b inations and drawing
combinations, respectively. Win­
ning combinations are normally
broken down further into two sub­
groups, i. e. winning combinations
ending in mate and winning combi­
nations in which at the end of a
combination o n e side has achieved
a material superiority that is suffi­
cient for a win.
According to their ending, drawing
combinations can readily be sub-di­
vided into the following four sub­
groups : combinations ending in
stalemate ; positions characterized
by endless chase or perpetual
check; positi on s in which the mate­
rial balance is s u ch that the game is
certain to end in a draw and lastly
foqified positions or blockades.
This is a rough sketch of the plan
used at present for classifying com­
binations. Most authors of books
dealing with the middle game try to
make further sub-divisions of the
various groups. A distinction is of­
ten made for instance between com­
binations ending in mate on the last
rank and those in which mate oc­
curs on a file or diagonal. But there
is no general agreement on this
qu estion and we shall not go into it.
I believe that the classification of
combinations according to the ob­
jective into winning and drawing
combinations is ful ly justifiable.
First such a classification gives an
exact description of the tasks set
and secondly, it enables studies to
be automatically included in this
classification, which are differen_­
tiated on the basis of this character­
istic. We should not forget of
course that combinations designed
to win material or to bring about
material balance with a view to
achieving a draw can be essentially
very similar, although they belong
to different groups.
As far as the further breakdown of
combinations is concerned, I be­
l ieve it would be expedient to conti­
nue using the same approach. But I
suggest the classification be based
on the objects of attack and the
contacts that are co-ordinated
against them in the final phases of
combinations. We already know
that co- ordinated connection of
contacts is the essence of combina­
tions.
23 1
In terms of objects of attack, win­
ning combinations can be classified
into three groups . The first contains
combinations aimed at the king, the
second comprises those combina­
tions whose purpose is to capture a
piece, whilst the third involves those
connected with queening.
In view of the special role of His
Royal Majesty in the chess game (if
the king is lost, all is lost) I believe it
is appropriate to put all combina­
tions that are connected with an at­
tack on the king-and not only mat­
ing combinations-in a separate
group. We shall call them combina­
tions aimed at the king. This group
can be broken down into three sub­
groups according to the connection
of contacts occurring in the final
stage. The first sub-group com­
prises combinations whose final
stage features a mating connection
of contacts. It consists not only of
all mating combinations which end
i n mate but also those in which m ate
is prevented at the price of losses in
material. The characteristic feature
of the final stages of such combina­
tions is an active contact aimed at
the king.
The second sub-group comprises
mating combinations consisting of a
double attack and consequently of
two active contacts. These combi­
nations are characterized by the
fact that they consist of a m ating
threat coupled with a stron g threat
or attack. This kind of combination
does not necessarily lead up to a
m ate ending, but in cou ntering it
the other side usually sustains losses
in material.
232
The third sub-group covers combi­
nations involving double attacks,
one of whose objects is the king.
Mating threats do not occur in this
sub-group, there are only attacks
on, or threats to, the king. Why did
we find it necessary to put combina­
tions with double attacks, in which
one of the objects of attack is the
king, in a sub-group of their own ?
The decisive consideration was that
the king is forced either to step
aside or to protect itself against the
attack as soon as it is attacked. By
contrast with an attack on any other
piece, an attack on the king may
never be disregarded, because it
may not be left exposed to direct at­
tack.
Combinations whose purpose is to
win pieces can be broken down into
two sub-groups according to the
kind of contact. The first consists of
combinations which end in the im­
mobilization and capture of the
piece against which a combined at­
tack is launched. The reader will re­
call that in such situations there is
an attacking contact and confining
contacts . In other words only one
active contact occurs.
The second sub-group comprises
combinations involving the double
attack. They are characterized by
an attack on a piece coupled with a
strong threat or an attack on an­
other chessman. To qualify for this
sub-group a combination has to
have two active contacts in the final
position.
Queening combinations can also be
broken down into two sub-groups
according to the contacts that occur
in them. The first sub-group com­
prises combinations in which the
queening threat in the form of an
active contact between the pawn
and its queening square results
either in queening or in a decisive
gain in material. The other sub­
group covers combinations involv-
Combinations
against the king
Winning
combinations
Combinations
against pieces
and pawns
Queening
combinations
It is expedient to classify drawing
combil}ations according to the same
prir nple on the basis of the objects
of attack and of the contacts com­
ing into being towards the end of
the combination.
The first group again comprises
ing double attacks in whose final
phase the queening threat is cou­
pled with other strong threats or at­
tacks on pieces. Two active contacts
are involved here again.
The following diagram shows all
the different kinds of combination
considered by us.
Mating combinations
(one active contact aimed
at the king)
Double attack
(two active contacts
including a mating threat)
Double attack
(two active contacts
including an attack
on the king
Combined attack
(one active contact)
-{ Double attack
(two active contacts)
Simple queening
(one active contact)
-{
Double attack
(two active contacts
including a queening
threat)
combinations involving an attack
on the king towards the end of
which there are only confining con­
tacts and no active attacking con­
tacts. These are stalemate combina­
tions. Their main feature is that al­
though the king is in a quandary, it
233
is able to save itself by a stalemate
combination.
The second group is made up of
combinations designed to achieve
perpetual check. Such combinations
have an attacking contact, but the
confining contacts are not sufficient
for giving mate. At the same time
the freedom of movement of the de­
fender's king has been curtailed to
such an extent that it cannot shake
off its assailants no matter where it .
turns.
Combinations of the third group do
not differ basically from those of
the second group. We call them
combinations designed to achieve
perpetual chase. The only differ­
ence is that in the third group it is
not the king but the other pieces
that are subjected to endless perse­
cution.
Combinations involving blockades
and the building of fortifications
form two separate groups. In com­
binations with a blockade, one
piece is usually caged in to prevent
it from taking part in the struggle,
which, as a result of the small num-
ber of pieces participating in the
struggle, leads to a drawn game.
In the second group, the player
forced to go on the defensive suc­
ceeds in fortifying his position in
such a way that the opponent is un­
able to penetrate it. Or the attacker
finds himself unable to co-ordinate
his forces against the obj ects of at­
tack. It is also conceivable that for
some reason the objects of attack
are inaccessible to enemy attack.
The last group comprises combina­
tions leading to a balance of forces
that ensures a draw. In such situa­
tions usually the player with weaker
material resources restores material
equilibrium.
These comb inations are not essen­
tially different from the winning
combinations d iscussed earlier. Ma­
terial balance can be restored by an
attack on the ki�g, an attack on
pieces or pawns or by queening.
The reader will realize of course
that each of the aforementioned .
combination groups can be broken
down into two sub-groups.

Stalemate combinations

Perpetual check combinations

Drawing Perpetual pursuit combinations

combinations
Blockade combinations

Fortification combinations

Combinations aimed at restoring mate­

rial equilibrium and ending in a draw

234
The first of these would comprise 1. Combin atio ns
combinations in their purest form,
ag ains t the king
the second covers double attacks
coupled with some other manoeu­
vre. We shall deal with this question Mating combinations account for
in greater detail when we examine the bulk of this group. Their main
some relevant examples. characteristic is the realization of a
All cases of drawing combinations mating combination.
are covered in the diagram shown ·
Let us take a closer look at the two
above. kinds of such combinations. The
In actual practice we may some ac­ first kind inevitably ends in mate.
ross some convoluted combinations
taking place in several phases and Radulov-Soderborg
overlapping each other. Such intri­ Helsinki 1 961
cate combinations are always classi­ 477
fied on the basis of the last, decisive
phase. For instance a combination
initiated with a view to win material
can transform into a queening com­
bination.
The determining feature of such a
combination is the queening ma­
noeuvre, the gaining of a piece
being the introduction.

White t o play
Winning combinations
Four white pieces are ominously
poised to strike at the opponent's
You will recall that winning combi­
K-side. But how is he to get at the
nations can be classified into three king, how does he eliminate the
groups based on the object of at­ pawns protecting it? He does it by
tack. The first comprises combina­ sacrificing the rook. This is not only
tions aimed at the king ; the second a self-destructive but also a diver­
those aimed at pieces and pawns sive sacrifice.
and the third queening combina­
tions. Let us now examine combina­ 1 El X g7 ! 00 X g7 2 ® g4 + 00h8 3
tions belonging to each of these sep­ ® hs
arate groups. Note the important detail that
White was able to play the queen to
g4 and hS only because Black's king
had been drawn into the sphere of
influence of the bishop at eS. The

235
mating mechanism-by this we
mean the queen and the bishop sup­
porting it in its attack on square
h7 -has been set in motion. The
bishop at e5 is also indirectly taking
part by preventing the knight at f6
from taking any counter-measures.
Mate is inevitable, because 3 . . .
cbg8 is followed by 4 � x f6 tl e 8 5
'® x h7 + cbf8 6 ®h8 mate.
In the next example there is also a
mating combination of contacts,
but what makes it different from the
preceding position is that mate is
thwarted at the cost of heavy losses
in material.
Kopayev-Averbakh
Leningrad 1 946
478
Black to play
The king was forced to come out
from behind his pawn defences,
while its pieces are on the other side
of the board and give it no protec­
tion. No wonder that the king im­
mediately finds itself under heavy
fire.
1 . . . h4 2 ® c3 + 00d7 3 ®es . . .
White's queen double-times to her
lord's rescue, but . . .
236
3 . . . hg + 4 00 X g3 lj h3 + !
Here too the rook sacrifices itself
for a mating manoeuvre. This sacri­
fice steers the king from the safety
of square f3 and to the death-bring­
ing square h3. Black's queen imme­
d iately takes advantage of this.
5 et> x h3 ® f3 + 6 cbh4 � e7 +
The mating combination betwen
the queen and the bishop is estab­
lished, the queen playing a passive
b ut very important role by creating
a confining contact. The bishop is
g iven the honour of delivering the
decisive blow. White has no choice
but to play 7 ®g5. But after 7 . . .
'® X f4 + 8 00h3 ® X g5 the game
ts over.
Mating combinations can also be
coupled with double attacks when
the mating threat is accompar.ied by
some other strong threat or attack.
Such a combination obviously does
not have to end in mate, but it can
win material.
Here is a typical example.
Parr- Wheatcrofi
La:-:d!!'!'! ! 938
479
White to pl ay
 By 1 1:1 h5 ! White launches an effec­
tive double attack. Black's queen is
under fire, and 1 . . ® X d7 is fol­
. de Black cannot play 5 . . . 0 x e4
lowed by mate in two : 2 f) g5 +
dlh8 3 0 x h6 mate.
Such combinations accompanied by
a double attack are not usually the
result of a consistent king's attack,
in most cases they occur unexpec­
tedly.
' fhe following position is quite
characteristic.
Soloviev-Averbakh
Moscow 1 945
480
Black t o play
The deployment of Black's pieces is
more active, but at first sight there
are no signs that a quick ending is in
sight. But the main weakness in
White's battle order is the posting
of his queen. The second player ex­
ploits this.
1 . . f) e2 + 2 \t>ft f) f4 3 ® h7 . . .
.
White's queen is trapped. It just
has a single squ are to move to.
Black can apparently triumph im­
mediately by 3 . . . ® e6 , mate in
three moves being threatened. But
White can refute this threat by
bringing his queen into play by 4
®e4, because after 4 . . . ® x e4 5
because of 6 0 d8 mate. But Black
has a more effective reply.
3 ® f7 !
0 0 .
With this move he threatens first to
en cage the queen by moving his
knight to g6 and then to capture it
by moving his rook to h 8 . To coun­
ter this threat White is forced to
take decisive steps.
4 .El c l f) g6 5 �g3 c6
To 5 . . . .El h8 White would have
answered with 6 0 x c7 + 'l!!:t x c7 7
® x g6.
6 .El c5 . . .
The last attempt. If Black went
6 .. EX h 8 , White's reply would
.
have been 7 !l f5. Instead, he
launches a double attack prepared
well in advance, which puts a quick
end to the struggle.
6 . . . � c3 !
Attacking the queen and threaten­
ing mate at e 1 . White resigned.
Combinations with double attack,
in which the king is one object of at­
tack, without an actual mating
threat are conceivable.
237
Botvinnik-Menchik T h e second kind consists of combi­
Hastings 1 93 5 nations with double attacks in
481
which various pieces are subjected
to simultaneous attacks and threats,
i. e. they usually involve two active
contacts. The next diagram shows a
typical example of a combined at­
tack.

Botvinn ik-Stepanov
Leningrad 1 934
482

Black to play

Black's position is difficult. In an at­

tempt to exchange her bishop
against White's bishop, whose line
of fire c overs the squares e6 and f7 ,
Black played 1 . . � d 5 . But then
.

came 2 fe � x b 3 3 e7 + ! A brief
but extremely effective combina­
tion ! The p awn sacrifice entangles White to p lay
Black's king in a double attack. Af­
ter 3 . . 00 x e7 4 � c6 + White's
. The black queen's freedom of
materia l superiority is sufficient for movement is perceptibly curtailed,
a win. Black resigned. but for the time being it does not
seem to be in danger. White does
not bother to cover his pawn at d4
2. Combin atio ns and goes 1 f5 ! If Black played 1 . . .

agai nstpi eces andp awns Et x d4, White would have

answered with a double attack 2 fe
We differentiate between two kinds � x e6 3 � x e6 fe 4 � g6 Et e8 5
of combination in this category. e5 � fd5 6 ®f2 attacking the rook
The first comprises combinations in and threatening a breakthrough at
which a combined attack results i n f7. That is why Black-apparently
the capture of a piece, i . e. the piece thinking that he could take the
is attacked and at the same time its p awn at leisure-plays 1 . . ef. But
.

mobility i s restricted so that it can­ this move is a mistake, for it allows

not evade the attack. In other words White to work up a combination
there is one active attacking con­ against the precariously placed
tact, while the rest are confining queen : 2 � x f7 + ! Et x f7 3 � c4,
contacts. and the queen is captured.

238
And here is an example of an origi­
nal combination, which peaks in a
piece capture.
Novotelnov-Rown er
Moscow 1 946
483
Black to play
White has attacking chances,
threatening to take the p awn at b6
and to advance his pawn to f6 . But it
is Black's move and he detects a ma­
jor weak spot in his opponent's po­
sition : the inadequately protected
second rank. He sacrifices a piece in
order to steer the rook to square fS .
1 . . � X fS ! 2 �
. X fS � X fS 3
r:! x fs r:! ed 8
The queen has t o step aside t o a
square from which it can watch the
first rank.
4 ® c4 r:! ac8 5 ®e2 b:l x c2
This swap is just as effective as the
sacrifice made earlier, for the queen
is lured to the square c2.
6 ® x c2 ® c8 !
The ensuing position is quite un­
ique. White is powerless against this manages to capture his opponent's
double attack finale in which
Black's queen attacks th e enemy
queen and rook. Unabl e to take the
queen or cover the roo k, he resigns .
3. Qu eening combina tio ns
Material advantage can usually be
obtained by advancing a pawn to
the queening rank. An elementary
combination illustratin g this point is
the pawn breakthrough. H ere is an
instructive example of su ch a com­
bination that occurred i n a n end ­
game.
Averbakh-Bebtchuk
Moscow 1 964
484
White to play
Before White can advance one of
his pawns to the queening rank,
White h as to clear the way . In the
position shown here this i s done by
sacrificing three pawns.
1 eS ! fe 2 gS hg
Black does not fare bette r by 2
cbd7 either. The king stop s the
white pawn's advance after 3 £6
cbe6 4 fg cidf7 5 gh bS, bu t White
pawns by 6 d:le4 b4 7 <1>d3 !
239
After 3 f6 gf 4 hS the sole White sur­ double attack consisting of a two­
vivor's advance cannot be stopped. fold attack (on the rook and the
king) and a concealed queening
There are of course also queening
threat.
combinations involving double at­
tacks. In the next position White decides
the game in his favour in no time by
Weltmander-Polugayevsky unleashing a double attack.
Sochi 1 958
Dufresne-Harrwitz
485
486

B lack t o play White to play

The first i mpression is that White He played 1 ®e8 ! If 1 . . . .!d x e8,

has the better of the game, since then 2 f7 + with a double attack
Black's knight e2 is lost. But the consisting of an attack on the king
outwardly harmless, but actually and at the same time an attack on
strong passed pawn at c3 enables the rook coupled with queening.
Black to reali�e a winning combina­ Black preferred to play 1 . . . d/h7,
tion in which the knight is sacri­ but after 2 f7 he res igned .
ficed.
1 . . . � g3 + !
Thanks to this sacrifice the f-file is Drawing combinations
opened for a decisive double attack.
It is advisable to take a closer look
2 fg ® f6 + 3 ® f2 . . . particularly at those drawing com­
The only defence, otherwise White binations in which the king is the
loses the rook. object of attack. Perpetual check
positions are typical examples of
3 . . . }d X e1 + ! 4 d/ X e1 f!t:J X f2 + this kind of combination. Stalemate
5 dl x f2 c2, and the pawn queens. also belongs in this category, al­
It is worth taking another look at though as a rule we stalemate our
this operation. 2 . . . ® f6 + is a own king and not the opponent's.

240
1. Pe rpetu al check
Perpetual check occurs most fre­
quently when the offensive forces
are not sufficient for a winning at­
tack and when they are unable to
disorganize the position around the
enemy king and pursue it. In such
situations it is usually the queen that
is used to give perpetual check.
This is illustrated by an old exam­
ple, which has become classic.
Alekhine-Lasker
Moscow 1 91 4
Black to play
The pos ition shown in the diagram
occurred after the following open­
ing moves :
1e4 e5 2 � f3 � c6 3 d 4 ed 4
� x d4 � f6 5 � c3 �b4 6 � x c6
be 7 � d3 d5 8 ed cd 9 0-0 0-0 1 0
� g S � e6 1 1 ®f3 �e7 1 2 b:I fe 1 . . .
In an attempt to resolve the problem
created by the posting of his oppo­
nent's black-square bishop, Lasker
played 1 2 . . . h6. Alekhine took ad­
vantage of this in order to develop a
drawing combination by making
two destructive sacrifices.
13 � x h6 ! gh 1 4 b:I x e6 ! fe 1 5
®g3 + 00h8 1 6 ®g6 . . .
White's queen supported by the
bishop is hovering ominously close
to the hostile king, but Black's
knight reliably defends the square
h7. White does not have sufficient
reserves to make mating thr�ats. Af­
ter 1 6 . . . ®e8 he can do no more
than give perpetual check :
1 7 ® X h6 + 00g8 1 8 ®g5 + 00h8
(by no means 1 8 . . . 00f7 on ac­
count of 1 9 ®g6 mate) 19 ®h6 + ,
and the game could be agreed
drawn.
Sometimes perpetual check is the
last resort; it happens when the at­
tacking player's resources are ex­
hausted before a breakthrough can
be made.
Smyslov- Vasyukov
Moscow 1 961
488
White t o play
The first impression suggests that
White's attack has broken down
and material losses are unavoidable.
But White played 1 b:I h5 + ! This
destructive sacrifice lays bare the
king and enables White to give per-
241
petual check by 1 . . . gh 2 ® d6 +
®g6 3 ® f8 + .
Perpetual check threats involving
double attacks can also be coupled
with other strong threats.
W: von Holzhausen, 1 899
489
White to play
White's position looks hopeless.
How can h e stop the terrible ene my
pawns ? And yet there is a way : 1
� b 8 ! This is a double attack, be­
cause it contains two concealed
threats. The first is 2 � c6 with a
mating threat on b 8 . That is why
1 . . a 1 (®) is worthless, because
.
this would even give White a win by
2 � c6 '1!!1 a7 + 3 � x a7 \!/ x a7 4
lJ g6. Consequently the king has to
move, but after 1 . . cba7 2 rl b4 !
.
the second concealed threat comes
into play : White has set up the me­
chanism of perpetual check. After
2 . . . a t (® ) 3 � c6 + d/a6 4
� b 8 + dlaS 5 � c6 + the game
ends i � a draw.
242
2. The s talem ate
Stalemating combinations occur
rarely in actual games, especially in
the middle game. They occur most
frequently when the player out to
win the game underestimates his
opponent's defence resources or
fails to make a realistic assessment.
Here is one of the best- known ex­
amples of this kind.
Evans-Reshevsky
New York 1 963
490
White to play
It is e asy to tell that White's posi­
tion is miserable. He is one piece
down and his king is in mortal
danger. Apparently it is high time to
acknowledge defeat. But Evans did
not think of quitting. Instead he
m ade a move which Black must
have interpreted as a move of des­
peration.
1 h4 ! rl e2 + 2 cbh t ® x g3 ??
Expecting capitulation any minute,
Reshevsky did not take the trouble
to see what his partner was up to
and took the pawn. But this left
White's king withou t a move, so
 1 hat all he has to do is free himself no means the last straw that the
of his superfluous pieces, which he drowning man clings to. The stale­
does by making two self-destructive m ating threat is a perfectly legiti­
.\ :-tcrifices. m ate means of defence which
should never be discounted .
.) ® g8 + ! dl x g8 4 � x g7 + , and
regardless of who takes on g7,
Taimanov-Geller
White is stalemated .
Moscow 1 95 1
Beware of stalemate threats in end­ 492
games, especially when there are
only a few pieces left and the oppo­
nent's king is perceptibly hemmed
1 11 .

%agoryansky- Tolush
Moscow 1 94 5
491

Black t o play

White's position is hopeless, for

there is no defence against the nu­
merous threats. And the game lasts
only another three moves : 1 . . .
� x h2 2 ® e3 � a8 3 � g7 + (a
desperate sacrifice, which Black
B l ack t o play simply disregards) 3 . . . \t>h6, and
White resigned. It was discovered
White's prospects are bleak, but l ater that this was what might bt;
Black still has to overcome the last called a "comedy of errors". The
few obstacles. Instead of fortifying very first move could have given
his position systematically,' Tolush Black a win. He should have p layed
engineered a trap b ased on the 1 . . . � g4 ! , e . g. 2 l:l h3 � c l + 3
double attack. He played 1 . .
. dle2 � g2 + .
ddh3 ? His opponent seemingly But the move made by Black in the
obliged him by 2 'f!1l X f4 ! g2 + 3 actual game gave White a chance to
ddf2 � f6 . Black's idea was to coun­ realize an elegant stalemate combi­
ter 4 ® :f6 by the decisive 4 . . . n ation . It begins with an exchange :
g e4 + . But the game was drawn af­ 2 � x g8 ! et> x g8 3 � g3 + ! ! This
ter the unexpected 4 dlg l ! ! decoy sacrifice deprives the White
As you see, stalemate is indeed king of all moves. After 3 . . .
sometimes the last resort. But it is by ® x g3 the king is in a stalemated

243
position, which means that all
White has to do now is get rid of his
queen by 4 ® b 8 + 00g7 5 ®g8 + !
to get a draw.
Stalemate has always fascinated
problem composers. They have
created numerous outstanding
works illustrating how stalemating But now White's king, which has
connections come into being in the locked up its bishop, has no move
course of a contest.
V. and M. Platov, 1 90 7
493
Draw
It is impossible to intercept Black's
pawn, which makes White's checks
looks like gestures of resignation.
1 h7 + d/h8 2 � g7 + d/ X h7 3
�a1 + ! . . .
What is the point of this ? The pur­
pose of this move is unclear at first.
3 . . . dlg6 4 Jd x c6 + d/hs
There are no more checks in sight
and White's position suggests that
he is on the verge of resigning. But
his defence resources are far from
depleted. By posting his bishop on
a 1 , White has prepared a stalemate
nook for his king well in advance .
244
5 dlb2 ! ! . . .
This move is also a double attack.
White threatens to stop the pawn
by 6 tl cl and forces Black to
q ueen.
5 . . . h 1 (®)
left itself. In consequ ence the self­
destructive sacrifice 6 tl h6 + ! re­
sults in stalemate.
The threat of stalemate can also be
coupled with a variety of other
threats in a double attack.
A. Motor, 1 972
494
Draw
White's position is critical. The
pawn e2 can be stopped only by 1
tl g 1 , but this is followed by 1 . . .
f4 , and the attempt to create mating
threats is easily refuted, e. g. 2 dlf7
h6 3 tl b 1 dlh7 and then 4 . . . f3 .
No better is 3 dlg6.c4 4 tl b l c3 .
What can one do?
1 tl g2 ! !
An amazing move ! White literally
forces his opponent to pull out his
pawn and to give check to boot !
I . . . e 1 (®) + 2 c,tlf7 . . . be co-ordinated . This makes it pos­
The situation has been resolved sible for the defending player, when
now. White threatens to give mate there is not much material left on
the board, to use the blockade as a
:md perpetual check s imultane­
means of neutralizing an oppo­
ously, i . e. again a double attack.
nent's piece, for instance the king.
'rhe perpetual check is based 0�
.
stalemate. Thus, after 2 . . . h5 1t I S Here is a very characteristic exam­
possible to continue by 3 bl g8 + ple.
dlh7 4 bl g7 + ! dlh6 (after 4 . . .
� x g7 White is stalemated) 5 P. Ilyin, 1 94 7
!d. g6 + . Having forced Black to
queen, White gets a chance to
threaten stalemate. But the struggle
is not over yet. Black can protect his
position with the bishop and shield
the rook.
2 . . . � g7
White finds himself in a delicate si­
tuation again. But he has a fitting
reply.
3 bl e2 ! ! . . . Draw

Black is again confronted with ex­

tremely bothersome problems. If he
takes the rook, White is stalemated
again. Moreover, the text move is a
double attack aimed at the queen,
and if it moves aside then the king is
checkmated on e8. Strange as it
seems, Black's only defence is 3 . . .
�c3, but after that the rook returns
with a mating threat : 4 bl g2. In this
manner White manages to achieve a
draw in the end-game by alternately
threatening mate and stalemate.
3. The blo ckade
In an attack it is very important that
the attacking pieces should support
each other and that their operations
White is faced with the d aunting
problem of stopping the black
pawn. Apparently in despair, he
gives check with the knight.
1 � e7 + . . .
The king h as to dodge in such a
way that the knight cannot capture
the pawn by means of a double at­
tack. If for instance 1 . . . dlf8 or
1 . . . dlh 8 , then 2 � g6 + and 3
� f4. But if 1 . . . cJJ f7, then 2 � c6
e2 3 � eS + , and the pawn is
stopped after 4 �f3. This means
that the king has only one square i t
can move t o with impunity.
2 g6 + ..1hh
1 . . . rlhh7
'4d \b/ 8
The king cannot evade to h6 be-

245
cause of 3 � f5 + and 4 � x e 3 . But
White has run out of checks and the
pawn is stil l beyond his reach.
Suddenly the idea occurs to him
that it might be possible to exploit
the fact that the king is confined to
the corner. Besides, a lone queen
cannot do anything against a king
and knight. Pursuing this line White
sends hi:; king on a support mission
to the knight.
3 \tb4 ! e2 4 \tc5 e 1 (®) 5 \td6 . . .
Here you can see for yourself that
the "all-powerful" queen is unable
to lift the b lockade, e. g. 5 . . . ® e4
6 \td7 -& eS 7 \td8 tlt1e6 8 00e8
tlt1d6 9 00f7 t!t1d7 10 00f8 ®d8 +
1 1 c11 f7, etc.
There is one point that deserves spe­
cial emphasis. If after 5 \td6
Black's queen had been posted on
e2 instead of e 1 , Black could have
broken the blockade by 5 . . . tlt1g4 !
6 d:>c7 t!t1 d 4 !
Not only the king, but any piece,
the queen included, can be blocked.
F. Zimkhovitch, 1 92 7
496
Draw
246
White has a dangerously weak
pawn b2. 1 � x c4 would therefore
be worthless on account of 1 . . .
rl b 1 , in which case Black captures
the object of contention on b2 and
achieves victory. But White finds a
surprising line that gets him out of
trouble.
1 � g4 + 00d6 2 � f5 ! . . .
Preventing the rook from going to
b 1 . But Black too comes up with a
no less threatening move, which ha­
rasses the pawn .
2 . . Il a2 !
.
The tide of battle seems to have
turned in Black's favour, but . . .
3 � x a2 ! ba 4 OOcl a t ( ® ) + 5
� b 1 , and both sides are marking
time. Black's queen has been corn­
. pletely neutralized . Therefore a
draw !
And here is an example showing the
blockade of a rook.
A. Troitzky, 1 930
497
Draw
Black seems to be well on his way to
victory. He is a rook up and White's
pawns can be stopped easily, e. g. by
1 \tlh5 .bl x e7 2 \tlh6 �e5 3 � d5
\tle8 4 � f7 + \tlf8 , a nd White can
give up, or 1 �d5 .bl x e7 2 � f7
� e5 3 \tlh5 � g7 with the same re­
sult. In the second line, 1 . . .
.bl x g6 would have been a mistake
because of 2 � f7.
White can save himself only if he
succeeds in surrounding the rook.
Let us see how this can be done. It is
very important not to let Black go
1 . . .bl x e7, for this would enable
.
him to launch co-ordinated opera­
tions with the rook and bishop
against the g-pawn.
This is done by 1 � c6 + ! \tl x e7,
because 1 . . . \tl x c6 loses to 2 g7.
Now White initiates a double attack
by 2 �d5 ! which contains two
threats. If the rook evades the blow
on the e-file, then White goes 3 g7
and wins. That is why Black is
forced to play 2 . . . 1:1 x g6, which
enables us to carry out our plan : 3
00h5 rl g7 4 g6 00f6 5 � f7, and the
rook is imprisoned.
But this is not the end of the b attle.
Black can try to force the opponent
to loosen his grip on the rook. But
after 5 . . . 00f5 6 00 h4 ! c,tlf4 7
c.1;lh3 ! � e7 8 c,tlg2 �h4 9 c,tlg 1 a
draw is certain.
On the other hand, attempts to
keep the king in the vicinity of the
black rook end in a disaster. For in­
stance, if 6 c.1;lh6 ? � e5 7 \tlh5 � f6 !
8 00h6 00g4 9 � e6 + c,tlh4 1 0 � f7
� d4, then White is forced to re­
lease the rook from pr�son and
Black wins after 1 1 � e6 rl e7 1 2
� f7 � e 3 + 1 3 00g7 \tlg5 1 4 c,tlfs
00f6 .
In a double attack, a blocking
threat can also be coupled with
other threats.
G. Zakhodyakin, 1 94 9
498
Draw
White's position looks hopeless.
There is no way to stop the Black
pawn.
1 � a6 e2 2 � c7 + cj;lfg 3 � e6 +
c,tlgs
White has run out of checks and the
only way he can save himself is by
blocking the king. But how?
4 � f7 ! . . .
A simple solution. Again we see a
double attack here. If the king takes
the knight then White retaliates
with the concealed threat of stop­
ping the pawn by 5 � g5 + and 6
� f3 . But if the pawn queens then
White locks up his adversary's by 5
� fg5 ! ·and throws away the key.
4. Perpetual pursuit
Not only the king but any other
piece on the board can be exposed
247
to unceasing attacks or perpetual G. Nadareishvili, 1 9 70
pursuit. This aspect of the game has 500
been exhaustively dealt with by
composers of studies.

S. Birnov, 1 928
499

Draw

A sharp position in which White's

p awns are an important strength
factor. But after 1 g7 + 00£7 2
Draw � e8 + ! dig S White's onslaught
seems to have been repelled. At this
White's position could hardly be point White played 3 � c6 ! threat­
worse. The passed pawn aS is easily en ing mate with his bishop at d S .
controlled by the bishop, and Blacks's reply 3 . . ® x d4 does
.

Black's h-pawn cannot be stopped. not provide real protection in view

Although at first sight the pos ition of 4 � d S + ! ® x dS stalemate.
is hopeless, White can still defend This means that the bishop's move
himself. It is only a question of find­ to c6 contains two threats : a mating
ing the right solution. threat and a concealed stalemating
threat. Black can protect himself
1 a6 � x c4 2 e4 + ! . . .
against the m ate by 3 . ®aS.
. .

This thrust catches Black off guard This is followed b y 4 � cS ! ® d 8 S

and confuses his defence. Later on � d6 ! ®gs 6 � e S ! ® d2 7 � d4 !
the king obstructs its bishop's free­ These repeated self- destructive sac­
dom of movement. rifices are made in order to keep up
a perpetual pursuit of a piece.
2 . . . dl x e4 3 a7 �ds 4 c4 ! �a8 S
I should like to point out that the
ct>b8 � c6 6 dlc7 � a8 7 ct>b8 etc.
term "perpetual pursuit" is more
The steering of the black king into comprehensive than the term "per­
an unfavourable position played an petual attack". In the first example
important role. discussed there is no attack at all,
but only the threat of stalemate,
which forces the b lack Queen to
d ance a strange waltz with the
White bishop .

248
Occasionally we also encounter 3 . . . � h2 � 4 �f2 � g4 5 � f3
perpetual pursuit in practical chess. (mate is also possible : 5 ®fl ? � h2
Here is just one example. mate) 5 . . . � h2 with a draw.

Surachov-Buchmann
Kiev 1 96 7 5. Building/o rti.ficatio ns
501
Building an impregnable fortress is
an effective means used in practical
chess. The building of fortifications
occurs most frequently in the end­
game.

Ilivitsky-Averbakh
Moscow 1 950

502

Black t o p l ay

Black's pieces are not exactly what

one might call well placed ; their
main weakness is that they have lost
their coherence. Although the
queen is not being attacked, it is out
of the game, and e4-e5 is threat­
ened. But the cramped position of
the white pieces on the K-side en­
Black to play
ables Black to carry out a perpetual
attack combi nation.
Black seems to have walked into a
1 � g4 � 2 hg (after 2 � f2
. . . trap :
� x e2 3 � x e2 � x f2 + Black
1 . . � X g4 + 2 bi X g4 J:1 X g4 3
has nothing to fear) 2 . . . � X g4 3
.

Q c8 + 00d6 4 � x g4 cbc6 .
E1 fe 1 . . .
White has won a piece, but Black
The link between the bl ack queen
has captured his adversary's passed
and its bishop in itself does not en­
pawn and built an impregn able for­
danger White's king in view of the
tress. Black's king retreats to b 8 , af­
cover given by the knight g3 to
ter which White is unable to get at
squ are h l . But attention is focuss ed
_ the pawn a7. After a few more
not so m u ch on attacking the white
moves the contenders a g reed to call
king as on curtailing the queen's
it a draw.
freedom of movement to make it
possible to pursue it constantly.

2 49
Flohr-Lilienthal And ag<:.in we have to ask study
Budapest 1 9 50 composers for help, for they have
503 done a very thorough job of ex­
amining the building of fortifica­
tions.

F. Zimkhovitch, 1 926

504

Black t o play

White has a queen against Black's

rook and is one pawn down, but the
struggle takes place exclusively on
one side. Its result depends on
whether or not White can take the Draw
fortress by storm. What followed
was : White is unable to stop the pawn.
The next best thing seems to try
1 . . � f5 ! 2 � x f5 . . .
building a fortress for his king to
.

The more persistent line was of hide in by 1 f6 . Now let us see if

course to play 2 � f4 and then to try Black succeeds in destroying it.
breaching Black's fortress by advan­
1 f6 h2 2 00f8 h 1 (�) 3 00g8 dld7
cing the h-pawn. But White thought
he could win easily after the ex­ Black's only chance is to advance
change thanks to his passed pawn. his king to f5 and then-by offering
up his queen against the knight-to
2 . . . gf 3 00g2 f4 4 h4 00h7 5 00f3
capture the pawn at g5. If he suc­
bl e3 + 6 00 x f4 bl e6 7 00f5 bl g6 8
ceeds in carrying out this plan he
h5 il h6 9 OOg5 . . .
wms.
White has pulled up his forces close To frustrate this undertaking once
to the enemy's defensive wall, but and for all, White simply has to
he was unable to overcome it. The cover the pawn at g5 with his
contest then took the following bishop. To do this he has to trans­
course : pose the bishop to square h6 or to
the diagonal c l -f4.
9 . . El e6 1 0 � d 8 00g7 (the queen
First, l e t us try transferring the
.

must not be allowed to move to f8)

bishop to h6.
1 1 � d4 + 00h7 12 � h4 bl h6 1 3
�b4 00g7 1 4 ®b8 !l e6 draw. 4 � f8 ®a8 !

250
The bishop must not be allowed to moving to the saving square h6. Let
get to h6. us try building our fortress again.
5 dlg7 dle6 6 digS dlf5 7 dlg7 . . . 5 f6 h2 6 � f8 ! h1 (®) 7 � h6, and
Black can do nothing to keep the
Again the bishop is prevented from
king from moving to the safe re­
going to h6. Black now has the upper
fuge.
hand, which he can retain easily.
White had to work with surgical
7 . . . ® h 1 8 dlg8 ® X h7 + 9
precision up to the very end. If he
00 X h7 \1) X g5 etc.
had played his king behind the pro­
But what would have happened if tective wall straight away and con­
instead of 7 dlg7 White had simply tinued with 6 00f8 h 1 (®) 7 00g8 ,
played 7 dlh8 ? Black would have then Black would have had a strong
won by 7 . . . ® x f8 + ! 8 � x f8 reply at his disposal : 7 . . . ®h2 ! ,
dl x g5 9 dlg7 OOf5. which would have tipped the scale
Perhaps the bishop should be in his advantage. 8 � f8 would then
placed on the diagonal c l -f4 ? Let have been followed by 8 . . . ®b8 !
us see what happens. 9 dlg7 dld7 1 0 00g8 dle6 1 1 dlg7
dlfs 1 2 00g8 ®a8 ! 1 3 dlg7 ® h 1
4 �b4 00e6 5 �d2 dlfs 6 � e3 . . .
a n d 1 4 . . . ® X h7 + .
Black succeeds in driving the bishop The building of a fortress is not al­
off the vital diagonal. ways the obvious thing to do ; in fact
sometimes it is anything but obvi­
6 . . . ® f3 7 �d2 ®e2 8 � c l ® e t !
ous.
9 �a 3 ® h 1 and then 1 0 . . .
® X h7 + and 1 1 . . . 00 X g 5 .
V. Chekhover, 1 94 7
W e can conclude from this that 1 f6 505
is not enough for a draw. Perhaps
we were a bit too hasty in deciding
to build a fortress? Let's take an­
other look at the original position.
The black king's posting prompts us
to stop the pawn with the bishop .
Let us try it:
1 � f6 00d6 2 � e7 + OOeS 3 � d8 !
dld6 4 � e7 + dlc6
By threatening the king in this fash­ Draw
ion we have forced it to step on the
diagonal h 1 -a 8 . Now the newly White's three extra p awns seem to
created queen cannot move to a8 . outweigh the disadvantage of hav­
This means that Black is no longer ing only a bishop against Black's
able to prevent the bishop from rook. But such superficial assess-

251
ments are often incorrect. Black can f. Hasek, 1 93 7
threaten to penetrate with his rook
and capture the pawns on f- and g­ 506

files, after which victory would only

be a matter of time.
Two variants to exemplify this : 1
cbc2 Et h2 2 � fl Et x f2 3 � d 3
!l g2 or 1 �f3 !l f8 2 � d 1 Et x f2 3
g4 Et g2 4 cbc2 Et g3 , and Black tri­
umphs by transposing his king to
e4. We can concl ude from this that
the loss of a single pawn on the K­
side can lead to defeat. D raw
So is there anything White can do
to prevent the loss of a pawn ? There White's pos ition looks hopeless in
is, although the line is very u ncon­ view of the incurably weak pawns
ventional and involves the offering g2 and c2. For this reason conven­
up of the bishop. This idea seems tional lines such as 1 Et e6 cbg7 2
absurd at first. It is indeed paradoxi ­ El X e5 !l h 8 3 !l f5 !l h2 4 E1 X f4
c a l that White, through an appar­ El X g2 5 El g4 + d/f6 6 d/b 1
ently senseless sacrifice of a bishop.' !l g 1 + 7 cba2 El c l 8 E1 x g3
manages to drive the rook out of hts Et x c2 + cannot help improve
own camp and to build an impreg­ White's position because Black re­
nable fortress. tains the upper hand by capturing
There is only one way of carrying one of the pawns b3 or d3. There is
out this plan : 1 did 1 ! Et h2 2 die 1 only one line that leads to a draw :
(or cbe2) Et x g2 3 cbfl Et h2 4 1 ebb 1 ! cbg7 2 Et h6 ! ! . . .
cbg1 E1 h6 5 f3 ! Et e6 6 cbfl cbf7 7
cbf2 In this highly unusual manner
White wins decisive tempo enabling
A unique position � Black is a rook him to move his king to the v ital
up but he is unable to tip the scale in squ are e2 in time.
his favour. White has built a bastion
from his pawns whose entrances 2 . . . d;> X h6 3 d/c1 digS 4 did 1
and exits are reliably watched by Et h8 5 cbe2 Et h2 6 cbfl Et h 1 + 7
the king. d;>e2 Et cl or 7 . . . Et g 1 stalemate !
Sometimes the building of a fortress The alternative line after White's
is motivated by plans to achieve 6th move 6 . . . Et X g2 7 d;> x g2
stalemate, etc. Here are two exam­ d;>h4 8 cbg 1 cbh3 9 d;>h 1 g2 + 1 0
ples of this. d;>g 1 with a draw is not any better.
The next study is even more remar­
kable.

252
f. Hasek, 1 932 he is unable to achieve overwhel­
507
ming m aterial superiority.

Draw
Black is a rook up and the inconve­
niences caused by the cramped plac­
ing of his pieces are apparently o nly
of a transitory nature. It is not easy
for White to find a solution. Eve n
an experienced grandmaster will be
astonished at the solution .
1 � a7 + ! . . .
White has little enough material as
it is and now he gives away his l ast
piece. But in this position each
tempo is valuable. To win time
White does not hesitate to sacrifice
6. Bala nce
offorces leadi ng
to a draw
When a player on the defensive
who is weaker in terms of m aterial
manages to restore material equilib­
rium through a combination , then
th is comb ination does not differ ba­
sically from the winning combina­
tions designed to win material,
which we have already dealt with.
What we are interested in here are
combinations which do not restore
equilibrium but which bring about a
balance of forces that theoretically
ensures a draw.
Let us start out with a very old posi­
tio n .
G. Greco, 1 623
508

his bishop.
1 . . . E1 X a7 ( 1 . . . dl x a7 2 b6 +
dlb8 etc. changes nothing) 2 b6
l:l a8 3 dlc3 � x g4
Black's plan is clear : h e wants to
penetrate into the e nemy camp with
his rook along the o nly open file.
4 dld2 dlc8 5 OOe 1 d/d7 6 �f1
l:l g8 7 ddg1 ! �h 5 + 8 OOh l ! Black to play, draw

White's king has at l ast reached the

secl uded spot. In view of the threat­ The solution is elementary :
ened stalemate Black's rook can n ot l . . . El a 1 + 2 rl f 1 El x f1 + 3
penetrate at g2, which means that dl x f1 � h3 !

253
Black's position will theoretically
lead to a draw by his exchanging
the bishop for the pawn g2.
The next example is much more
complex.
M. Botvinnik, 1 9 52
509
Draw
Black is a knight up against White's
extra pawn. But the knight is badly
placed. What line should White
take to ensure a draw? Theoreti­
cally he has two options : the first is
to capture the knight in exchange
for a pawn, the second is to ex­
change his three pawns for his op­
ponent's two.
It is not difficult to capture the
knight. After 1 dlc3 it is trapped.
What worries him is the counter
1 . . . fS ! that Black has at his dispo­
sal. He cannot meet it by 2 eS be­
cause of 2 . . . f4 3 dlb2 gS 4
dl x a 1 g4 5 dlb2 f3 , and a pawn is
on its way to the queening squ are. 2
ef + would not do either, because
of 2 . . . dl x fs 3 dlb2 dle4 4
d/ X a 1 dl X d4 5 d/b2 gS 6 dlc2
dlc3 7 did 1 ffif2 8 g4 dlf3, and
l31 ack wins.
254
Consequently h e cannot take the
knight with impunity. Let us try
finding a continuation which will
enable White to exchange his three
pawns against his adversary's two.
Let us see what happens for in­
stance if we go 2 d S + .
If Black answers with 2 . . . dld6
then the pawn exchange plan is fea­
sible : 3 ef gf 4 g4 ! fg 5 dld4 ! � c2 +
6 dle4, and the pawn at g4 is lost.
But now let us assume that Black
decides to play 2 . . . dieS. In that
case the d-pawn lends a helping
hand again : 3 d6 ! dl X d6 4 ef gf S
g4 ! fg 6 dld4 ! and again a draw.
Everything seems to be clear. But
Black springs a surprise. 2 dS + is
followed by 2 . . . dlf6 !, a magnifi­
cent move which brings White's
plan to naught, because after 3 ef
dl x fS 4 dlb2 dieS S dl x a 1
d;> x ds 6 dlb2 dld4 7 dlc2 dle3
Black is in control.
We h ave made a few mistakes, but
now we know where the right �olu­
tion lies. The thrust with the cl­
pawn has to be made when square
f6 is still occupied by the pawn.
Thus :
1 dS + ! dies 2 dlc3 fs 3 d6 ! dl :d6 4
ef gf S g4 ! etc.
If Black goes 1 . . . dld6 this time
then 2 ®c3 dieS 3 g4 ! with a clear
draw.
A balance of forces that ensures a
draw can n aturally also be achieved
with the aid of a double attack.
M; Perelman, 1 9 5 5 rook nor leaves his bishop un­
covered : 4 . . . Ll eS. But after 5
510
� f4 + dlg4 6 � d3 White captures
the bishop and in view of the bal­
ance of forces a draw is ensured.

The attack
on the king

How do contacts c.o me about in the

Draw course of a game ? How are they co­
ordinated, or more precisely, what
White is a whole rook down here . are the contributory factors that
To achieve a draw he has to solve help bring about combinations ? We
the daunting task of saving the propose to make a more detailed
knight and capturing the oppo­ study of these questions on the basis
nent's bishop. He has no choice but of the attack on the king . Normally
to try a pawn move. an attack on the king is the upshot
1 e7 bl e4 + of a strategic plan whose objective
is to move pieces (and sometimes
Where should he move his kin g ? also pawns) up to the enemy king so
Since he h as to keep an eye on the that they can attack or at least
bishop, he has only two moves to threaten to attack the king from
choose from : 2 did 1 and 2 dlfl . their positions.
If White opts for 2 did 1 , he finds We will deal here only with those
that after 2 . . r:l x eS 3 � d6
.
cases in which the forces have al­
bl x e7 4 � fS bl eS he has gained ready been moved up to the king.
nothing. But to 3 � c7 bl x e7 4 The attacker's pieces are ominously
� d S Black has the reply 4 . . . bl d7 hovering over the opponent's king
at his disposal. and waiting for a tactically oppor­
In view of the fact that the knight tune moment. Combinations sig­
got in the king's way perhaps it was nalling the final act of the attack de­
wrong to move it to d 1 . In that case velop .
let us try 2 dlfl r:l x eS 3 � c7
r:l X e7 4 � d S ! . . .
The white knight wants to "fork"
Black's pieces, i. e. a double attack
is in the offing. Black has only one
reply which neither costs him the

255
Hewitt-Steinitz Black may not have sufficient mate­
London 1866 rial to do him any harm.
51 1 4 ci>h4 � g2 + 5 ci>h5 . . .
Shaking a little and surprised at his
own audacity, the lone king plunges
into battle against overwhelming
odds. Having dispatched the
mighty queen and the rook, he tries
to prove his prowess in hand-to­
hand combat. But despite the fact
that Black has much less material
left on the board than White, his
Black to play strength on the King's side is still
entirely sufficient to seal the fate of
We have in front of us a typical at­ the enemy king. The hour of reck­
tacking position. Black pieces have oning has come.
taken up positions directly in front
of the king's fortress, the rook has 5 . . . El f5 + 6 d/g4 hs + 7 ci>h3
already penetrated the defences. .E1 f2 mate.
The question now is how to attack Please note the mating mechanism,
the king, how to stage a combined which is actu ally in the form of a
attack on the king and how to double attack. Whilst protecting the
checkmate it. knight, the rook paves the way for
Steinitz solved the problem in an in­ the bishop to deliver the decisive
structive fashion. blow. At an early stage Black had
1 . . . !j X g2 + ! realized drat White's bishop and
pawn would play an important part
This sacrifice is the mine that in hemming in the king.
sweeps away the last bastions·. The There was another line Black could
white sovereign is left virtually h ave taken to win the game, but it
without pawn protection. wou ld have meant making an extra
2 d/ X g2 ® h3 + ! ! 3 \tJ X h3 move : 6 . . . E! f4 + 7 ci>hs � e 8 +
� e3 + 8 d/g5 h6 mate. What with the king
contending unaided and exposed
Black forces the king to come out against three pieces and two pawns,
into the open. If White had refused the outcome is not surprising.
to accept the sacrifice and retreated This very impressive combination
with his king to gl instead, there of Steinitz consists of three phases :
would have been no defence against 1 . Destruction of the king's fortress
mate after 3 . . . El f2 . But now the and cork-screwing the king. 2. Pur­
king is forced to face his destiny, suit of the king. 3. Setting up of the
harbouring the secret hope that mating mechanism.

256
As a rule mating combinations are 2 d/ X f2 � g4 + 3 ffif3 . . .
made up of three phases. In the first
The king has to c.o me forward , for
phase the king being attacked
if it retreats it is checkmated by the
usually loses its p ::�.wn defence
bishop on e3. This is the start of a
through a sacrifice and comes un­
chase whose purpose is to involve
der enemy fire from hostile pieces.
an additional piece in the attack and
In the second phase the king is
with its aid to force the king back
driven to the spot where the last act
into the mating net.
takes place, i . e. the combined at­
tack and the mating thrust. 3 . . . e4 + � 4 ffi X e4 . . .
Here is another example of such a If the pawn takes, then mate with
three-phase mating combination . knight d7 :lt eS follows i mmedi­
ately.
0 'Kelly-Defosse
4 . . . � df6 +
Belgium 1 936
Check can also be given with the
51 2 other knight ; the outcome is the
same.
5 ffif3 � eS + 6 cbf2 � fg4 + 7
ffig 1 � e 3 mate.
Black's pieces have collaborated
magnificently, while White's vast
army stood around and watched.
The rook e 1, the bishop h 1 and the
pawns h2, g3 and e2 even played a
negative part by hemming in their
Black to play
king. Not every mating combina­
White has j ust played the knight to tion is necessarily made up of three
d4 hoping to wrest the initiative distinct phases, of course. For in­
from h is adversary. If Black takes stance, if the king's position has
the knight, White recaptures with been destroyed in the course of the
his queen and h is plan is vindicated. game then it is no longer necessary
But events take an entirely different to smash the king's fortress.
course.
1 . . . ® x f2 + ! !
With this magn ificent sacrifice
Black breaks open the king's for­
tress . White's kin g immediately
finds itself in a cross-fire from the
enemy pteces.

257
Mackenzie-Mason carried o u t in a number o f different
Paris 18 78 ways and assume different forms.

513
Here is a simple but characteristic
case of an attack with two rooks.

Polugayevski-Szilagyi
Moscow 1 960

514

White to play

Black hopes to get his king safely to

the Q-side, e. g. if 1 � hfS + then
1 . . d)fs 2 -&1 h6 + d)e8 . But the
.

weakened position of Black's king

White to play
enables White to launch a decisive
mating combination without delay.
Black's king is in danger. It has been
t &th6 + ! ! . . . cut off from the g-file, and there is
no escape for it if White manages to
Dragging Black's king out i nto the
transpose his other rook to the h­
open.
file. This is very easily done : 1
1 . . d) x h6 2 � hfs + �
. x fs 3 � f8 + ! El x f8 . The sacrifice di­
� X f5 + . . . verts Black's rook from i ts task of
keeping watch on its White oppo­
Starting a hunt that does not l ast
site n umber. The White bishop has
very long.
to be take n, for if the king retreats it
3 . . . d)h5 4 g4 + d) x g4 5 rl g3 + is i m mediately checkmated. But
d)h5 6 � e2 m ate. after 2 El d3 Black is left without
defences.
Since the king had to contend with
superior White forces unassisted,
there was another, even easier line
White could have taken to win the
game : 4 El h3 + d)g4 5 � h6 m ate.
As soon as the king is deprived of its
defensive pawns it can be threat­
ened by different pieces, which is
why the combined attack can be

258
Miszto-Kloza Now let us examine a mating attack
Poland 1 95 5 with a rook and a knight.

51 5 Rossolimo-Reissman
San Juan 1 96 7

White t o play

The first impression is that Black is White to play

better off, White's bishop at d4 be­
ing uncomfortably pinned. B ut first As soon as White h as eliminated the
impressions are misleading. The knight d5 through exchange, he
bishop can support the two rooks launches an attack on the king dis­
despite the pin and two rooks are a regarding its seemingly reliable
force to be reckoned with. But how pawn defences.
can they be mobilized for an attack? 1 � X d5 cd 2 g f6 + 00h8 3 ®g6 ! !
By sacrificing the queen.
t ® h7 + ! ! . . . Moves such as this are a source of
In this manner the black king willy­ delight. Although exposed to attack
from two pawns, the queen cannot
nilly finds itself drawn into the con­
flict. be touched because if 3 . . . fg then
4 g x g6 + hg 5 � h3 mate. At the
1 . . . 00 X h7 2 � X g7 + 00h8 3
s ame time mate is threatened by 4
� g8 + ! . . .
® X h7. To ward off this threat
As you see the pinned bishop plays a Black covers the point h7 with his
very active role indeed. Since there queen.
is no defence against double check
3 . . . ® c2 4 � h3 ! . . .
the king has to step aside.
This quiet m ove dots the i.
3 . . . ®h7 4 � l g7 + ®h6 5
EI g6 + ®h7 6 � 8g7 + ®hs 7 4 . . . ® X g6 5 g X g6 + fg 6
Ei h6 mate. � x h7 mate.
By covering the rook at g7 the The ability to recognize in time
bishop also plays an important role contacts that are characteristic of
in the final stage of the game. certain pieces is a great art.

259
Mackenzie-N. N
London 1 891
51 7
White t o play
Here the contacts needed for a co­
ordinated connection are already
extant. The rook d 1 and the bishop
g5 create favourable conditions for
a double attack on point d 8 . The
problem consists in opening the cl­
file and in diverting the opponent's
queen and rook from covering the
square d 8 . The sclution is relatively
easy.
1 El e5 + ! de 2 ® e5 + ! ®
x X
Black could have prolonged the
game by 2 . . . � e7 , but this would
not affect the outcome : 3 ® x h8 +
�f8 4 ® f6 with the threat of 5
� c6 + ® x c6 6 .!d d8 + and mate
on the next move.
3 �c6 + .!d x c6 4 .!d d8 mate.
The heroes of the final stroke, i. e.
pieces assigned the job of check­
mating the king, are not always the
obvious ones to pick for this assign­
ment.
260
518
White to play
The first move- 1 ®g5 !-is easy
enough to find, but how does White
keep up the pressure after 1 . . . g6
without giving his opponent time to
recover? Although not so obvious as
the first move, there is a line White
can take which will force a win and
in which the inconspicuous bishop
plays the leading part.
2 ® h6 ! ! . . .
Co-ordinating the contacts against
point h7, and then immediately
against fl
e5
·
x
2 gf
. . .
The task facing White now is to re­
m ove the pieces obstructing the dia­
gonal b 1 -h7. This is easily done :
3 bj_ g4 + ! fg 4 � X h7 + d/h8 5
�g6 + cbg8 6 ®h7 + dlf8 7
® x f7 mate.
Another example along these lines.
Bauer-Gol!ner loosened , they had not been
Berlin 1 956 severed , because the knight had to
519
restrict the freedom of movement of
the bishop b2. The bishop's full con­
tacts with squares g7 and h8 were
restored by diverting the pawn g7
and the knight f6 .
To avoid unpleasant surprises one
should pay attention to developing
contacts especially in sharp situa­
tions.

Stein-Portisch
White t o p l ay Stockholm 1 962
520
It would seem that Black should
have no trouble repulsing White's
attack. Thus 1 � f5 is naturally fol­
lowed by 1 . . . d1h8, and the dia­
gonal a 1 -h8 seems to be securely
protected. But is the long diagonal
really as well protected as it seem s ?
I f it were free, then White could
win by checkmate by playing the
bishop to fS . And the diagonal can
be opened-by two consecutive s ac­ Black to play
rifices :
The second player did not want to
l l:l X h6 + ! gh
exchange on c4 having played
Black cannot avert disaster even if � d7-b6 ? instead with a view to
he plays 1 . . . d1 x h6 because this forcing his opponent to decide
is followed by 2 ®gS + d1h7 3 where to leave the white-square
® h4 + d1g6 4 fS mate. bishop. But this resulted in a brief
weakening of Black's K-side de­
2 ® g8 + ! . . .
fence, which is somewhat uneasy in
This sacrifice diverts the bishop and the presence of the two minor
at the same time hems in the king. pieces. Stein made masterly use of
this circumstance and played :
2 . . . � x g8 3 � fS mate.
1 � X g7 ! ! . . .
That's how easy it is ! We should
bear in mind, however, that al- A bolt from the blue ! White can af-
though the contacts of the knight f6 ford to forsake not only his bishop
with the squares g7 and h8 had been at c4 b ut even his queen.

261
1 . . . � X c4 2 �f6 . . .
Threatening mate in two by 3
� f5 + 00g8 4 � h6 mate. Black's
last reply was 2 . . . � e7, but he re­
signed after 3 ® f3 and justifiably
so, because 3 . . . � x f6 4 ® x f6
� d7 leads to mate on account of 5
Lj X d7.
If the pieces needed for the mating
mechanism are already in position,
then the attacker's task sometimes
consists in counteracting co-ordi­
nated operations of the opponent's
forces.
The following study is an admirable The desperate attempts made by tlw
illustration of this procedure. black pieces to help their king are w
no avail. Mate is unavoidable.
0. Duras, 1 93 9
521
White to play and win
The king is obviously in a precar­
ious situation, but it feels confident
with so many pieces on its side.
They will not fail to come to its as­
sistance i n time, or will they? And
besides, the pawn e2 is on.the
threshold of queening. And yet . . .
1 f7 ! . . .
White diverts the bishop from the
262
fourth rank by sacrificing his l ast
pawn.
1 . . . � X f7 2 Lj g4 ! �b3
The alternative is mate. But now the
bishop blocks an i mportant square
for the knight a 1 .
3 rl g l ! � c2 4 rl gs . . .
If White had played the rook to gS
one move earlier, Black would have
s aved himself by 3 . . ds 4 00b6 d4
.
5 dlc7 � d 5 6 ld x ds � b 3 !
4 . . . d 5 5 d:>b6 d 4 6 00c7 !
The importance
of concrete calculation
of variants
We have seen that in the course of ::1
game the number of various con­
tacts between pieces increases as the
opposing armies draw closer. At
first they h appen casually and can
be co-ordinated only in exceptional
cases-usually as a result of a bad
blunder. But as the heat of the battle
intensifies more and more contacts
occur and it becomes increasingly
difficult to keep track of them. The
resulting co-ordinations of con­
tacts, i. e. combinations, often come
as a surprise to the opponent. This
happens because he has not been
following the development of
events with sufficient care or he has
been too engrossed in pursuing his
own plans to give due consideration
to those of his adversary.
You will agree that critical tactical
situations warrant close attention.
General reflection is not enough
here-precise and concrete calcula­
tions are called for. Contacts, which
we have dealt with exhaustively, are
beacons designed to help the chess­
player find his way in a sea of tacti­
cal complications. As soon as we see
that contacts are beginning to take
shape between our pieces we should
seek ways of co-ordinating them,
i. e. we should keep an eye open for
possible combinations. At the same
time we should be on the lookout
for contacts between the oppo­
nent's pieces and either eliminate
them as a precaution or at least take
any measures we can to prevent
their co-ordination. Combinations
require precise and sometimes long­
term calculation. It is naturally
sometimes impossible to calculate
all lines to the very end. In such situ­
ations one's experience and faith in
oneself and in one's i ntuition will be
helpful. Nevertheless, the main lines
have to be calculated and checked,
otherwise one may be in for some
unpleasant surprises. Precision and
care are nowhere as crucial as in
combinative complications. An in­
significant change in the order of
moves or failure to take the strong­
est line can instantly decide the
success or failure of a combination.
Let us consider some examples.
Martynov-Go/stein
Soviet Union 1 9 76
522
White to play
White's disposition of forces threat­
ening the King's side is more than
enough to make up for the disad­
vantage of his being �ne pawn
down. All he has to do is to castle
and go over to the offensive. But
White felt that he could launch and
assault on the hostile king even
without castling, and initiated a
complex and very effective combi­
nation.
1 � c6 ! . . .
This sacrifice opens the diagonal
for the bishop b2.
1 . . . � x c6 (if 1 . . de, then 2
.
bj X g7 + ! 00 X g7 3 ® g4 + 00h8 4
® h 4 ® x d3 5 � x f6 + , and
White wins) 2 bl x g7 + ! 00 x g7 3
® g4 + 00h6 !
The only move. Weak is 3 . . . 00h8
because of the unavoidable mate af­
ter 4 � x f6 + � x f6 5 ® h 5 .
4 ® f4 + 00 h 5 5 � e2 + ( ?) . . .
An inaccurate continuation, show­
ing that White had not anticipated
263
the king's being played to h6 or that opportunities this position offers.
he had acted on general considera­ Although Black has other lines of
tions such as "The king has moved d efence at his disposal, none of
out into the open, there must be a them seems to be sufficient to save
way to checkmate it". But Black has him.
achieved substantial material supe­ One thing is clear : White's attack is
riority that has made it imperative on the razor's e dge, but it gives him
to make at least a rough check of good prospects even if Black finds
the line of play. the best moves , which in some cases
The correct and logical continua­ is not easy to do. But after the woe­
tion of the attack was 5 0-0-0 ! If ful queen check the attack suddenly
Black replies 5 . . . � dS then 6 peters out.
� e2 + d/g6 7 ® g4 + with m ate at
6 . . . d/h6 7 � e4 . . .
next move.
Granted, the sally 5 . . . � d5 is no 7 ® h4 + brings nothing. The sec­
good because Black himself re­ ond player flees by 7 . . . d/g7 8
moves the knight defending the ® g 5 + d/h8 9 � e4 Id. g8 and de­
king. The correct reply was 5 . . . fends himself successfully.
� g4 ! with a view to replying to 6
7 . . . Id. g8 8 ® h 4 + d/g7, and
h3 with 6 . . . � x f2 . The attack
Black won.
can then take the following course :
7 � e 2 + d/g6 8 Id. g l + �g5 9 It can be boldly stated that in criti­
® x f2 ! for the time being White is cal tactical situations, nonconcrete
one rook down, but thanks to the approaches based on general con­
threat 10 h 4 he wins a piece and siderations are fatal. Especially in
Black's king remains in an exposed sharp positions everything depends
position. on very precise and concrete calcu­
lation.
5 . . . d/g6 6 ®g3 + ? . . .
We should like to acquaint you wilh
White has lost the thread. Of course
a case in which the author of this
he should have played 6 �d3 + in
book came out second best.
order to be able to counter 6 . . .
d/h5 by 7 0-0-0 thereby initiating
the variant just discussed.
Black, however, has an effective
counterstroke at his disposal 6 . . .
� e4 ! which is based on 7 � x e4
� b4. But even if Black embarks on
this course White retains a very
strong attack, e. g. 7 � e2 ! fS 8
® f3 ! with the formidable threat of
9 ® h 5 m ate. Naturally this variant
does not take full advantage of the

264
A verbakh- Tal may find that the ensuing position
Riga 1 958 has some subtleties which refute the
523
original assessment.

Simagin-Beilin
Vilna 1 946

524

Black t o play

In this position Tal sacrificed a

piece for two pawns :
1 . . . g x e4 2 � x e4 � f5 3 � fd2 White to play
� x d5
To win a pawn , White embarked on
White could have repulsed the at­ a combination involving-as he
tack and retained an advantage in
thought-the temporary sacrifice of
material by 4 � g3 ®e7 5 � f3
a piece .
.!d. ad 8 6 .!d. e t ! although the game is
still very involved and sharp. I had l � g6 fg 2 ,!d. X e6 'f!tlf7 3 � X b7
gone through several similar var­
iants when I sudd enly decided to
Black may not touch the knight be­
recapture a pawn immediately by 4
cause of 4 .!d. x f6 + , but in this case
� x d6. After that I hoped to be
White is a pawn up.
able to find a way out of the situa­
Apparently this is as far as White's
tion with one p iece up against a
calculation went. But there fol­
pawn. Not having thoroughly ex­
lowed an unexpected blow.
amined the consequences, I took (

the pawn on d6. But already after 3 . . . � d5 !

4 . . . g f6 it was clear that White
immediately loses a piece and winds
up with a pawn minus. That is how I
was punished for my nonconcrete
approach .
If one decides to play a combination
it is very important to make a cor­
rect and accurate assessment of the
ensuing position . Otherwise one
Black used a standard d ouble attack
combination. He threatened mate
after giving check on f2, and the
knight on b7 is hanging. White had
no alternative :
4 .!d. e2 ® x b7 5 .!d. c5 E! ad8 6 E! e 5,
but after 6 . . . ® f7 he was unable
to recover the piece and lost.

265
The danger of such unexpected � f3 and the queen is trapped.
counterstrokes is particularly great
And here is an example in which my
in sharp and tactically intricate situ­
opponent underesti mated the
ations . That is why in calculating a
strength of an attack.
combination one should examine
very closely even the obvious lines
Averbakh-Zita
to avoid unpleasant consequences.
Szczawno Zdr6j 1 9 50
Guldin-Averbakh 526

Lugansk 1 9 5 5
525

White to play

The position is extraordinarily

Black to play sharp. Black's king stands exposed ,
but it is protected by a queen and a
I had been working up to this posi­ rook and there seems to be no way
tion for several moves inwardly of getting at it.
hoping for a concealed combina­ On the other hand, the White king
tion. My partner had hardly played is relatively safe for the present, al­
� f3-d2 before I instantly replied though Black has two strong centre
1 . . � X f2 2 d) X f2 ® h4 + .
.
pawns which are about to move for
The answer 3 g3 seems to be forced, ward and threaten the king.
for if the king tries to evade the at­ On balance, therefore, the two
tack by moving to e2 then 3 . . . sides have good chances . The only
� x e3 + follows, but if it moves to question is who will be the first to
g 1 then the rook e 1 is lost. What pose real threats.
h appened then was 3 . . . ® x h2 +
4 � g2 � f6 ! with a strong black at­ 1 ® fs d3
tack which quickly led to victory. What should White do next? If he
But when I went about preparing a takes the intrusive pawn he expose�
commentary on the game I dis­ his king, if he does not, then the
covered that there was a "hole" in Black pawn threatens to continue
the combination. After 2 . . . its advance .
® h 4 + the king could still return to In situations such as this one should
g l , because if 3 dlg l ! ® x e l then 4 weigh every change in position very

266
CJ.refu lly no matter how insignifi­ Playing in such situations can be li­
cant, because they may give rise to kened to a balancing act u nder­
astonishing combinations. neath the roof of a circus tent where
the slightest inattentive ness leads to
2 r1 f4 . . .
loss of balance and fall.
White simply disregards Black's
We shall conclude this chapter with
threat and throws another rook into
an example from games played by
the attack.
candidates contending for the
2 . . . d2 + 3 did 1 e3 world title.
White's king is in mortal danger
Fischer-Larsen
now. Black threatens e3-e2 +
Denver 1 9 71
against which there is no straight­
forward defence . 4 r1 e4 is still fol­
lowed by 4 . . . e2 + and Black re­
tains the upper hand through 5
ld x e2 'd:!:t x e2 + 6 dl x e2
dl (®) + . To 4 E! g4 + Black also
has a combination at his disposal :
4 . . El g6 and if 5 r1 x g6 + hg 6
.

® x g6 + then 6 . . . dl x f8 and
Black is a roo k up.
And yet White takes up the gauntlet
because he has spotted a way of re­
futing the combi nation. Black to play

4 ld g4 + El g6 5 h6 + ! . . . Larsen is usually known for his ex­

This pawn sacrifice has brought
about a radical change in the situa­
tion . The king is steered to square
h6 and the h-file has been opened .
5 . dl X h6 6 Lj X g 6 + h g 7
. .
'd:!J h3 + ! . . .
This is the pu nch-line ! Black is
traordinary i nventiveness. In this
position his king is in a very un­
pleasant situation. His opponent is
obviously i n control of the situation
here, but for all that the Dane
found a brilliant combination .
1 . dlg6 !
. .

doomed beyond salvation : 7 . . . His intention was to wrest the in­

dlg7 8 ® h 8 mate or 7 . . . digS 8
®h4 mate.
One may venture to say that eve n
first-rate grandmasters are hardly
immune to mistakes in sharp and in­
tricate tactical situations, especially
when playing u nder time pressu re.
itiative from Fischer after 2 ld X eS
® X eS 3 ® X d7 by 3 . . . ld ad 8
with threats against the king. And
indeed they look quite dangerolls.
But Fischer accepted the challenge
after having made a thorough cal­
culation of the contingencies.

267
 and which formed the crowning
2 El X eS 'f!t1 X e5 3 'f!t1 X d7 El ad8 4
® x b7 ® e3 + 5 \1lfl El d2 conclusion of the contest. In analys ­
ing these games the reader should
White seems to stand bad , but
pay special attention to the forma­
Fischer had foreseen everyth ing.
tion of contacts and their co-ordi­
6 ®c6 + ! l'l e6 7 � cS � . . . nation against a certain object of at­
tack. We hope the analysis of the
White's cou nterplay culmi nates in
games will give you pleasure and
this move. Larsen should have an­
help you get a firm grasp of the the­
tici pated this development, but he
oretical principles of contacts and
did not. He fo und a way of winning
combinations o n which this book is
the queen, but the price was too
based.
high.
In acquainting you with the combi­
7 . bl f2 + 8 d/g l bl X g2 + 9
. . national creations from the past and
d:> x g2 ®d2 + t o \1lh l rl x c6 1 1 present we should like to stress once
� X c6 'f!t1 X c3 1 2 El g 1 + ffif6 1 3 again that the combination is the
� x a7, and White manages to summit of creative chess and that it
bring his superiority in material to appeals equ ally to the intellect and
bear. the heart of m an.
Only very recent examinations have Queen 's Indian Defonce
shown that Black could have main­ Zukertort-Blackburne
tained the equilibrium by playing London 1 883
4 . . . 'f!t1 x c3 ! instead of 4 . . .
®e3 + . A line worked out by l c2-c4
I. Saitsev runs as follows : 5 � c6 +In the 1 9th century, when the open
digs 6 � c l + f4 7 h4 + ! d:>fs ! s games dominated the scene, Eng­
g4 + � fg 9 d/g2 'f!t1d4 1 0 d;J X g3lish m asters frequently started the
'f!t1 x a 1 , and White gets nothing game with a double step of the
more than a perpetual check. bishop's pawn. This is how the
name "English Opening" origi­
nated . But this opening often side­
tracks i nto other beginnings such as
S ample games
the Queen's Gambit, Queen's Pawn
Opening or-as in the present case­
In conclu sion we should like to the Queen's Indian Defence with
acquaint you with some instructive the colours reversed. At any rate
games, all of which ended in an ori­ this is what we would call the devel­
ginal combination. We tried to pick opment adopted by Zukertort.
those duels from a vast nu mber of
1 ... e7-e6
such games in which the combina­
2 e2-e3 � g8 -f6
tion was in the making throughout
3 � g l -f3 b7-b6
the preceding course of the game
4 � f l -e2

268
The current preference in such posi­ could have availed himself of
tions is to move d2-d4 and bring 1 2 . . . de ! which i s an ingenious
out the bishop to d3 where it is of move based on concrete calcula­
course more actively placed . tio n . If White hits back with the
bishop Black can immediately
4 ... � c8-b7
threaten point g2 by 1 3 . . . ®gS !
5 0-0 d7-d5
and after 1 4 g3 .EI fc8 he can look
6 d 2-d4 � f8-d6
forward to a secure future .
7 4:l b 1 -c3 0-0
White can of course capture on c4
8 b2-b3 4:l b 8-d7
and, in his turn, threaten 1 4 � a3 .
9 � c l -b2 ® d8-e7
But in this case Black c a n again play
Up to this point Blackburne has 1 3 . . . ®gS with sharp and not dis­
been playing faultlessly, but here he advantageous play.
commits a strategical mistake. He After the stereotyped move with the
permits his opponent to trade off knight, events are determined by
the black-squ are bishop, which has strategic considerations. It is not
an important role to play in this easy for Black to come u p with any­
line. One us ually continues with thing effective to counter White's
9 . . . a6 or 9 . . . c S , with prospects well-devised and imaginative king­
of an easy game for Black. Even the side attack.
exchange of the bishop is tolerable ,
1 3 f2-f3 4:l e4 x d 2
provided that Black finds a plan
1 4 tfrt d 1 X d2 dS X c4
suited to the situation.
Black improves h i s pawn position
10 4:1 c3-b5 4:l f6-e4
and opens the c-file at the same
1 1 4:J bS X d6 c7 x d6
time . He tries to initiate counter­
Blackburne takes with the pawn on play operations on the queen's side.
d6 and tries to get counterplay
1 5 � e2 x c4 d6-d5
along the open c-file.
16 � c4-d3 .EI f8-c8
1 2 4:1 f3-d2
B lack assumes that he cannot bring
Zukertort follows a strategical plan home the e3-e4 thrust an yw ay and
typical of such positions. He has in d oubles the rooks on the open c­
mind f2 -f3 and then e3-e4 with a fi le, hoping for an opportunity to
view to staging an offensive on the throw them into the battle. But they
K-side with the aid of the pawn cen­ only shoot holes in the air, while
tre and his two bishops . White's attack is becoming danger­
ous. Probably 16 . . . aS was a more
12 . . .
promising defence plan, especially
This obvious reply deserves cen­ i n view of the fact that 1 7 . . . � a6
sure. Black should be exploring all threatens to trade off the bishop
possibilities of rendering White's and makes 17 . . . a4 possible. To
plan as difficult as possible. He 1 7 a4 Black could then answer

269
1 7 . . . .EI fc8 with the threats of Relying on the force of his doubled
1 8 . . . . ®b4 and 1 8 . . . � a6. In rooks, Black apparently believed he
this case White could indeed have would be able to withstand White's
had to reckon with Black's counter­ pressure by playing the knight to e4.
play.
22 eS X f6 � e8 X f6
1 7 .Ei a 1 -e 1 .EI c8-c7
If Black h ad captured on f6 with his
1 8 e3-e4 .EI a8-c8
queen White would have com­
19 e4-e5 � f6-e8
p letely wrested the initiative from
20 f3-f4
his opponent by 23 ® e t f) g7 24
White's offensive is developing .EI eS, but he would still have had to
without interference. 2 1 fS is al­ demolish his adversary's b astions.
ready being threatened, thereby But Blackburne was planning, by
opening files on the king-side . tactical means, not only to stem the
Black can prevent this by playing hostile attack but to take over the
20 . . . fS himself and replying to 2 1 reins. White frustrates this plan in a
e f by 2 1 . . . � x f6, thereby threat­ brilliant fashion.
ening to post the knight on e4. The
2 3 f4-f5 ! !
best reply for White is to play 22
.EI e5 in order to counter 22 . . . This does credit to Z ukertort's tac­
� e4 with 23 ®e2, repeating the tical perspicacity. To decide to
threat of f4-f5 . Black must defend make this move he must h ave calcu­
himself by 2 3 . . . ® f7 , and if 24 lated its consequences far in ad­
®g4 then 24 . . .EI c2 ! 25 � x c2
. vance and clearly visu alized the
.EI x c2 26 � c l � c 8 . Although course of future events . That is why
White wins an exchange, the game this move deserves two exclamation
now proceeds along tactical lines m arks . This is the beginning of a
and thanks to his strong forward­ forced development.
based knight on e4 Black now has
23 . . . � f6-e4
prospects of cou nterplay.
24 �d3 x e4 dS x e4
20 . . . g7-g6
2 1 .Ei e l -e3 f7-f5 528

Compared with the variation ex­

amined above, the line emb arked
upon by Black is decidedly inferior.
White made a useful move by play­
ing the rook to e3, whilst Black
weakened his king's position. For
this reason 21 . . . � g7 would have
been more consistent, and only af­
ter 22 g4 to continue by 22 . . . fS
2 3 ef ® x f6.

270
This position had apparently ap­
pealed to Black as wel l . The threat
of 25 . . . El c2 apparently gives
White no time to capture on g6.
And yet !
2s fs x g6 ! .bi c7-c2
26 g6 X h7 + dlg8-h8
27 d4-d5 + e6-e5
529
28 ® d2-b4 ! !
This had escaped Blackburne's no­
tice ! But such stunning moves can
be overlooked occasionally.
White's queen, which is being frivo­
lously sacrificed here, is actually
unassailable because after 28 . . .
® X b4 29 � X e S + d/ X h7 30
El h3 + dlg6 3 1 E! g3 + White's
second rook enters the arena and
White, with his superior forces, eas­
ily mates his opponent's king.
What is Black to do now? If he
plays his queen to e8 then 29 ® d6
decides immediately, for White
threatens to take on e5 or to play 30
®f6 + cj) X h7 3 1 Ej h3 + .
If the rook on c2 is interposed on c5
then 29 ® x e4 follows with an un­
avoidable strike at e S . There is no
alternative to playing the other
rook to c5.
28 . . . .bi c8-c5
29 E! f1 -f8 + !
White does not relent. Incidentally
White can also change the sequence
by playing first 29 ® x e4 .bi x b2
and then 30 E! f8 + ® X f8 3 1
® x eS + ®g7 3 2 ®e8 + dl x h7
3 3 Et h3 + ®h6 34 ®f7 + with
mate on next move. The fact that
there are two lines of play leading
to victory in no way detracts from
the value of the combination, it only
serves to stress the impetus of the at­
tack.
29 . . . d/h8 X h7
No better is 29 . . . ® x f8 30
� X eS + d/ X h7 31 ® X e4 + with
a quick mate .
30 ® b 4 x e4 + dlh7-g7
3 1 � b2 X eS + !
The penultimate sacrifice ; but 3 1
E! f5 was also possible.
31 . . . d/g7 X f8
3 2 � e 5 -g7 + !
The final chord ! Black resigned be­
cause he is mated after 32 . . .
\i) X g7 33 ® X e7 + .
In this game the combination
evolved as a result of the fact that
B lack tried to take over the initia­
tive by force in a situ ation that did
not provide sufficient j ustificati on
for such a course. In the decisive
moment both of White's rooks and
a bishop were levelled at the hostile
king. The shattered position of the
black king was an equally important
factor for the successful outcome of
the combination, giving added
force to the attack.
27 1
Bird's Opening White's bishops are clearly directed
Lasker-Bauer at the opponent's king-side and the
Amsterdam 1 88 9 threat of 1 4 � hS is in the offing.
l f2-f4 d7-d5 The simplest way of refuting it
2 e2-e3 � g 8-f6 would be 13 . . . � e4 in order to be
3 b2-b3 e7-e6 able to reply to 14 � hS by 14 . . . f6 .
4 � c l -b2 � f8 -e7 This advance with the pawn would
5 � fl -d3 also have been made in the case of
14 ®g4. Why did Black disregard
The bishop stands insecurely on this
the threat? Perhaps he was planning
square, justifying the critique lev­
to play 1 3 . . . B fc8 in order to
elled at Lasker by his contempora­
stage a breakthrough of his own by
ries for this text move. They were of
c5-c4, but then it d awned on him
the opinion that Black could ex­
that he loses his queen after 1 4
change the bishop to his advantage
� b S . At any rate this may be the ex­
later on in the game.
planation for his move.
5 ... b7-b6
6 � b 1 -c3 �c8-b7 1 4 � g3-h5 ! � f6 x hs
7 � g 1 -f3 � b 8-d7
8 0-0 0-0 530

9 � c3-e2 c7-c5
It would naturally have been more
sensible to play 9 . . . � c5 1 0 � eS
� fd7 ! After an exchange of pieces
in the centre Black would have had
no worries about his future. But
now the bishop d 3 is safe and Wh ite
is getting set for an assault on the
king-side in which this piece will
play an important part.
1 0 � e2-g3 ®d 8-c7 Black relies on the automatic con­
tinuation 1 5 ® x hS fs after which
Again Black fails to take advantage his position becomes quite tenable
of the opportunity to stabilize the because the bishop d3 cannot parti­
situation in the centre by 1 0 . . . cipate in an attack on the king's
� e4 and to nip White's aggressive side.
intentions in the bud. His plan is to But let us leave the game for a mo­
transpose his queen to c6 and ment and take a closer look at the
threaten m ate on g2 at an oppor­ position. The bishops are set to
tune moment. strike at the king's fortress, which
1 1 � f3-e5 � d7 X e5 they can demolish at a moment's
12 �b2 X eS ®c7-c6 notice because there is not a single
13 ® cl 1 -e2 a7-a6 Black piece defending it. To make

272
matters worse the bishop e7 and
rook f8 hem in their own king mak­
ing it impossible for it to flee if the
need arises .
Consequently it is not diffic ult to
see that after sacrificing his two
bishops the least White can do is
give perpetual check. But is this all
he can d o ? It is not : he can easily
mobilize the rook via f3 in an attack
on the king's side. This means that
all conditions are satisfied for
launching a combinative assault.
Another questio n : was there no way
for Black to put up a stouter de­
fence against 14 � h5 ? Granted ,
1 4 . . . � e 8 would have failed to
stop the bishop sacrifice at g7 be­
cause of 1 5 . . . � x g7 16 ®g4.
But an experienced player on the
defensive would have chosen the
lesser of the two evils and played
1 4 . . . E! fc8 instead of 14 . . .
� x h 5 and surrendered a pawn by
1 5 � X f6 + � X f6 1 6 � X f6 gf 1 7
®g4 + d:>f8 for the opportunity of
evacu ating his king and offering
tough resistance.
1 5 � d3 x h7 + ! d/g8 X h7
1 6 ®e2 X h 5 + <ilh7-g8
1 7 �e5 X g7 ! d/g8 X g7
Black has to accept the second sac­
rifice as well , for if 1 7 . . . f6 then
the easiest way to the objective is 1 8
E! f3 ®e8 1 9 ® h8 + <ilf7 2 0 ® h 7.
18 ® h5-g4 + dl g7-h7
1 9 E! fl -f3 e6-e5
This is the only way for Black to
prevent being mated .
20 E! f3-h3 + ®c6-h6
2 1 E! h3 X h6 + d/h7 X h6
Fo r the time being Black feels am­
ply compensated for the queen with
two bishops and a rook. But with
his next move , which Lasker must
have seen far in advance, White
mou nts the final double attack,
which gives him a decisive superior­
ity in material.
22 ®g4-d7 � e7-f6
23 ®d7 X b7
The rest is not interesti ng. Actu ally
Black could have resigned, but the
game wore on for another fifteen
m oves.
In the meantime, Lasker's combina­
tion based on the double bishop sac­
rifice has become a classic. Here we
see it in conj unction with a double
attack, which strictly speaking was
decisive for the outco me of the
combination. The sacrifice of the
two bishops after this pattern has
since occurred, in various constella­
tions, in several chess tournaments.
Italian Game
Steinitz-Bardeleben
Hastings 1 8 9 5
1 e2-e4 e7-e5
2 � g l -f3 � b 8 -c6
3 � f l -c4 � f8-c5
4 c2-c3 � g8 -f6
5 d2-d4 e5 x d4
6 c3 X d4 �c5-b4 +
7 � b l -c3
This two-edged knight mo�e was
introduced in the early 1 8 th century
by Gioacchino Greco, a native of
Calabria. Trying to outstrip his op­
ponent in developing the pieces ,
White does not mind sacrificing
273
material. The ensuing positions White clearly has the advantage .
have been thoroughly analysed a He has outstripped his opponent in
long time ago. It was found that development, and besides, Black
Black is forced to conduct his de­ has no time to castle.
fence with great precision if he
14 . . . f7-f6
wants to maintain equilibrium.
Was this weakening necessary ? Ap­
7 ... d7-d5 parently yes. If Black tries to make
According to modern theory it is do without it he is in for even bigger
better to capture on e4 with the trouble, e. g. 1 4 . . . ®d6 1 5 ®e2
knight. The move made by Black is c6 16 rl ad t ri d8 17 d5 with the in­
rejected because it puts White at an itiative clearly on White's side. By
advantage. playing 14 . . . f6 Black hopes to
continue with 1 5 . . . 00f7 and
8 e4 X d5 � f6 x d5 16 . . r! he8 and in this way to
.

9 0-0 complete the artificial castling op­

Black has to contend with great dif­ eration.
ficulties now. For instance, after 15 ® d 1 -e2 ®d5-d7
each side has captured on c3, White 1 6 B a t -c l
always has a telling attack after a
sortie with the queen to b3. This move raised doubts among
many commentators . They saw a
9 ... �c8-e6 more precise continuation in 1 6 dS
dif7 1 7 El ad 1 , e. g. 1 7 . . . El he8
After 9 . . . � X c3 10 be 0-0, 1 1 ( 17 . . . � x d5 1 8 � g5 + fg 1 9
� g S looks very dangerous, but the ®f3 + \t>g8 20 El X d5) 1 8 ®c4
move made by Bardeleben is appar­ dif8 1 9 ®b4 b6 20 d6 with strong
ently not a bit better. pressure. But Black improves his si­
tuation somewhat by 1 7 . . . r! hd8
10 � c 1 -g5 �b4-e7
and then by sacrificing the ex­
In worse positions like this it is very change after 1 8 � d4 by 1 8 . . .
difficult to choose the least of many � x d5 1 9 � e6 c6 ( 1 9 . . . ri e 8 20
evils. Perhaps 1 0 . . . ®d7 was not ® hs + ) 20 � x d8 EI x d8. Al­
so unfortunate, e. g. 1 1 � x d5 though White has the advantage, it
� X d5 12 � X d5 ® X d5 13 a3 is no easy job for him to get at
� d6 14 ri e 1 + ®f8 , and Black Black's position.
threatens to castle artificially by 16 . . . c7-c6
1 5 . . . f6 and 1 6 . . . ®f7 .
Why didn't Black reply 1 6 . . . dlf7,
1 1 �c4 X d5 � e6 X d5 after which he could have immedi­
12 � c3 X d5 ®d8 x d5 ately cornered the knight at d5 ? At
13 �g5 X e7 � c6 x e7 any rate, White would have gained
14 ri fl -e l nothing from the exchange sacrifice

274
1 7 ® x e7 + ® x e7 1 8 El x e7 +
d;l x e7 1 9 El x c7 + cbd6 20
El x g7 after 20 . . . El hc8 21 g3
El c7 .
Much later, Romanovski pointed
out another interesting attacki ng
possibility, i . e . the knight sacrifice
1 7 � e 5 + . After 1 7 . . . fe 1 8 de it is
not easy for Black to defend him­
self. For instance if 1 8 . . . ®e6
then 19 El x c7 El hb 8 20 ® f3 +
cbg8 2 1 El x b7 and White gets
three pawns for one piece. Still
stronger apparently is 19 ® f3 +
®fs 20 e6 + 00g6 2 1 ® x b7 El hb 8
22 ® x c 7 , a n d 2 2 . . . El x b2 is
worthless on acco unt of 23 ®g3 +
® gS 24 ® f3 � with a double attack
on f7 and a 8 . The lines examined
show that Black's is not an easy lot
no matter how much effort he puts
into his defence, even if White plays
1 6 l5. ac l .
1 7 d4-d5 �
A very energetic move � Thanks to
this pawn sacrifice White paves the
way for the knight to e6 via d4.
17 . . . c6 X d5
After that the course of events is
forced. Black's prospects would
have hardly been any brighter if he
had tried to simplify matters by
17 . . . dlf7 18 de be 19 ®c4 +
®dS, for after 20 ® x dS cd 2 1
ld. c7 El he8 22 � d4 cbf8 23 � e6 +
dlf7 2 4 � x g7 ! White has won a
pawn, and the rook end-game after
24 . . . 00 X g7 25 )5. C X e7 +
l5. x e7 26 l5. x e7 + cbg6 27 El d7
l5. c8 2 8 g3 l5. c2 29 El x ds l5. x b2
30 El aS gives him good winning
chances. Instead of 20 . . . cd, Black
can give up the pawn i mmediately
by 20 . . . � x d S , but after 2 1
E1 x c6 E1 hd8 2 2 g3 he gets no
compensation for it.
18 � f3-d4 cbe8 -f7
19 � d 4-e6 E1 h8-c8
20 ®e2-g4 g7-g6
21 � e6-g5 + 00f7-e8
531
Bardeleben's last moves were prac­
tically forced, White making threats
and Black warding them off. This
gave rise to a very tense situ ation ,
which offered many tactical possib­
ilities.
22 E1 e 1 x e7 + ! !
The dance begins. To capture the
rook with the queen would be bad
in view of 23 E1 x c8 + l5. x c8 24
® x c8 + . But if it is captured by
the king, then White still retains the
upper hand after 22 . . . dl X e7 23
El e 1 + cbd6 24 ®b4 + 00c7
(24 . . . bl c5 25 El e6 + ) 2 5 � e6 +
00b8 26 ® f4 + .
But the circumstance that the white
rook at cl is without cover enables
Black to make a surprising move,
which he had planned in advance.
275
In this m ::tn n er h e hoped to refute opportunity of concluding his mag­
Wh ite's combi n::ttion. nificent combination. But immedi­
ately following this i ncident, he de­
22 . . . d:>e8 -f8 �
monstrated the finale to the specta­
.-\ worthy reply. "Help yourself tors .
pl c::tse", s::tys the black king sm iling This is how the game could have
i ro n i c::t l ly, "take the queen and get ended :
\' O U rself checkm::tted on the first
25 . . . d/h8-g8
�:-� n k �, But S teinitz had thought one
26 El h7-g7 + d:>g8 -h8
step fu rther.
27 ® g4-h 4 + 00h8 X g7
23 � e7-f7 + �
At long last White's efforts are
Another bri l l i ant move. The rook crowned with success. In the end he
renu i ns invu lnerable. would have forced his u nwilling op­
ponent to accept the sacrifice. And
23 . . . ffif8 -g8
now he attacks Black's king with
24 � f7-g7 + !
death-dealing blows from the
The obtru sive roo k feels at home in queen and the knight.
Black's camp.
28 ® h 4-h7 + d:>g7 -f8
24 . . . d:>g8-h8 29 ® h7-h8 + <1>f8-e7
25 � g7 X h7 + � 30 ® h 8 -g7 + cbe7 -e8
If Black evades to d6 he is mated in
532 three moves . The king is in no hurry
to move to d8 and has to be
prodded a little by White.
31 ®g7-g8 + cbe8 -e7
32 ® g 8 -f7 + d:>e7 -d8
33 ® f7-f8 + ®d7-e8
34 g g5-f7 + <1>d8-d7
35 ® f8-d6 mate.
The final act of Steinitz's magnifi­
cent combination . The entire com­
At this point the g::tme came to an bination, which begins with the at­
::tbrupt end. Realizing that he was tack on e7, consists of fourteen
losing, Bardeleben lost his compo­ moves, not counting the different
s u re and stormed out of the tourna­ possible variants. But Steinitz will
ment hall neve r to return. Of course have hardly calculated them all up
the game was adj udicated as having to the last move. He didn't have to,
been lost by h i m . for it was obvious that the com­
This unfair behaviour on the part of bined attack with the three Wh ite
Bard eleben d eprived Stei nitz of the pieces would lead to victory from

276
the moment White's queen pene­
trated at h7.
But let us retu rn to the position af­
ter Black's 2 1 st move . Only in re­
cent years was it discovered that
there was another line, although a
much more modest one, that would
have given White a win. Instead of
22 El x e7 + ! ! he could have played
22 � x h7 � With this simple combi­
nation based on double attack
White captures a pawn. After
22 . . .Ej X cl 23 � X f6 + ffif7 24
.
to provoke an exchange on c4 in or­
der to capture the pawn on c4 with
the bishop with out losing the
tempo. But this approach is too leis­
urely, especial ly if one considers
that the position can be opened at
any time, with a telling advantage in
development for Black.
10 . . . ® d 8 -e7
A fine move. In response to White's
deliberate play, Rubinstein sacri­
fices a pawn. If the offer is accepted
by 1 1 cd ed 1 2 � x d5 � x d5 1 3

.Ej X c l 'l!iJ X g4 2 5 � X g4 White

wins the end-game in the long run. ® x d S , the position i s opened ,
whereupon Black can conj ure up a
dangerous attack by sacrificing the
Tarrasch Defence knight at b4. For instance 1 3 . . .
Rotlevi-Rubinstein � x b4 � 1 4 ab � x b4 + 1 5 dle2
Lodz 1 907 ( 1 5 � d2 El d8 ) 1 5 . . . � e6 1 6
1 d2-d4 d7-d5 ® d4 fS , and White i s faced with a
2 � g 1 -f3 e7-e6 tedious defence on account of his
3 e2-e3 c7-c5 king's unfortunate posting.
4 c2-c4 � b 8 -c6 1 1 � fl -d3
5 � b1 -c3 � g 8 -f6
W hite concedes that his preceding
6 d4 x c5
move was abortive. Perhaps Rotlevi
This exchange poses no difficulties had originally intended to follow up
to Black. with 1 1 .b! d 1 , but he then realized
Nowadays one usually continues that he achieves nothing after
with 6 a3, to exchange only after 1 1 . . .b!d 8 . 1 1 cd ed 1 2 � e2 was
.

6 . . . � d6 7 de � x c5 8 b4. This worth considering. One gets the im­

leads to the s ame position as in the pression that White should by no
game discussed, except that White means pave the way for the bishop
gains an extra tempo. on c 8 .
6 ... � f8 x c5 11 ... dS X c4
7 a2-a3 a7-a6 12 � d3 x c4 b7-b5
8 b2-b4 � c5-d6 13 � c4-d3 .b! f8-d8
9 � c l -b2 0-0 14 ® d2-e2
10 ® d 1 -d2 Placing the queen opposite a rook
The introduction to a capricious on a file is a risky thing to do, but
plan . White takes his time in bring­ White loses another tempo if he
ing out his bishop to d3. His aim is opts for the text move.

277
I4 . . . �c8-b7
IS 0-0
It is not difficult to see that Black
has overtaken his opponent by two
tempi in development. No wonder
tlut he gets an opportunity to dic­
tate the cou rse of the game.
IS . . . � c6-e5
I 6 � f3 x e5 �d6 X eS
There is a clear difference in the
posting of th e pieces. Already
1 7 . � X h2 + 1 8 cj) X h2
. .
®d6 + is threatened with a double
attack against the king and bishop.
In situations such as this, in which
the oppo nent' s pieces are threaten­
ingly poised against the king's side,
one should exercise utmost caution
in organizing one's defence . One
should not weaken one's position
un less there is an urgent need to do
so.
The correct continuation here was
1 7 E Hd 1 � If 1 7 . . . ®c7 with
threats to h2 and c3 then either 1 8
f4 � x c3 1 9 .El ac l or the supple 1 8
.El ac t �' discou nting the threat of
the thrust at h 2 . What happens if
1 8 . . . � x h2 + is replied to by 1 9
dl h 1 �eS 2 0 � x b S ? In this line
1 9 . . . ®eS fails to 2 0 � e4 , and if
19 . . . ® b 8 then 20 � x h7 +
d;l X h7 2 1 EI X d8 ® X d8 22
dl x h2 is possible, thereby recover­
ing the pawn a and getting approxi­
mately even chances.
In an attempt to put a radical stop
to the development of Black's initia­
tive, White begins to set his centre
pawns in motion. But in so doing
he comes out of the frying pan
278
into the fire, because by opening the
d iagonals for Black's bishops he
dangerously weakens his position.
17 f2-f4 �e5-c7
18 e3-e4 EI a8-c8
19 e4-e5 �c7 -b6 +
20 dlg 1 - h 1 � f6-g4 !
21 � d3-e4
533
White pins all his hopes of a suc­
cessful defence on this move. He
seems to have thought that he could
take the wind out of the sails of his
opponent's offensive by exchanging
the white-squ are bishops .
But even a superficial examination
shows that White has little reason to
have much confidence in his defen­
sive system in the centre, which is
exposed to fire from Black's long­
stepping pieces. The only question
is how it can be blasted into smith­
ereens.
The sober line 2 1 . . . � x h2 22
®hs � x e4 23 � x e4 � x fl ,
with advantage in material, or 22
� x b7 � x fl with the threat of
23 . . . � g3 + could have been
taken. But the approach chosen by
Rubinstein is of cou rse much more
beautiful and impressive.
21 . . . ®e7-h4
2 2 g2-g3
White goes along with the effective
main version of the combination.
Now the overwhelming Black for­
ces close in on the enemy ki ng like a
pack of wolves .
22 . . . bl c8 x c3 ! !
2 3 g3 X h4 bl d 8-d2 ! !
The entire white edifice, whose
bishop e4 is much too weak as a
foundation, collapses like a house
of cards.
A fantastic position ! Four black
pieces are exposed to attack but
none of them can be taken , because
m ate o r loss in material would be
the inevitable consequence.
If for instance 24 � x c3 then
24 . . . � X e4 + 2 5 cl'1 X e4 bj X h 2
mate ; b u t if 2 4 ® x g 4 then 24 . . .
� X e4 + 2 5 rt f3 rt X f3 26 '<!!J g 2
bl fl + 27 bl x f1 � x g2 m ate ; and
finally 24 � x b7 is followed by
24 . . . bl x e2 2 5 � g2 bl h3 ! 26
� x h3 bl X h2 mate. There re­
mains only one chance, and White
takes it.
24 ® e2 x d2 �b7 x e4 +
2 5 ® d2 -g2 rt c3-h3
White gave up, for there is no de­
fence against mate on h2. Let us see
how Black's attack would have de­
veloped if White had def�nded him­
self by 22 h3 ?
There are two lines here that lead to
the goal. The first is a combina­
tional one. Although not as impres­
sive as after 22 g3, the moves are
still sufficiently telling.
22 . . . bl x c3 ! 23 � x c3 . . .
The best reply. White cannot escape
being mated after 23 ® x g4
bj X h3 + ! 24 '<!tJ X h3 "1.!11 X h3 + 25
gh � x e4 + 26 rj;h2 bl d2 + .
2 3 . . . � X e4 24 '<!tJ X g4 . . .
Here, too, 24 ® x e4 ®g3 25 hg
® h4 would have led to mate.
24 . . . '<!tJ X g4 25 hg bl d 3 ! 26 rj;h2
The only defence against 26 . . .
bl h3 mate.
26 . . . bl x c3, and Black is bound
to win in the long run with two
bishops against a rook.
Perhaps the second line is even sim­
pler :
279
22 . . . � x e4 23 � x e4 � e3 ! 24
!'He t (there is nothing else White
can do to prevent Black's rook
penetrating at c2) 2 4 . ® X f4
. .
with advantage in material for
Black.
In this game we saw a combina­
tional discharge, Wh ite havi ng left
his position wide open by advan cing
his pawns.
White i s ahead i n development and
immediately launches active opera­
tions. But this approach is too
straightforward, and a precise de­
fence should effect a compl ete
equalization. A more interesting
game would follow after 9 b3 and
10 �b2.
9 ... �c6 x ds
The natu ral respo nse. 9 . . 0-0
.

would have been followed not by 1 0

Philidor Defence
� x e 7 + ® x e 7 1 1 � gS, because
Adams- Torre
Black's tJrospects are not worse af­
New Orleans 1 920
ter 1 1 . . ® x e4 12 � x f6 gf 1 3
.

1 e2-e4 e7-e5 ® x f6 El ae 8 , but by 1 0 c4, a solid

2 � g 1 -f3 d7-d6 move giving White an advantage in
3 d2-d4 e5 x d4 space.
4 'i.!!J d 1 X d4 10 e4 X d5 0-0
In an attempt quickly to mobilize 1 1 � c l -g5
the pieces, White takes the pawn
with his queen. 536

4 ... � b 8 - c6
5 � f 1 -b5 �c8-d7
6 � bs x c6
Entirely logical. Not to lose any
time by retreating, White exchanges
a bishop for a knight. In this way
the queen does not have to quit its
active posting.
6 ... � d7 x c6
7 � b l -c3 � g8 -f6 This position seems very quiet. No­
8 0-0 thing seems to portend the combin­
A plan involving long castling mer­ ative storm that is to flare up in sev­
its special attention here : 8 � g5 en moves. But if we regard the posi­
� e7 9 0-0-0 . At any rate the b at­ tion from the point of view of our
tle would have been more l ively if notions regarding attacks and
this continuation h ad been chosen. threats we find that two white
pieces are posted more actively than
8 ... � f8 -e7
their Black counterparts . The
9 � c3-d5
bishop has made enemy contact,

280
while the qu een occup ies an effec­
tive position in the centre, has a
wide sweep and is attacking the
pawn at a7 .
There is only one free fi le on which
White might feel inclined to double
his rooks .
And what cou rse should Black
take ? What can he do to co unter his
opponent's obvious intentions ? If
one wanted to simplify the game
one could play 1 1 . . . � d7 1 2
� x e7 ® x e7 1 3 .EHe 1 ® f6 . In
this line it would be useful to play
h7 -h6 in between, i. e. 1 1 . . . h6
12 � h4 � d7 . In view of the u ncer­
tain position of the bishop on h4,
Black might be inclined to make the
tactical move 1 2 . . . � x d 5 . To
recover material equality, White
has to continue with 13 ® x d5
� X h4 1 4 � X h4 'C!tJ X h4 1 5
'f!t1 x b7, but after 1 5 . . . ®c4 !
Black has no problems.
But if Black is disinclined to make
simplifications and prefers to com­
plicate things, he might reply to 1 2
� h4 with 1 2 . . . gS 1 3 � g5 � h5
and then continue with 1 4 . . . � f6,
thereby activating the bishop. Torre
h as a different plan in mind. He
opens up the Q-side file, hoping to
get counterplay there .
11 . . . c7-c6
1 2 c2-c4
Moves such as these one makes a!-
most automatically. But the possib-
ility of playing 1 2 de be 1 3 c4 merits
consideration.
12 . . . c6 X d5
Once again Black could defuse the doubtedly have decided in favo ur of
situ ation if he wanted to by playing
1 2 . . . � X d5 13 cd � X g5 1 4
� X g 5 'C!tJ X g5 1 5 d e be 1 6 'f!t1 X d6
®b5 1 7 b3 .El ad8 .
In the end -game with heavy pieces ,
White can hard ly convert his slight
positional advantage, wh ich con­
sists in the better placing of his
pawns .
1 3 c4 X d5 .El f8 -e8
No wonder that this move is the
root of Black's demise. White in­
tends to double th e rooks on the e­
file anyway, which means that even­
tually the rook at e8 can be exposed
to the threat of a double attack. The
correct reply was 1 3 . . . ®d7 ! 1 4
rl fe l a6 1 5 Ei e2 .Ei ae8 1 6 .El ae 1
� d 8 in order to relieve the pressure
on the e-file.
14 .El f l - e l a7 -a5
1 5 .El e 1 -e2 rl a8 -c8
Without suspecting anything bad
Black forfeits the opportu nity of
creating an "air pocket" for himself.
By exchanging the pair of rooks by
playing 1 5 . . . h6 1 6 � d2 � f8 he
could have perceptibly improved his
defence. But now the white pieces
h ave created a vigorous two-fold
active contact, threatening to attack
the rook e8 , which, along with the
missing air pocket, triggers the deci­
sive mating attack.
16 .El a 1 -e 1 ® d 8 -d7
1 7 � g5 x f6 .,.�.. el x fb
�..
If Black had foreseen the endless
u npleasantnesses which he has to
contend with now he would un­
28 1
1 7 . . . gf. In this case an attempt to i t i s exposed t o a double attack by
launch a double attack by 1 8 the two white rooks.
l:1 x e 7 r1 x e 7 1 9 r1 x e 7 'f!tJ x e7
18 �d7-b5 �
20 � g4 + followed by 21 ® x c8
0 0 0

would h ave failed dismally to
1 8 . . . ® X e7 ! Of course we have to
take into account that 17 . . . gf
creates gaping weaknesses on the
K-side, enabling White to capitalize
on them by 1 8 g4 and the threats 1 9
g h4 and 2 0 g fs .
537
But Black still has some ammuni­
tion left. While evading the obtru­
siveness, the queen threatens to take
on e2.
19 � g4-c4 !
A magnificent reply, which is also
based on the double attack. The
White queen is itself unassailable,
but it threatens its opposite number,
forcing it to retreat.
19 0 0 0 ®b5-d7
20 � c4-c7 ! �

538

1 8 ® d4-g4 !
The irony of fate ! Up till now Black
could have opened the "valve" by
h7 -h6 several times, but he never
took advantage of this opportunity.
And now fate takes its terrible re­
The white queen is becoming a
venge, for the absence of the incon­
nuisance. It seems to feel quite at
spicuous but fatal air pocket enables
home in the enemy camp. And all
White to carry out a splendid com­
this is the aftermath of the mating
bination. Black's queen covers the
threat at e 8 .
rook e 8 , but White tries with a text­
book move to distract it from this 20 0 0 0 ®d7-b5
task. At the same time he stages a 21 a2-a4 !
double attack. One cannot say that
This sacrifice of an inconspicuous
this coming to a head of the situa­
pawn is no less telling than the pre­
tion came as a surprise. You will re­
ceding queen sacrifice offers. In an
call that the rook stepped on the fa­
involved combination all things are
tal square e8 five moves earlier, and
of equal importance : the sacrifices,
now, as could have been expected,

282
the attacks and even the simplest Irregular Opening
piece moves . White's scheme can­ Reti-Alekhine
not work without this double step, Baden-Baden 1 92 5
for 2 1 ® x b7 would have been
1 g2-g3 e7-e5
harshly refuted by 2 1 . . . ® x e2 !
2 � g 1 -f3
21 . . . ®bs x a4 An experiment. White intends to
22 � e2-e4 ! ® a4-b5 play the Alekhine Defence and win
2 3 ® c7 x b7 ! a tempo by g2-g 3 .
A triumphant attack ! The queen 2 ... e5-e4
falls on the battlefield. Black re­ 3 � f3 -d4 d7-d5
signed. Black undertakes nothing to refute
What actually gave rise to the com­ the opponent's plan. But 3 . . . cS 4
bination in this game ? First, to se­ � b 3 c4 5 � d4 � c S 6 c3 � c6
cure the open file, White doubled would have deserved consideration
his rooks on it. It was this battery in as a means of winning time for de­
the end that determined the out­ velopment.
come of the game. Secondly, the ac­
tive posting of White's queen also 4 d2-d3 e4 x d3
played a role and enabled White to 5 ®d1 x d3 � g8 -f6
offer several times to sacrifice it. 6 � fl -g2 � f8 -b4 +
The fact that the principal pieces in­ For the time being Alekhine does
volved in the combination, namely not get involved in the struggle for
the queen and the two rooks, were the centre, but tries to mobilize his
effectively placed is unquestionably forces as quickly as possible.
an important factor. But this alone 7 � c l -d2 �b4 X d2 +
was not decisive . No less significant 8 � b 1 X d2 0-0
in carrying out the combination 9 c2-c4
were the three weaknesses in
Black's position. First, there was the The opponents are pursuing differ­
rook e8 that was exposed to double ent strategies. Alekhine castled
attack ; secondly, the unfortunate early, while Reti is in no hurry to do
posting of Black's king, i. e. the so. Instead, he increases pressure in
missing air pocket, on account of the centre and wants to force open
which there was a mating threat on the diagonal of the bishop g2. Ap­
the 8 th rank thirdly the important parently he intended to reply to
fact that Black's queen was not 9 . . . cS with 1 0 � 4b 3, after which
covered . If the pawn had stood not 1 1 � x cS and 1 1 cd are threatened.
on aS, but on a6, Black could have 9 ... � b8 -a6
parried 1 9 ®c4 with 19 . . . E1 ed 8 10 c4 x dS � a6-b4
or 1 9 . . . E1 f8 . 1 1 ®d 3-c4 � b4 x d s
12 � d2-b3 c7-c6
1 3 0-0 E1 f8 -e8

283
The initiative in this position is ob­ bishops. But Black now has the pos­
viously in White's hands . In my s ibility of getting counter ch ances
opinion he would have done better on the K-s ide .
to concentrate his efforts on playing
20 h7-h5 !
e2-e4. For this reason 1 4 E He l de­
0 0 0

served serious consideration. In this The second player hopes to weaken

case Black would have been forced the position of his opponent's king
to seek a solution to the problem of by advancing his pawn.
d eveloping the bishop c8 . But Reti
21 b2-b4 a7-a6
opts for another plan : he tries to put
22 Et a 1 -c l h5-h4
pressure on the Q-side and does not
23 a2-a4 h4 X g3
touch the e-pawn.
24 h2 X g3 ®d8-c7
1 4 Et f 1 -d 1 � c8-g4 !
A weak spot g3 has occurred in the
1 5 bl d 1 -d2 ®d8-c8
white camp , and Alekhine is willing
16 � b3-c5 � g4-h3 !
to take advantage of it. White co uld
In this way Black succeeds in rid ­ p arry the immediate threat by 25 e4
ding himself of the problem of de­ � b6 26 ® b 3 , but this would mean
veloping the bishop c8 and in achiev­ closing the b ishop's diagonal.
ing complete balance. Black's idea
2 5 b4-b5
is based on the consideration that
White may not attempt to win a Reti doggedly sticks to his plan, but
pawn by 1 7 � X h3 ® X h3 1 8 is taken by surprise by an effective
� x b7, because if he does, then tactical stroke.
1 8 . . . � g4 1 9 � f3 � de3 ! 20 fe
25 a6 X b5
� X e 3 and 2 1 ® x f7 + cannot
0 0 0

26 a4 X b5
save White on account of 21 . . .
d/ h 8 22 � h4 Et f8 , after which
mate is threatened on f l . 539

1 7 � g2-f3 �h3-g4
1 8 � f3-g2
White did not want to exchange his
strong b ishop, because he pinned all
his hopes on it. But he could h ave
had a good game by playing 1 8
� X g4 � X g4 1 9 e4.
18 0 0 0 � g4-h3
19 � g2-f3 � h3-g4
20 � f3-h 1 26 0 0 0 Et e8 -e3 !
In the end White decided to avoid White seems t o h ave underesti­
move repetition and exchanging mated the strength of this move.

284
The rook may not be taken on ac­
count of 27 . . . 'd:ti x g3 + 28 � g2
� x e3, but how can Black protect
himself against 27 . . . Ei x g3 + ?
27 � d4-f3
Alekhine rightly said that the only
defence was 27 � f3 ! � x f3 2 8 ef!
But he was of the opinion that Black
would have the advantage after
28 . . . cb 29 � x b5 ® aS, assum­
ing that 30 Ei x d5 is rendered The most obstinate line. Although
worthless after 30 . . . El e 1 + 3 1 White saves the pawn at g3 by 3 1
El x e 1 ® x e 1 + 3 2 cbg2 El a l . \t>f1 � X g 3 + 3 2 fg � X f3 33
This assessment has to be corrected. � X f3 El X f3 + 3 4 \t>g2 Ei aa3 3S
If we continue this line, we find that .Ei d 8 + d:lh7 36 Ei h l + d:>g6 37
Black's attack is refuted after 33 El h3, he is faced with insurmoun­
El d8 + d:>h7 34 'd:ti h4 + d:>g6 35 f4. table difficulties after 37 . . . El fb� !
Obviously 32 . . . � X dS 33
® x dS and only then Ei a8-a 1 31 . . . � f6-e4 !
would have been more precise, but The second player's tenacity is ad�
Black can then give perpetual mirable. He introduces a new corn,
check. bination, capitalizing on White's
27 . . . c6 x b5 disjointed pieces. Despite the fact
2 8 ® c4 x bs � d 5-c3 ! that the rook at e3 has been under
29 ® b s x b7 fire for five moves it remains unas­
sailable, because if White plays 32
If 29 ® c4 then Alekhine could have fe Black replies with 32 . . . � x d<
played 29 . . . b5. and White loses the exchange.
29 . . . ® c7 x b7 32 El c 1 -c4 !
Black is not afraid to exchange A brilliant reply ! If Black played
queens. Worse was 29 . . . 32 . . . � x d2, White planned to
� x e2 + 30 El x e2 ® x b7 be­ counter with 33 � X d2 !, and if
cause of 3 1 El x e3 ! with some hope 32 . . . � x f3 then 33 El x e4 !
of salvation. Alekhine thought this to be the best
30 � c5 X b7 � c3 x e2+ defence, but the obvious weakness
31 d:> g 1 -h2 of this line of play is that eventually
the rook at c4 would hang. I am of
the opinion that 32 El d8 + El x d8
33 fe would have given White bettet
chances of getting a draw. For in
this case 33 . . . � x cl 34 � x d8

28S
would be of little use. Black has to 4 � f3 X d4 � g8 -f6
continue with 33 . . . rl b8 34 r1 c7 5 � b 1 -c3 d7-d6
� 2 x g3, winning a pawn. This ad­ 6 g2-g3 � b 8-c6
vantage is by no means easy to con­ 7 � f l -g2 � c8-d7
vert on account of the loss of 8 0-0 a7-a6
pawns .
After the text move, Black also wins The two-edged line adopted by
a pawn, and the rooks remain on Black is characteristic of many
the board, which is of great import­ opening variations of the Sicilian
ance for the further course of the Defence. The second player does
game. not castle for the time being, initiat­
ing active operations on the Q-side
32 . . . � e4 x f2 in the hope of achieving certain ad­
33 � h 1 -g2 � g4-e6 ! vantages there. Such an approach
This is the core of the matter. Black usually leads to complex, tactically
is attacking the rook and at the sharp positions, in which the Black
same time threatens check with the king can suddenly come u nder en­
knight at g4. emy fire if it is not removed from
the centre in time. This is why
3 4 rl c4-c2 � f2 -g4 + White is forced to take vigorous
3 5 dlh2-h3 steps if he does not want to risk los­
There is no alternative , for if 3 5 ing his superiority.
dl h 1 then. } 5 . r1 a 1 + decides
. .

9 � c l -e3 bl a8-c8
the issue. There follows an elegant 10 ® d 1 -e2 b7-b5
finish. 1 1 a2-a3 � c6-e5
35 . . . � g4-e5 + 12 ll a 1 -d 1 � e5-c4
36 dlh3-h2 bl e3 x f3 !
One could get the impression that
37 ri d2 x e2 � e5-g4 +
Black is in full control of the situa­
3 8 dlh2-h3 � g4-e3 +
tion, especially in view of the fact
39 dlh3-h2 � e3 x c2
that his knight has occupied a key
40 � g2 X f3 � c2-d4
position in the enemy camp and is
White resigned. After 41 bl e3 attacking a pawn and that the posi­
� X f3 + 42 bj, X f3 � d 5 ! the tion of White's knight at c3 is ex­
knight at b7 is lost. tremely s haky. But there is no rea­
son at all to overestimate Black's
prospects. While Black lost time
Sicilian Defence
transferring his knight, White cen­
Ravinski-Panov
tralized his pieces and is now poised
Moscow 1 943
for an attack-Black's king has yet
1 e2-e4 c7-c5 to leave the dangerous area. In
2 � g l -f3 e7-e6 s hort, White, too, h as some trumps
3 d2-d4 c5 X d4 up his sleeve. The situation has

286
reached a critical point and the
scrimmage can begin any min ute.
13 � e3-c1 � c4 :a3
541
Moves such as this, especially when
the king's position is insecure, call
for precise calculation and all at­
tacking and defence resources have
to be scrupu lously weighed. 1 3 . . .
� e 7 was, of course, a more cau­
tious move, but Panov apparently
had full confidence in his position .
1 4 e4-e5 !
White at once goes on the counter­
offensive in the centre and opens
lines for his pieces.
14 . . . d6 X eS
1 5 � d4-c6 ® d 8 -c7
1 6 � c6 x e5 � a3 -c4
The Moor has done his duty, let
him go ! Having devoured a pawn,
the knight goes on to help his king
out.
1 7 � e5 x d7 � f6 X d7
1 8 � c3-d5 ® c7-a7
19 � d5-f4 !
White has obtained a strong attack
along the opened centre files.
Black's position begins to burst at
the seams. A knight sacrifice on e6
is threatened.
19 . . . � c4-e5
Almost as if plagued by bad con­
science, the knight, which has
brou ght about the whole calamity,
is now defending its king recklessly
disregarding all danger.
20 Ei d 1 X d7 !
The hour of reckoning has come .
B y sacrificing a n exchange, White
distracts the knight from the protec­
tion of e6, thereby throwing Black's
defence into disarray.
20 . . . � e S X d7
2 1 � f4 x e6 ! f7 x e6
22 ® e 2 x e6 + � f8 -e7
Very bad is 22 . . . dld8 23 �gS +
dlc7 24 ®c6 + dlb8 2 5 � f4 + ,
and White mates Black's king.
23 E1 fl -e 1 ® a7-c5
Sparing no effort, Black covers his
bishop, at the same time blocking
the white bishop's access to square
gS.
24 b2-b4
White wants to distract the queen
from guarding square gS.
24 . . . � d7-f8 !
The only defence. Black is already
willing to surrender part of his pos­
sessions for the sake of simplifica­
tion, but White of course has no
thought of exchanging queens.
2 5 ® e6-g4 !
28 7
In this manner the threat of 28 ®g4-d 1 + !
� c l -g5 remains and at the same
A precise move ! Black's king is
time the g-pawn finds itself under
forced to flee to the Q-side, because
fire. Black's reply is practically
if 28 . . . cbe6 or 28 . . . dieS , Black
forced, because 25 . 'dtt c7 26
loses by 29 ® d S mate. But here,
. .

� X g7 � g6 27 �g5 � f8 is out of
too, it comes under the cross-fire of
the question on account of the in­
hostile pieces.
evitable mate after 28 �c6 + !
� x c6 2 9 � x e7 + cbd 8 3 0 28 ... cbd6-c7
E! f7 + . 29 � g5 -f4 + cbc7-b6
30 ® d l -d6 + cbb6-a7
25 . . . ®c5-c3
31 ®d 6-e7 +
542 Black resigned on acco unt of
3 1 . . . � c7 3 2 �e3 + cbb8 33
®d8 + EX c8 34 ® b6 mate.

Catalonian Opening
Mikenas-Averbakh
Moscow 1 943
1 d2-d4 � g8 -f6
2 c2-c4 e7-e6
3 g2-g3
26 E! e l x e7 + !
This opening was introduced by
Having eliminated the last of the T artakover at the International
king's guards, Whiu begins chasing Tournament of B arcelona in 1 929,
the king. All of White's forces are hence its name : Catalonian Open­
involved in the pursuit, whilst the ing. White posts his bishop on the
black pieces are relegated to act as diagonal h 1 -a 8 , intending to exert
onlookers. pressure on the Q-side.
26 . . . cbe8 x e7 3 ... d7-d5
27 � c l -gS + cbe7 -d6 4 �f1 -g2 dS X c4
Advancing towards the centre, the Black tries to speed up his develop­
king has not given up hope of sav­ ment by capturing the pawn at c4 .
ing his skin . But if it had chosen to
5 ® d l -a4 + � b 8-d7
move to the side of the board, it
6 fl b 1 -d 2
would have ended in a cui-de-sac,
e. g . 27 . . . cbe8 28 ® e2 + cbf7 29 · Black's plan is to reply to 6 ® x c4
�d5 + cbg6 30 ®e4 + (the last with 6 . . . a6 and 7 . . . b S . But
sacrifice) 30 . . . cb x g5 31 ® f4 + White i nten d s to take the pawn with
cbhs 32 � f7 + � g6 33 ®h4 mate. the kn ight, e. g. 6 . . . a6 7 fl x c4

288
b5 8 cl'Jc2 bl b 8 9 � aS �b4 + 1 0 geous for White to adva nce the e­
�d2 , with better p rospects for pawn.
White. 13 . . . � b6-d 5 �
6 ... c7-c6 Black rem:-tins h ithfu l to his con­
Thanks to the threat of 7 . . . bS, ception of piece p by in the centre .
Bbck forces his opponent to take If 1 4 e4 then 1 4 . . . � b 4 1 5 cl'J e2
the pawn with the queen. c5 1 6 � b 5 �g4 1 7 f3 �e6 can fol ­
low.
7 cl'J a4 x c4 e6-e5
1 4 a2-a3
An attempt to solve the problem in
the centre and the development Consistent play. Wh ite is trying to
problem at the same time. p revent the opponent's pieces fro m
taking up active stations, i. e. he
8 � g 1 -f3 wants to limit their effectiveness.
It would seem that 8 de � :e5 9 But this costs him v:-tluable time,
cl'l c3 would be a more precise move which Black uses to regroup his for­
that would take advantage of the ces to his advantage.
unfavo urable posting of the knight 1 4 . . . cl'J d 8 -b6
eS. But now the posting of the white 1 5 e2-e4 � d5-c7
knight poses problems. 16 � c l -e3
8 ... � d7-b6 If the whi-te h-pawn were on h3,
9 cl'J c4-d3 eS X d4 o n e could s ay with certainty that
10 0-0 � f8-e7 the placing of White's men is better.
It would be contrary to the spirit of But without this restricting move
this game to defend the pawn by and in view of the endangered
10 . . c5 , because this would open
.
knight d4, Black has numerous pos­
the diagonal for the White king's sibilities of cou nterplay.
bishop and slow down Black's de­ 16 . . . � f6-g4
velopment. 1 7 � e3 -d2 a7-a5
1 1 � f3 X d4 0-0 1 8 h2-h3 a5-a4
12 � d2-b3 Ll f8 -e8 543

White intends to advance the pawn

e2-e4. Black wants to counter this
by piece pressure in the centre. Who
will be vindicated ?
1 3 ®d3-c2
The queen has to move away, be­
cause in view of the threatened
1 3 . . . c5 it would be disadvanta- -

289
Each side unwaveringly pursues its 22 . . . ®bs ! (22 . . . �bs 23
own plan : White tries to convert his ® c2 !) 23 ld ac 1 ® x e2 24 ® x cS
opponent's hanging pieces to his � d3 2S ®c2 � x c l . But the attack
advantage and Black tries to do the penetrates .
same.
22 . . . � c7-bS
The position has sharpened per­
23 ®c3-c2 � c8 x fs
ceptibly. White believes he can re­
24 e4 X fS � b5 -d4
fute his opponent's plan with the
following effective reply prepared Black's pieces penetrate White's po­
well in advance. sition in short order. White's q l!een
is sti ll busy covering its ill-placed
19 � d2-aS
knight. In the meantime the first
Black's position seems critical. If for threats cast their shadows on the
instance 19 . . . ab 20 � x b6 be other flank.
then 2 1 � x c7 and White is a pawn
25 ®c2 x a4
up.
There is no better reply.
19 . . . ld a8 x aS !
2S . . . � d4-e2 +
This exchange sacrifice, which the
26 d:>g 1 -h2 � cS x f2
enemy is virtually forced to accept,
promises Black good attacking Black's initiative has developed into
chances in view of the badly placed an attack on the king.
white knight. But for the present
2 7 ® a4-c2
there are no combinations in sight.
They develop somewhat later.
544
20 � b3 x aS � g4-eS
21 ®c2-c3
If 2 1 ® x a4 then 2 1 . . . ® a7 ! , and
the pinning of the knight at aS gives
Black an immediate advantage.
21 . . . �e7-cS
22 �d4-fS
The knight has to withdraw, be­
cause 22 ld ad 1 � x d4 23 bl x d4
� bS is very bad. But this retreat is White would like to defend himself
not the best one. The only correct against ®e3 by playing his queen to
reply was 22 � e2 !, whereby the e4.
knight would have been able to
27 . . . � e5 -g4 + !
guard all accesses to its camp from
the outset. In this case Black would White has to accept this sacrifice. It
have achieved material equality by destroys the defence round the op-

290
ponent' s king and prevents the de­ the forces. And besides, d i d n ' t
fens ive move ®c2-e4 . Lasker once say that first the k u i g h t .,
and then the bishops should he
28 h3 X g4 ®b6-e3
brought out? And yet the b i s h o p d "
The attack triumphs � Bl ack's queen velopment has its positional j u s t i li
has approached the white sovereign cation . The first player tries t o :1 1 1
with obvious intentions : mate i n ticipate the opponent's activ i t i c .�
three · i s threatened . The rest is sim­ and to restrict his freedom of a r
ple and easy to understand. tion. Moreover, the move () � g �'
contains a trap, because 6 . . c .'1 ? 1 �
29 E! fl X f2 ®e3 x g3 +
.

followed by 7 de de 8 ®' x d H
30 <j)h2-h l ®g3-h4 +
E1 x d 8 9 � dS with mate r i a l g a i 1 1 .
3 1 � g2-h3 ®' h4 X f2
B u t after 6 . . . h 6 White i ntc i H i s t t '
White resigned. There is no effec­ recall the bishop t o e3 i n o rd e r t u
tive defence against threatened win a tempo with the d cv e l op • • u· n t
3 1 . . . � g3 mate . Perhaps the most move 6 ®d2 .
interesting aspect of this game is
6 ... c7-c5
that the black knight threatening
7 d4-dS a7-a(,
mate at g3 undertook a truly heroic
8 a2-a4
journey, from b 8 to
d7 -b6-dS -c7 -bS-d4 and thence This move is also part of Wh i t e \
to e2, while his opposite number by plan ; i t is intended t o cou nt n
contrast moved to aS and had to Black's Q-side play.
stay there for the duration of the
8 .. ® d H -a S
game.
.

9 �gS-d2 �
King 's Indian Defence It would not be to White's ad v a 1 1
Averbakh-Pan no tage t o play 9 ®d2 b S !, beca use
Buenos Aires 1 954 Black would get the init i ati v e for
the sacrificed pawn .
1 d2-d4 � g8-f6
The fact that the bishop h a s to t l l l ' l l
2 c2-c4 g7-g6
back does not mean a loss of t e m p o
3 � b l -c3 �f8-g7
,

because this move forces the q u 1T 1 1

4 e2-e4 d7-d6
t o leave the square aS .
S � fl -e2 0-0
6 � c 1 -gS
This was one of the first times that I
used this bishop sortie. Viewed su­
perficially it contradicts the princi­
ples of development of the opening,
because White crosses the demarca­
tion line without h aving completed
the normal p rocess of mobilizing
9 ... e7-c5
Strange as it may sound , t h i s p;t W I I
sally i s a serious and perhaps c v c 1 1
decisive mistake. By blocking t l w
centre Black gives h i s opponc 1 1 1 ; 1
free h and for an offensive o n t l w
K-side. The correct continua t i o 1 1 i \
9 . . . e6 1 0 � f3 ed 1 1 ed � g 4 I )

)'1 1
0-0 with a sl ight advantage for 15 � g4 x c8 ® d 8 x c8
White. For this reason accord ing to 16 � g 1 -f3 � g7-f8
modern theory Black should imme­ 17 d:>e l -e2 !
diately play 8 . . . e6 instead of
8 . . . ®aS. 545

1 0 g2-g4 !
Since the centre is closed, White im­
mediately launches a pawn offen­
sive on the K-side.
10 . . . � f6-e8
1 1 h2-h4 f7-f5
Black tries to organize cou nterplay.
But his plan has the obvious flaw
that the Black king' s position is Impregnable bastions have taken
weakened by it. shape in the centre of the board,
that is why the white king feels saf­
12 h4-h5 f5-f4
est here. White now wants to
Realizing that opening the play on double the heavy pieces on the h­
the K-side ( 1 3 hg hg 1 4 gf was file, which he plans to open at an
threatened) would only be playing opportune moment.
into White's hands, Black tries to
17 ... Ei f7-g7
keep the files closed there.
18 Ei h 1 -h4 � b8-d7
1 3 g4-g5 ! 19 hS X g6 h7 x g6
20 ® d 1 -h 1
Threatening to play hS to h6 in or­
der to exclude the bishop g7 from White has regrouped his pieces and
active play and preparing to ex­ placed them as effectively as possi­
change the white-square bishops in ble. He is poised to launch the deci­
order to open up the white squares sive offensive. Black's king has to
in Black's camp. try to evade the s allies, which are to
be expected from the opponent's
13 . . . E1 f8-f7
pteces.
1 4 �e2-g4 ® a5-d8
20 . . . � f8 -e7
Black sounds the retreat and takes
up an entirely defensive stance. An The threatened king had to be evac­
attempt to become active on the Q­ uated forthwith because from now
side could have had unpleasant con­ on the moves are forced.
sequences, e. g. 1 4 . . . � x g4 1 5
2 1 bl h4-h8 + d:>g8 -f7
® x g4 ®b4 1 6 hg hg 1 7 ®c8 !
22 ® h 1 -h6 � d7-f8
Ei e7 1 8 b3 ! , and 1 8 . . . ® x b3 is
followed by 19 E1 b1 and 2 0 E1 x b7. Black relied on his K-side position

292
being ad equately fortified once he In the end, White's plan triumphs :
has removed his king from squ are the queen penetrates the enemy
g8 and covered the pawn g6 with camp through the weakened white
the knight. But the black knight is squ ares.
far from being as safe as it looks.
26 . . . � d7-f8
The pieces huddled close round it
27 bl h 8 X f8 + !
afford only makeshift protection .
But White still has to blast a breach The only defender of the squ are e6
in the enemy stronghold in order to is eliminated.
penetrate it. The following move is
dlf7 x fs
made in preparation for this breach .
;� .(!; h 3-e6 bl g7-g8
2 3 El a l - h l � 29 � f3 -h4 !
The mating attack is executed with
546 clockwork precision.
29 . . . � e7-d 8
30 � h4 X g6 + d/f8-g7
Or 30 . . . El X g6 31 n h8 + d/g7
32 �g8 mate.
3 1 � g6 x e5
Black resigned. If 3 1 . . . de then
the simplest reply is 32 n h7 + ' and
if 3 2 . . . dl X h7, then 33 ® h6
It is not d ifficult to recognize that m ate. In this game the very re­
Black is powerless against the sacri ­ stricted position of the black king,
fice of the bishop at f4, e. g. 23 . . . which was wedged in between its
� d 8 2 4 � x f4 ef 2 5 El h4 dle7 26 own pieces, resulted in White's
Ij X f4, an d he c a n no longer avoid making a brea�h in his opponent's
telling losses in material. king's stronghold by a sacrifice and
penetrated it with his pieces. This in
23 . . . El a8-b8
the end determined the outcome .
Since there is no way out anyway,
this move is as good as any.
Spanish Opening
24 �d 2 X f4 clYc8 -c7 Averbakh-Furman
Moscow 1 961
Black had to parry the threat of 2 5
� x e5 ! If, e . g. 2 4 . . . � d7 then 1 e2-e4 e7-e5
25 ®h3 � b6 26 � X e 5 ! 2 � g l -f3 � b 8-c6
3 � f1 -b5 a7-a6
2 5 ® h6-h2 � f8 -d7
4 � b5-a4 � g8-f6
2 6 ® h2-h3
5 0-0 � f8-e7

293
 6 Et fl -e 1 b7-b5 � x f6 gf, Black has good piece
7 � a4-b3 d7-d6 play in the centre.
8 c2-c3 0-0
12 . . . d6 X cS
9 h2-h3
My opponent had prepared this
The Spanish Opening has been
move specifically for our encounter,
thoroughly and comprehensively
thereby boldly accepting the chal­
analysed. Some versions extend
lenge.
well into the middle game, others
up to the end-game. For all that, 1 3 d4 X e5 � f6-e8
there is still room for theoretical 14 e 5-e6 f7 x e6
disputes. 15 �b3 x e6 + d/g8-h8
9 ... � c6-b8
547
This knight manoeuvre was used by
the Hungarian master Breyer in the
twenties. Eventually its memory
faded, but thanks to the research
work done by the Soviet masters
Furman and Borisenko, it became
popular again and currently it holds
an established place in modern
tournament practice.
1 0 d2-d4 � b 8-d7
1 1 c3-c4 The critical position. Although only
White takes advantage of the brief a few moves have been made, there
period in which Black diminishes is already a marked shift in forces.
pressure on the centre, and immedi­ White has chances of attacking the
ately mounts active operations king, whilst Black has a strong
there . 12 c5 is threatened, but now­ p awn phalanx on the Q-side. In
adays one u sually plays 1 1 � bd2. such sharp situations usually that
player is successful who is faster in
11 . . . c7-c6
bringing the advantages of his situa­
1 2 c4-c5
tion to bear.
Looks dangerous for Black, be­
1 6 � b 1 -c3 � e 8-c7
cause his pawn grid in the centre is
1 7 � e6-f5
destroyed, for 12 . . . ® c7 is fol­
l owed by 13 cd � x d6 14 � g 5 . White is clearly training his sights
Only subsequent experience with on Black's weakest poi nt, which is
this line of play has shown that h7. If Black should try to exchange
Black can play 1 4 . . . ed without the bishop fS by 1 7 . . . g6 1 8 � g4
any misgivings. Although his king's h5 then White plays 1 9 ®d2 ! with a
position is weakened after 1 5 strong attack. This · explains why

294
Black is in such a hurry to realize
his plan of occupying important
squares in the centre.
17 . . . c5 -c4
1 8 � c l -f4 � c7-e6
1 9 � f4-g3 � e6-c5
In this manner tension in the centre
is maintained, but as the further
course of events will show, this is to
Black's own detriment. Perhaps he
should h ave tried to simplify the sit­
uation. He coul d have done this by
playing the other knight to c5.
20 � f3-d4 ®d8-b6
2 1 e4-e5 l'l f8-d8
Black's p lan looks very logical.
While regrouping his forces, Black
not only increases pressure in the
centre but also intends to dispatch
the knight to f8 in order to protect
his only weak spot at h7. Black
needs only one move to get excel­
lent play. But White's pieces are al­
ready poised to attack the K-side.
Another advantage White has is
that the black forces are so far off­
side. At this point he goes on the of­
fensive .
22 � fs x h7 ! 00h8 X h7
548
The decisive mistake was to accept
the sacrifice. Much stronger was
22 . . . � f8 !, which would have
saved Black from direct attack and
given him every opportunity of ac­
tive counterplay.
23 ®d l -h S + Cbh7-g8
24 � d 4-f5
By winning this tempo, White is
able to enlist the knight in his at­
tack. And a queen acting in concert
with a knight are not a b ad combi­
nation when it comes to weaving a
mating net.
24 . . . � e7-f8
25 ® h5-g6
Moves such as this are normally re­
ferred to as " quiet", but the text
move contains anything but a
"quiet" threat 26 � h6 + 00h8 27
� f7 + 00g8 2 8 � g5 and inevitable
mate . The choice of replies is very
limited .
25 . . . 00g8-h8
26 � c3 -e4 � c5-e6
A desperate attempt to withstand
White's onslaught, for 26 . . .
� x e4 is out of the question on ac­
count of 27 l'l x e4, followed by
l:l h4 + with an irrefutable mate.
Black planned to answer White's 27
® x e6 with 27 . . . � x eS 2 8
® x e S l:l d S in order t o capture the
white queen after 29 ® e 8 � x fS !
30 ® x a8 .Et d 8 3 1 � c7 ® x c7 3 2
® x a6 � c 8 . Of course White does
not h ave to go along with this line
of play, especially in view of the fact
that the s imple excuse 30 ® hs +
00g8 3 1 � c3 would have given him
295
an advantage in material. But his at­ 5 ... c6 X d5
tack would have petered out after 6 � b 1 -c3 �f8 -g7
this. 7 e2-e3 0-0
27 � e4-f6 ! 8 � g 1 -e2 � b 8 -c6
9 0-0
White does not accept the sacrifice,
preferring to offer his own knight in Byrne's opening shows no signs
order to pull down the last fortifica­ o f ambition and, as was to be ex­
tion round the opponent's king. pected, he did not obtain any open­
Now 27 . . . gf fails to 28 rl e4 �g7 ing advantages . Black has no diffi­
29 rl h4 + <j)g8 30 � h6 + <j)h8 3 1 culties whatsoever.
� f7 + and 3 2 rl h 8 m ate. 9 ... b7-b6
27 . . . � d7 x f6 1 0 b2-b3 � c8 - a6
2 8 eS x f6 rl a 8-a7 1 1 � c l -a3 .bt f8 -e8
29 rl e 1 -e4 � e6-f4 12 ® d 1 -d2 e7-e5 !
At this stage Black is already in a The pawn position is fully symmet­
real fix. rical. It can be broken only if one is
30 rl e4 x f4 � c8 X f5 willing to accept weaknesses in
3 1 rl f4 x fs .bt d 8 -d5 one's own camp. Nevertheless
Fischer goes along with the game
A satisfactory defence against 32 opening, because he feels that he
.bt h 5 + <j)gs 33 f7 + .bt x f7 34 will be amply rewarded for the
® h7 mate is not to be found. weakness of his pawn at d5 with the
32 .bt f5 X d5 active posting of his light pieces .
Black resigned in view of 32 . . . cd 1 3 d4 x eS � c6 x e5
3 3 fg + and 34 ® X b6.
549

Grunfeld Indian Defence

R. Byrne-Fischer
New York 1 963
1 d2-d4 � g8 -f6
2 c2-c4 g7-g6
3 g2-g3 c7-c6
Black opts for the GrUnfeld Indian
Defence.
4 � f 1 -g2 d7-d5
5 c4 X d5 1 4 .bt fl -d 1
White wants to have a straightfor­ This seemingly natural move proves
ward game. The tension could have to be a serious error, which un­
been maintained by 5 ®b3. covers the po!nt f2 . White should

296
have played the other rook to d l . In
his book " My 60 Memorable
Games" Fischer writes : "Correct is
14 .El ad 1 ! Originally! had pro­
posed the following 'refutation' :
1 4 . . . � e4 1 5 � x e4 de 1 6
� X e 4 t!!J X d 2 1 7 r:i X d 2 � c4 1 8
� x a8 � x d2 1 9 Ei d 1 � c4 20 be
(the best mov� ) 20 . . . 1d x a8 with
pawn recovery and great end-game
advantage. But Averbakh dis­
covered a hole in my analysis, i. e.
20 � c6 ! (instead of 20 be, which I
had carelessly stated to be best)
20 . . . � x a3 2 1 � x e8 � x e2 22
1d d7, and it is White instead of
Black who wins.
"I spent an evening scrutinizing the
position after 14 r:1 ad 1 . . . The
longer I looked, the more I l iked
White's position � For instance :
1 4 . . . ld c8 ( 1 4 . . . � d3 is refuted
by 1 5 ® c2) 1 5 � x d 5 � x d5 1 6
� x d 5 � d 3 1 7 � g2 Ei c2 1 8
® x c2 � with a win. No better is
1 4 . . . ® d 7 1 5 ®c2, followed by
.El d2 and r:l fd l (if 1 5 . . . r:l c8 ,
then 1 6 ® b t �).
"Another attempt, equally unsatis­
factory, is 1 4 r:1 ad 1 ® c7 1 5 ® e t !
� e4 ! ? (otherwise White consoli­
dates his position by 16 ® b 1 ) 1 6
� X d5 ! t!!J X c l 1 7 � X c l � X fl
18 � x e4 � a6 1 9 � e7 + <ilh8 20
� x a8 r:1 x a8 21 f4, and White
stays one pawn up. Really, how is
White to equalize, to say nothing of
retaining the initiative ?
"At last I found 1 4 . . . ®c8 ! -the
only move to maintain the pressure.
1 5 � x d5 is followed by 1 5 . . .
� X d 5 1 6 � X d 5 Ei d8 1 7 f4
.El X d5 ! 1 8 t!!J X d5 � b7 ! 1 9
® d 8 + ( 1 9 ®d2 ® h3 ! 2 0 � d4
� g4 2 1 r:l fe 1 -or, 2 1 � c2 h5 with
a strong attack-2 1 . . . � x e3 !
should win) 1 9 . . . ® x d8 20
.El X d 8 + Ei X d 8 21 fe � X e5
with a better end-game . And after
1 5 Ei c 1 (after 1 4 Ei ad l
® c8 � -Y . A.) 1 5 . . . ®d 7 ! 1 6
1d cd 1 Ei ad 8 Black wins a valuable
tempo because his queen is on d7 in­
stead of d 8 . The relatively best
move after 14 . . . ®c8 ! is 1 5 �b2
( 1 5 ® c l � e4 1 6 � x d S � x e2 1 7
� x e 4 <ilh8 � wins the exchange . A
possible continuation is 1 8 ® x c8
Ei a x c8 1 9 � e7 r:l c7 20 Ei c t Ei d7
2 1 r:1 fe 1 � f3 �), although Black re­
tains the initiative by 1 5 . . . ®f5 . "
If one weighs t h e pros a n d cons of
what Fischer says above, we find
that the opening initiated with
1 2 . . . e5 ! gives Black a positional
advantage and-the initiative, but
White is able to stand his grou nd .
But the situation changes radically
after the text move, and White's po­
sition immediately becomes critical.
14 . . . � e5-d3 !
1 5 ®d2-c2
In this manner White wards off the
threat of 1 5 . . . � e4 and finds that
297
he has jumped out of the frying pan 19 d/g1 X g2 d5-d4 !
into the fire. D id he have a better 20 � e2 x d4 � a6-b7 +
alternative ? Fischer answers in the 2 1 d/g2-fl
negative and considers the follow­
2 1 dlg1 immediately fails to 2 1 . . .
ing lines to prove his point :
� d4 + 22 ® x d4 .!l e 1 + ! 23 dlf2
" (a) 1 5 � d4 � e4 1 6 � x e4 de 1 7
'{ttJ X d4 + 24 ,!l X d4 ,!l X a1 25
�b2 .!l c8 with strong pressure ;
EI d7 EI c8 26 El x b7 (26 �b2
(b) 1 5 � f4 � e4 1 6 � x e4 de (not
rl h l ) 26 . . . El X c3 27 .!l b8 +
1 6 . . . � x a 1 ·? 1 7 � d6) 1 7 rl ab 1
00g7 2 8 �b2 El x a2.
rl c8 1 8 � x d 3 � c3 ! 1 9 ®e2
But if 2 1 00f2 then the same would
� X d 3 20 @g4 f5 21 tftt h3 � X b 1 !
happen as in the game 21 . . .
22 ,bl X d8 .!l e X d8 23 � fl rl d 1 24
®d7 ! , e. g. 22 EI ac 1 ®h3 23 � f3
d/g2 �d 3 ! 2 5 � x d 3 �d and wins ;
� h6 24 ® d 3 � e3 + 25 ® x e3
(c) 1 5 f3 � h6 1 6 f4 (if 1 6 � f4 ? then
EI x e3 26 00 x e3 EI e8 + 27 00f2
1 6 . . . d4 ! ) 1 6 . . . � g7 ! resu mes
®f5 !
the threat of . . . � e4 . . . " But
These lines of play examined by
White can defend himself more vi­
Fischer prove -that White no longer
gorously. For instance in the case of
has a satisfactory defence.
the faulty move 1 8 � X d 3 White
should first play 1 8 � b4 ; only then 21 . . . ®d8-d7 !
may White play 19 � x d3. In this At this point White decided to cease
case he can offer stiff resistance. resistance, thereby depriving the
15 . . . � d3 X f2 ! spectators of the fun of witnessing
an elegant finish : 22 ®f2 ®h3 +
This knight sacrifice gives Black an
2 3 00 g 1 rl e 1 + ! 24 El x e 1 � x d4.
effective attack, which White ap­
parently had disregarded or over­
Sicilian De.fonce
looked altogether.
Averbakh-Aronin
1 6 d/g 1 X f2 � f6-g4 + Moscow 1 964
1 7 d/f2-g 1 � g4 X e3
1 e2-e4 c7-c5
1 8 ®c2-d2
2 � g 1 -f3 d7-d6
White was hoping for 1 8 . . . 3 d2-d4 c5 X d4
� x d 1 1 9 bi x d 1 , and he would 4 � f3 x d4 � g8 -f6
have no cause for concern. But 5 � b 1 -c3 a7-a6
Fischer is up to something com­ 6 � f l -e2 g7-g6
pletely different.
Usually one continues with 6 . . .
18 . . . � e3 x g2 ! e6, which results in the Scheve­
The bishop is eliminated, enabling ningen Variant; or with 6 . . . eS,
Black quickly to organize an a�tack which results in the Boleslavski Var­
on the long white diagonal takmg iant.
advantage of the clumsy posting of 7 � c l -e3 �f8-g7
the white pieces. 8 ® d 1 -d2 � b 8 -d7

298
Black pursues an original, but two­
edged plan. He delays castling,
choosing to develop his Q-side
pieces in order to exert pressure on
the enemy position from there.
would have taken on bS with his
knight, recovered the pawn and
ended up with a better position.
551

9 � e3 -h6 �g7 x h6
10 ® d2 X h6 b7-bS
It would be shadow-boxing to play
1 0 . . . ® b6 1 1 0-0-0 eS on ac­
C(l)Unt of 1 2 � b3, because 1 2 . . .
,&:1 x f2 is too risky in view of 1 3
� df 1 ! ® b6 ( 1 3 . . . ® X g2 1 4
� hg 1 ) 1 4 � X f6 � X f6 1 S ®g7
and 16 ® x f6.
1 1 0-0-0 � c8 -b7 1 6 � e2 X b S !
12 � h 1 -e l !
This· i s the punch-line ! Black had
White wastes no time covering the reckoned only with 16 � x b S ab 1 7
centre pawn. Its capture would � x dS � x a 2 and intended to re­
open the e-file, which is d angerous ply to 1 8 � X b S with 1 8 . . .
for Black, whose king is stuck in the rl a l + 1 9 \tld2 � x e 1 20 \tl x e 1
centre. ® x c2, because 2 1 � x d7 +
\tl x d7 22 ® x f8 is harmless on ac­
12 . . . ® d 8 -c7
count of the double attack 22 . . .
Prepares for long castling but . . . ® e4 + .
1 3 � c3-dS ! 16 . . . e7-e6
White offers another pawn sacrifice In response to 1 6 . . . ab White had
to open files. planned a difficult combination on
the th�me of double attack : 1 7
13 . . . � f6 X dS
� x bS ® aS (if 1 7 . . . ® c6 1 8
1 4 e4 X dS � b7 x ds
� x d S � b 6 then an effective finish
It would certainly have been more is possible by 19 � x d6 + \tld7 20
understandable to castle, although � c4 + ! !) 18 � x d6 + ! 00d8 1 9
14 � g4 would have followed, ® X f8 + ! � X f8 2 0 � b 7 + \tlc7
which would have been annoying. 2 1 � X a:S r! X aS 22 b4 bj bS 23 a4
Not wishing to lose the initiative, and White is an exchange up on
Black accepts the sacrifice. He h as Black.
confidence in his means of defence . For this reason it is fully under­
standable why Black declines the
1 S ®h6-g7 � h8 -f8
s acrifice. But his reply enables
If Black had castled long, White White to come up with another

299
combin:u ion . I must ad mit that it is have lost immediately because of 2 1
not c::tsy to find ::t satisfactory con­ Et x e 8 + d/ x e 8 2 2 !l e6 + did8
t i n u :uion for Black. 23 -diJg8 + � f8 24 ® X f8 + dJd7
for inst;"tnce, 16 . . . � x g2 would 2 5 .El e7 + cbd6 26 cl'Y f6 + .
be followed by the following at­ 20 ®c7 x d6
t:-tck :
0 0 0

2 1 .b! e6 x d6 00d8-c7
1 7 ®f6 eS 1 8 Et x eS + ! de 1 9
� e6 ! , which would have resu lted inWhite's attack h as resulted in a pec­
uliar position and an unusual rela­
a q u i c k win. Appa rently 1 6 . . . eS !

w:1.s the best defence in these cir­ tion of forces. This position is typi­
cumst:lnces, eve n if White has a cal of modern chess. White has a
strong reply : 1 7 f4 . queen and th ree pawns against
Black's rook, bishop and knight.
1 7 �b5-c6 ! �ds x c6 One could argue about who has the
Castli ng on the Q-side would h ave advantage in material. But the fact
resu lted in a bad position without that Black's kin g has been deprived
the slightest chance of counterplay of its pawn protection and that it
after 1 8 � x dS ed 1 9 !l e7 or even n ow has to be defended by pieces
19 ® x h7. That is why Black stakes gives White the advantage. Al­
everything on one horse and ac­ though to convert it, White still has
cepts the sacrifice. some tough problems to solve, for
Black's pieces can still be mobilized
1 8 � d 4 x e6 f7 x e6 to mount a vigorous attack on the
1 9 ld e l x e6 + 00e 8 -d8 king.
20 Jd d 1 X d6 !
22 !l d6-d2 Et a 8 -e8
23 b2-b3 !l e8-e 1 +
552
24 00c t -b2 ld f8 -e8 !
Black has doubled his rooks and
plans to exchange rooks.
25 ®g7-d4 � d7-b6
It was essential to parry the threat
of 26 ® a7 + .
26 a2-a4 a6-a5
27 ® d 4-c3 !l e t -eS
28 f2-f4 !l e5 -e2
The culmination of Wh ite's combi­
Brilliant defensive play. 29 cl'Y x aS
nation : he threatens 2 1 ® x f8 + ,
would be followed by 29 . . .
but if the king flees to c8 the n 2 1
!j X d2 30 tf!1 X d2 � X a4.
Et x c6 decides the game. Black
finds only one way of prolonging 29 !l d2 x e2 !l e8 x e2
the struggle. 20 . . . Et e8 would 30 ®c3 X aS � c6-e4

300
This seemingly active move actu ally Sicilian Defence
facilitates White's task. But even if Spasski-Petrosyan
Black had chosen the most stubborn Moscow 1 969
line, i. e. 30 . . . � x g2 3 1 ® a7 +
�b7 32 aS � c8 33 ®cS + White 1 e2-e4 c7-c5
would have won by advancing his 2 � g 1 -f3 d7-d6
pawns on the Q-side. 3 d2-d4 c5 x d4
4 � f3 X d4 � g 8 -f6
31 ® a5-e5 + d)c7-b7
5 � b 1 -c3 a7-a6
32 d)b2-a3 !
6 � c l -g5 � b 8-d7
Now the excessive zeal of the b lack
More frequent is 6 . . . e6.
pieces proves baneful to them, for
the bishop is pinned, while the pawn 7 � f l -c4
at c2 is unassailable.
It is held that dev.e loping the bishop
32 . . . � b6-d5 on the a2-g8 diagonal produces
33 a4-a5 � d5-c7 the most lasting effects in reply to
34 a5-a6 + ! the preceding knight move. It has
been demonstrated in practical
The finishing touch. The pawn may
chess that White gets a good game
not be taken, because 34 . . .
by sacrificing his bishop at e6 if
dl x a6 fails to 3 5 &J x c7, whilst
Black plays 7 . . . e6. Black is not
34 . . . � x a6 loses to 3 5 ®b S + .
much better off playing 7 . . . h6 be­
34 . . . 00b7-b6 cause of 8 � x f6 � x f6 9 ® e2 e6
1 0 0-0-0 ® c7 1 1 f4 with a strong
34 . . . d:>c8 loses to 35 a7 00b7 36
attack.
a8 ( ® ) + ! , and if 35 . . . h5 3 6
a8(®) + � X a 8 3 7 ®bS ! b:l e 1 3 8 7 ... ® d 8 -a5
�b4, then Black also loses a piece. 8 ®d 1 -d2 h7-h6
35 ® e5-d4 + 00b6-a5 This move is more meaningful here,
36 ® d4-b4 + d)a5 x a6 because the queen's posting at d2 is
37 ®b4-c4 + � c7-b5 + less effective than at e2.
38 d:> a3 -b4
9 �gS X f6
Black resigned.
A double-edged decision. Trying
In this game I was fortunate to
not to waste much valuable time,
carry out several combinations
White is getting set quickly to mobi­
based on the double attack. In the
lize his forces for a subsequent at­
end a position evolved in which
tack. But he nevertheless exchanges
bringing the advantage to bear be­
his important black-square bishop.
came a tedious matter ! White ma­
naged to do this by combinational 9 ... � d7 X f6
means, throwin g the concentrated 1 0 0-0-0 e7 -e6
force of his queen i nto the balance. 1 1 b:l h 1 -e l

301
 553
White has outstripped his opponent
in the development, although for
the time being there are no obvious
targets for White to pounce at.
Black is faced with the important
decision of where to put his king. If
we take into account the fact that
White's main thrust is aimed at the
king's side, it wou ld have been ad­
visable for Black to opt for castling
on the queen's side. Boleslavski has
proved that Black would have ob­
tained an entirely tenable position
after 1 1 . . . �d7 12 f4 0-0-0,
e. g. 13 fs ( 1 3 e5 � e8 14 f5 d 5 does
not look dangerous for Black)
1 3 . . . dS ! 1 4 fe de ! 1 5 ed +
n X d7, and White fails to turn his
better development to account. If
16 ®e2 then 16 . . . � b4 1 7
® x c4 + .El c7 . White's prospects
are not much better if 1 6 e5 � d5 1 7
� de2 � b4 1 8 ® d4 B hd 8 ! 1 9
® x c4 + d/b 8 . But Black's deci­
sion to castle short gives White a
free hand.
11 . . . �f8 -e7
1 2 f2-f4 0-0
1 3 � c4-b3
Before going over to the attack,
White puts his forces in battle or-
3 02
der. All he has to do is play his king
to b 1 and already his knight threat­
ens to make a sortie to d 5 . Since
Black has no counterplay, he tries in
advance to reinforce the square e6
and to vacate the squ are f8 for the
bishop at the same time.
13 . . . rl f8-e8
1 4 dlc t -b 1 �e7-f8
It is difficult to suggest anything
better. Perhaps he should have
risked 1 4 . . . eS. If White had then
played 1 5 � f5 � x fs 16 ef ef 1 7
® x f4 rl ac8 .
1 5 g2-g4 !
A bold decision ! White sacrifices a
pawn for the sake of opening a file.
15 . . . � f6 X g4
Perhaps Boleslavski was right in
saying that 1 5 . . . e5 16 � fS
� x f5 1 7 gf ef was the lesser of the
two evils.
16 ® d2-g2 � g4-f6
17 r1 e 1 -g 1 � c8-d7
1 8 f4-f5 d/g8 - h 8
White's onslaught becomes irresisti­
ble after the f-file has been opened.
For this reason Black should have
continued either 1 8 . . . ef 1 9 ef b5
20 ® g6 d/h8 21 � x f7 b4 (recom­
mended by Boleslavski) or 1 8 . . .
e 5 (suggested by Tal) .
1 9 B d 1 -f1 ® a5 -d 8
Bondarevski maintains that 1 9
® e 5 i s the only way t o resist the at­
tack.
20 f5 X e6 f7 x e6
If 2 0 . . . � x e6 then 2 1 � x e6 fe 6 r! ft -e l b7-b5
22 � e2 followed by 23 � f4. 7 � a4-b3 d7-d6
8 c2-c3 0-0
554 9 h2-h3 � c6-b 8
Compare this opening with that
used in the game between Averbakh
and Furman.
1 0 d2-d3
The standard reply is of course 1 0
d4. White chose the more cautious
line, without diminishing the ten­
SIOn.
2 1 e4-e5 ! 10 . . . �c8 -b7
By s acrificing another pawn, White 1 1 � b l -d2 � b 8 -d7
is able to throw the knight on c3 1 2 � d2-f1 � d7 -c5
into the final offensive . 1 3 � b3-c2 E! f8 -e8
1 4 � f1 -g3 � e7-f8
21 . . . d6 x e5 1 5 b 2-b4 � c5-d7
22 � c3-e4 ! � f6-h5 16 d3-d4 h7-h6
23 ®g2-g6 ! e5 x d4
Much too passive. In this line,
The re is no defence left. The which is currently very popular, one
tougher reply 23 . . . � f4 would usually plays 16 . . . g6 in order to
have been countered by White with fianchetto the bishop as quickly as
24 EX x f4 ef 2 5 � f3 ! ®b6 26 rl gS ! possible.
�c 6 27 � f6, which would have de­
cided the issue. 1 7 � c l -d2 � d7-b6
1 8 � c2-d3 g7-g6
24 � e4-g5 ! 19 ® d 1 -c2
Black resigned, for 24 . . . hg 25 White regroups his forces : he va­
'f!t1 X h5 + d/g8 2 6 'f!t:t f7 + d/h8 2 7 cates square d 1 for the queen's rook
E! f 3 leads t o mate. and gives the pawn e4 additional
cover at the same time.
Spanish Opening
Karpov-Spasski 19 . . . � f6-d7
Moscow 1 9 73 Black opens the diagonal h8-a 1 in
1 e2 -e4 e7-e5 order to put pressure on d4.
2 � g l -f3 � b 8-c6
20 .El a l -d 1 �f8 -g7
3 �fl -bS a7-a6
2 l d4 x e5 d6 X eS
4 �b5-a4 � g8 -f6
5 0-0 � f8 -e7 The alternative was to capture at eS

303
with the knight. But the text move
undoubtedly looks more logical.
22 c3-c4 !
An unstereotyped and logical move, c5-c4 :
since White could not have streng­
thened his position before opening
the game anyway. The irresolute
continuation 22 � e3 ® e7 23 � d2
would have enabled Black to play
2 3 . . . c5 .
22 . . . bS X c4
In response to 22 . . . cS White had have been followed by 29 . . . � a4
planned to sacrifice a piece : 23 cb
c4 24 � x c4 rl ac8 25 � x f7 +
\tJ X f7 26 '<!tJ b 3 + .
2 3 �d3 x c4 �d8-e7
The root of all evil. It was essential r1 c l with good attacking chances.
to trade off the bishop c4.
24 � c4-b3 ! c7-c5
The punch-line. In view of 26 aS
White need not fear the capture on
b4, and he had worked out the fol­
lowing plan to counter the thrust
25 . . . c5-c4
26 �b3-a2 � b7-c6
27 a4-a5 � c6-a4
2 8 � c2-c1 � b6-c8
Karpov later said that he had in­
tended to counter 2 8 . . . � x d 1
with 29 r1 x d 1 , which could then
30 � X h6 � X h6 3 1 '<!tJ X h6. As
things stand, 3 1 . . . � c3 is worth­
less because of 32 � x c4 with the
threat of 3 3 � x g6 + , whilst
31 . . . � f8 can be cou ntered by 32
29 � d2 x h6 � a4 x d 1
30 rl e l x d l � c8 -d6

555 556

Spasski sets a "trap", which puts

It almost looked as though Black
him in a losing position. His plan is
would wrest the initiative from his
b ased on the following continua­
opponent, because not only 25 . . .
tion : 3 1 � X g7 ffi X g7 32 '<!tJ d2
cb but also 2 5 . . . c4 is threatened .
il ad 8 33 ® x d6? � f8 , and Black
But Karpov had anticipated these
wins. But, White finds a better way
possibilities.
of converting the bla�k knight's
25 a2-a4 ! weak posting.

304
3 1 � h6 X g7 cbg8 X g 7
32 ® c l -gS !
Spasski had failed t o take this into
accou nt in his calculation . Since
trading off the queen leads to an im­
mediate loss of a piece, transferring
the queen to the king's side gives
White a decisive attack.
32 . . . f7-f6
If 32 . . Et ac8 then 3 3 Et x d6
.
keep your attempts to solve the
problems in step with your assimila­
tion of the subj ect m;:uter of this
book. The purpose of these exer­
cises is to reinforce the knowledge
you acquire in studying this work.
For instance, after you have gone
through a chapter you should try
your hand at solving the problems
covered in it.
Once again the problems are ar­
ranged in ascending order of in­

® X gS 34 � X gS � f6 3 5 � e2 c3
creasing difficulty, and it is there­
36 � x f7 and White obtains an ad­
fore advisable to work them out in
vantage in material.
that order. You should first try to
3 3 ®g5-g4 cbg7-h7 solve the problems directly from the
diagram . If you are unc:.ble to so!ve
Otherwise 3 4 Et X d6 followed by
the problem, then you should set up
35 � fS + is inevitable.
the pieces on the chessboard ac­
3 4 � f3-h4 cording to the diagram and try to
find the solution without moving
Black resigned, for there is no de­
the pieces. If the methods proposed
fence, for instance 34 . . Et g8 3 5
here for solving problems do not
.

� X c4 ,Et g 7 36 ,Et X d 6 cl} X d6 37

help you solve the most difficult si­
� hf5 ! or 34 . . . � f8 3 5 � x g6
tuations, then you should try to
� X g6 36 ® h 5 + cbg7 37 El X d6,
work out the move sequences
etc.
through analysis.

Mating combinations I
Exercises
Under this heading we have listed
for home study problems involving combined at­
tacks on the king. Each of the six
You can now check your combina­ positions contains a hidden combi­
tive farsightedness, you r skill in un­ nation, which leads either to mate
ravelling intricate combinative situ­ or-if it is warded off-to a decisive
ations and your ability to spot hid­ advantage in material.
den combinative possibilities. All
exercises are arranged in the order
of the classification we have pro­
posed, which will undoubtedly
make it easier for you to seek out
the combinations. We advise you to

305
Exercise 1 Exercise 4

557 560

W hite to play White to p l ay

Exercise 2 Exercise 5

558 561

Black to play. Can he p l ay 1 . . . .bi c3 ? White to play. Mate in four.

Exercise 3 Exercise 6

�9 �2

Black to p lay White to play

306
Mating combinations If Exercise 9

In this section are listed those cases 565

of double attack in which a mating

threat is accompanied by an attack
or another threat.

Exercise 7

563

White to play

Exercise 1 0

566

Black to play

Exercise 8

564

White to play

Exercise 1 1
567

White to play

White to play

307
Exercise 1 2 Exercise 1 4

568 570

White to play White to play

Combinations on the theme ofwin­ Exercise 1 5

ning material I
571

This group of problems deals with

cases in which the attack on the
king is accompanied by an attack
on another piece.

Exercise 1 3

569

White to play

Exercise 1 6

572

White to play

White to play

308
Exercise 1 7 Exercise 1 9

573 575

White to play White to p lay

Exercise 20
Exercise 1 8
576

574

White to play. He played 1 bl e7 . Was

White to play this move correct?

Exercise 2 1
Combinations on the theme ofwin­
577
ning material ll

This section comprises combined

attacks on different pieces.

Black to play

309
Exercise 2 2 Combinations o n the theme
ofwinning material
578 through pawn promotion

Exercise 2 5

581

Black to play

Exercise 23

579
Black to play

Exercise 26

582

Black to play. Can he save himself?

Exercise 24

580

White to play

White to play

310
Exercise 27 Exercise 30

583 586

White to play Black to play. He played 1 . r:l x h 1 .

. .

What i s the best way for White to reply ?

Exercise 2 8 Drawing combinations I

584
Combinations ending in perpetual
check are demonstrated.

Exercise 3 1

587

Black to play

Exercise 29

585

White to play

Black to play

311
Exercise 32 . Exercise 3 5

588 591

Bl ack to play White to play

Exercise 33 Exercise 3 6
589
592

Black to play
White to play

Exercise 34
Drawing combinations I/
590

This group of exercises comprises

stalemating combinations.

White to play

312
Exercise 3 7 Exercise 40

593 5 96

White to play White t o play

Exercise 3 8 Exercise 4 1

594 597

White to play White to play

Exercise 39 Exercise 42

595 598

White t o play White t o play

313
Drawing combinations Ill Exercise 45

This group of combinations is de­ 601

voted to blocking.

Exercise 43

599

White to play

Exercise 46

602

White to play

Exercise 44

600

White to play

Exercise 47

603

White to play

White to play

314
Exercise 4 8 Exercise 50

604 606

White to play White to play

Drawing combinations IV Exercise 5 1

607
Combinations involving perpetual
pursuit are dealt with here .

Exercise 49

605

White to play

Exercise 52

608

White to play

White to play

315
Exercise 53 Exercise 5 5

609 61 1

White to play White to play

Exercise 54 Exercise 56

612

White t o play
White t o play

Exercise 57
Drawing combinations V
613

The combinations presented here

are on the theme of fortification.

White to p l ay

316
Exercise 5 8 Drawing combinations VI

614 These combinations are designed to

achieve a b alance of forces in wh ich
a draw is possible.

Exercise 6 1

White t o play

Exercise 59

615

White to play

Exercise 62

618

White t o play

Exercise 60

616

White to play

White t o play

317
Exercise 63 Exercise 66

619 622

White to play White to play

Exercise 64

620

Solutions

No. 1 1 rl g2 ! , and the threat of 2

® X h7 + <.1> X h7 3 rl h3 mate can­
not be parried (Sultanbeyev-Colle,
Lesh-Antwerp 1 9 26) .
No. 2 1 . rl c3 was followed by 2
. .

de ! ri x d3 (2 . . . ® x a3 3 ® d 8 +
� f8 4 rl x f7 � c6 5 ® e8 ri x e3 6
White to play
e7 il e 1 + 7 � f1 rl x f1 + 8
ri x f1 ® x e7 9 ® x c6 and a win
Exercise 65 for White) 3 ef + <.1>h8 4 �b2 ! !,
and Black resigned (Kusmin­
621
Kotchiyev, Minsk 1 976).
No. 3 1 . . . rl b 1 + ! 2 <.1> x b 1
ri b6 + ! 3 <.1>ct � b2 + 4 <.1>d2
� c3 + ! 5 dlcl il b 1 + 6 <.1> x b 1
®b4 + 7 <.1>c1 ® b 2 mate (Vlk­
Duras, Prague 1 8 99) .
No. 4 1 � d 8 + <.1>a7 2 rl x a6 + ba
3 ® d7 + <.1>b8 4 ® c7 + ctla8 5
®c8 + <.1>a7 6 � b6 + ! <.1> x b6 7
White to play ® b 8 mate (D. L. Ponziani, 1 769) .

318
No. 5 1 Ll b 1 d3 2 � a 1 e5 3 Ll b2 00f7 Et h8 (3 . . n e7 + 4 dJf8
.

dJd4 4 Ll b4 mate or 1 . . . e5 2 dlb8 5 Et d3 and White wins) 4

�d8 d3 3 � b6 dJd4 4 n b4 mate � d 4 ! E! d8 5 Et a3 + dlb8 6 �e5 +
(W. Shinkman, 1 8 72). dieS 7 Et c3 + 00d7 8 .El c7 mate
(E. Dobrescu, 1 97 1 ) .
No. 6 1 ld b7 ! ® e 5 2 �d1 + dJa5 3
b4 + dJa6 4 � e2 + ! ® x e2 5 dJb8 No. 1 2 1 b4 ! ® x b4 2 Et ab 1
®e5 + 6 dJc8 ® e8 + 7 dJc7 ® x c4 3 � e2 ! ®c2 4 � d 3 ! ®c3 5
� x d5 (7 . . . ®e5 + 8 d6 .El b 8 + Et c8 6 ® x c3, and Black
'f!!J X d6 + 9 d) X d6 dJ X b7 1 0 resigned (Hulak-Romanishin,
a 8 ( ® ) + ctJ x a8 1 1 dJc7, and win Moscow 1 977).
for White) 8 a8(®) + ® x a8 9
Ll b6 + dJa7 1 0 b5 � b7 1 1 r1 a6 + ! No. 13 1 � f7 + ! , and Black re­
� x a6 1 2 b6 mate (A. Kasantsev, signed on account of 1 . . . dl x f7 2
1 953). e6 + (Katalymov-Ilivitski, Frunse
1 959) .
No. 7 1 . . . ld x f3 ! 2 Y!!J x g5
n X f2, and Black wins (Stahlberg­ No. 1 4 1 � d7 ! and Black resigned
Alekhine, Hamburg 1 930) . (Fischer-Shocron, Mar del Plata
1 959) .
No. 8 1 � c6 ! ® x c6 2 ® x f8 +
dl x f8 3 r1 e8 mate or 1 . . . ® x a3 N 0. 1 5 1 � X f7 + ! et; X f7 2
2 ld x a3. What happened in the ac­ )d X c7 + Y!!J X c7 3 Y!!J X h7 +
tual game was 1 . . . � c8 2 � x d7 (Mecking-Tan , Petropolis 1 973).
® x d7 3 r1 d3 ®c7 4 tmd6, and N 0 . 16 1 t!tJ X f7 + ! et; X f7 2
Black resigned (Lombardy­ Et X d7 + 'f!!J X d7 3 � X e 5 +
Kramer, New York 1 957). (Matanovic-Sokolov, Yugoslavia
No. 9 1 � f5 ! ! , e . g. 1 . . . ®g5 2 1 95 3 ) .
'f!!J X d8 + 'f!!J X d8 3 rJ c8 Y!!J X c8 4 No. 1 7 1 Ei b7 + ! ! dJ x b7 2
� x g7 + dJg8 5 �d 5 mate or � c8 + ! (Mason-Winawer,
1 . . . Y!!J X g2 + 2 Y!!J X g2 � X g2 3 Vienna 1 8 8 2) .
� x g7 + dJg8 4 Et c7 and White
wins. In the game actually played, 1 No. 1 8 1 Et c7 ! ® x h5 2 Et e7 +
® f2 ? eventually resulted in a draw dJf8 3 r1 x b7 + dle8 4 Et e7 +
(Yudovich-Ragosin, Tbilisi 1 937) . dJf8 5 Et X h7 + 00e8 6 Et X h5
(Antu nac-Htibner, Dresden 1 969) .
No. 1 0 1 � f6 + <l;g7 2 Ll h5 ! ®c7
(2 . . . gh 3 ®g5 + dJh8 4 ®h6 or No. 1 9 1 � d 8 dJd7 2 c 8 ( ® ) + !
2 . <;!; X f6 3 ® g5 + or 2 . . .
. .
dl x c8 3 � f6 (conclusion of a
Y!!J X f6 3 Y!!J h6 + dJg8 4 Y!!J X h7 study by G. Sakhodyakin, 1 939).
mate) 3 ld f1 , and Black resigned
No. 20 1 Et e7 is a weak move ; after
(Sahovic-Diaz, Vranjska Banja
1 . . . dJf8 2 Et x e6 dJf7 White lost
1 976) .
the exchange (Darga-lvkov, Hast­
No. 11 1 Ll c3 ! .El h8 2 dlg7 Et e8 3 i ngs 1 95 5 /56) .

3 19
No. 2 1 1 . . . � d 5 + 2 �d2 ®b6 !
3 ® x a8 + dldl 4 0-0-0 � cl 5
� aS � x a8 6 � x b6 � x b6, and
Black won (Nimzowitsch-Ale­
khine, Bled 1 93 1 ) .
No. 22 1 . . . b4 ! 2 E! x b4 b5 3 hS
�d6 4 rl b3 b4, and after 5 . . .
E! e 8 Black won by playing his king
to c4 (Andersson-Torre, Biel
1 9ll) .
No. 23 Black is unable to save him­
self, e. g. 1 . . . � c2 2 E! x d2
� b4 + 3 d/b t ! dJ X aS 4 E! d6 !
dla4 S dl x b 2 a S 6 rl d4 a 6 l E! d6
(conclusion of a study by W. Bron,
1 916).
No . 24 1 ® c l ! � x d4 + 2 dld3
® x eS 3 ®c8 + dlg7 4 ® h 8 + !
d/ X h8 S � X f7 + dJgl 6 � X e5,
and the black knight is trapped
(Matulovic-Zvetkov, Varna 1 96S).
No. 2 5 1 . . . � c3 2 E! d 1
E! x f 1 + ! 3 E! x f 1 e 3 (Hahn-Tar­ check (S. Kozlovski, 1 9 32) .
rasch, Halle 1 8 8 3) .
No. 2 6 1 ® e 8 + ! Black resigned
(Kondratyev-A. Geller, Leningrad dle4 S bl el + or 4 . . . dlc4 4
1 9 51).
No. 2l 1 � a7 ! �a1 2 d/b 1 � c3 3
dlc2 � a 1 4 �d4 ! � x d4 (4 . . . ed
S dld3) S dld3 � a 1 6 dle4
(P. Heu acker, 1 930) .
No. 2 8 1 . . . b4 ! 2 ab E! x h4 3 gh
g3 4 fg c3 + ! 5 be a3 (Lund-Nim­
zowitsch, Oslo 1 92 1 ) .
N o . 29 1 . . . E! b4 ! ! 2 c b a 4 3 bS +
d; x b5 4 � a3 c3 5 E! b 1 dlc4 6 f4
d; x d4 7 dlf2 dlc4 8 dle t d4, and No. 3 8 1 � e4 + d; any 2 � g2 !
Black won (Kmoch-Nimzowitsch, (conclusion of a study by A. Tro­
Niendorf 1 921).
320
No. 3 0 1 . . r1 x h 1 is bad on ac­
.
count of 2 gf! El x d 1 + 3 � X d 1 ! !
cl:! x d2 4 fg �e6 5 g8(®) + dld7 6
® X c8 + dJ X c8 l � X d2 (Tal­
N. N., s imultaneous game, Stutt­
gart 1 9S 8 ) .
N o . 3 1 1 � x e6 ! cl:! x e2 2
cl:f X g6 + ! fg 3 fl + dJh7 4 ri h3 +
� h6 S � gS + dlgl 6 � e6 + (Os­
m anagic-Gligoric, Sarajevo 1 963) .
No. 32 1 . . . � f6 ! 2 cd rl x bl 3 dc
El x c7 ! 4 cl:! x c7 � g4 ! S hg ® f2 +
6 dlh 1 ® h 4 + (Unzicker-Aver­
b akh, Saltsj obaden 1 972).
No. 3 3 1 . . . El b2 ! 2 cl:! x b2 c4 + 3
dlh 1 ® f2 4 E! g 1 �g4 ! 5 b1 g2
® e t + 6 El g 1 ®f2 7 b1 g2 ® e 1 +
(Tal-Gligoric, Bled 1 96 1 ) .
No. 34 1 El b6 + d/hs 2 ri bS +
dlh4 3 rl b4 + d/h3 4 b1 b3 + d/g2
S dla2 ! h 1 (® ) 6 rl b 1 ® h6 l El b2
and a draw through perpetual
No. 3 S 1 dlc8 + ! dlc2 2 El b7 !
f l (® ) 3 El c7 + dld3 4 rl dl +
b1 cl + dlbs 6 b1 bl + dla6 7
El b6 + dl al 8 b1 bl + dla8 9
El b8 + (H. Lommer, 1 93 3 ) .
No. 3 6 1 El a 1 d 2 2 dlb3 + � a S 3
El x aS + llb6 ! 4 �el llcl S
El cs + d/b8 6 �d6 + lla7 7
El aS + llb6 8 � el ! (E. Dobrescu,
1 97 1 ) .
No. 3 7 1 � e3 ® x e 3 2 ® f2 ® x f2
3 aS + (D. L. Ponziani, 1 1 82).
itzky, 1 924) .
No. 39 1 bl aB bl x a8 2 ® a2 + !
(Enigk-Effel, correspondence
game 1 9S 7) .
N o . 4 0 1 � f2 ! bl e l + 2 d:>d2
bl e2 + 3 d:>c3 bl x f2 4 � cS + !
dl x cS stalemate (H. Rinck, 1 924 ) .
N o . 4 1 1 g 7 bl b2 + 2 d:> f t r1 b t + 3
� d l ! (3 d/g2 � h2 ! 4 d) X h2 r1 b 8
and Black wins) 3 . . . bl x d 1 + 4
dlg2 bl d3 s g8( ® ) bl g3 + 6 d/h 1 !
(G. Nadareishvili, 1 9 S 7) .
N o . 4 2 1 h 6 d/fs 2 h 7 dlg7 3
r1 X f7 + d/ h 8 4 bl f8 + d/ X h7 S
r1 b 8 ! g2 6 r1 b l �e2 + 7 \tlaS !
� f1 8 bl b4 g 1 ( ® ) 9 E1 h4 + dlg6
1 0 bl g4 + (W. Bran, 1 927).
No. 43 1 dlb3 a2 2 cbc2 h3 3 � f3
h2 4 � aB � e2 5 � b7 � g3 6 diet !
h t ( ® ) + 7 � X h l � X h l 8 d:>c2
(Ercole del Rio, 17 50) .
No. 44 1 � c3 � aB ! 2 � a4 � g2 3
� cS � a8 ! 4 � b7 + d/d7 S dl x a8
dlc8 (B. Horwitz, 1 8 8 9) .
No. 45 1 � e7 dS 2 � ec6 de 3 d:>c3
followed by 4 et> x c4 (Ercole del
Rio, 1 750).
No. 46 1 � e5 e t (®) ( 1 . . . de 2
� f3 ! ef stalemate) 2 � g6 + hg 3
� x d5 ® e t 4 cbg2 ! ® d2 + 5
cbh3 ®d3 ! 6 � e4 ! ®d4 7 �d5
® f2 8 � c4 ®gl 9 �dS (J. Hasek,
1 96 5).
No. 47 1 bl f7 ! � d6 2 bl e7 � fs 3
bl f7 � h6 4 bl e7 � g8 S r1 f7 � h6
6 bl e7 � f5 7 ki f7 bl b l + 8 d:>h2
e2 9 � f6 r1 h l + 10 et> x h l
e t ( ®) + 1 1 d:>h2 ® h4 + 1 2 d:>g l
®d4 + 1 3 d:>h2 ®d6 + 1 4 d:>g l
� g7 1 5 dlh t ® b 8 1 6 d:>g l
(G. Kasparyan, 1 959) .
No. 48 1 � g6 + d:>g8 2 � ge7 +
dlh7 3 <1> x e6 ® x b 1 4 d:>f7 or
1 . . . cbe8 2 � d3 ! ® x d3 3
� g7 + dld8 4 � x e6 + 00c8 S
� e7 + ebbS 6 � c6 + cba8 7 � ed8
(G. Nadareishvili, 1 959-6 1 ).
No. 49 1 cbd8 ! (1 d:>d7 � d6 2 d:>c8
� e7 + 3 00d7 � d5 or 2 b6 d/ds 3
00c8 � e7 + 4 d/d7 � c6) 1 . . .
� d6 2 00d7 ! d/ds 3 00c8 � e7 + 4
dld7, and Black is in Zugzwang
(R. Reti, 1 92 8 ) .
N o . S O 1 � e4 ® x h2 2 r1 gs +
dlfs 3 Ei fs + dies 4 r1 h5 ! ®b2 s
El bS ! ®d2 6 bl hS (A. Gurvitch,
1 927) .
No. S 1 1 e7 00f7 2 e 8 (®) + dl x e8
3 dle6 E! f1 4 r1 a4 ! (4 r1 b4 ? r1 b l )
4 . . EI d 1 S EI h 4 (A. Selesniev,
.
1 9 1 9) .
No. S2 1 r1 e8 ! ® x f6 + 2 \t>d7 +
dlfs 3 e4 + cbf4 4 EI e6 ®g7 + 5
E! e7 ®g8 6 E! e8 ® h7 + 7 r1 e7
®g6 8 EI e6 (M. Klyatskin, 1 925) .
No . 53 1 d 8 (®) ! '@ x d8 2 � dS !
® b 8 ( 2 . . . ®e8 3 � cS + OOaS 4
� d2 + OObS 5 � c7 + ) 3 �f4 ®a7
4 � e 3 ® as S � d2 ®d8 6 �gS
(S. Birnov, 1 9 S2) .
N o . 5 4 1 � g 8 00f6 2 h 7 d/g7 3
dlbs ! (3 00b6 ? r1 a8 4 dlbs 00h8 5
hS 00g7 6 h6 + 00h8, and White is
in Zugzwang) 3 . . . EI a8 4 hS � b7
S h6 + 00h8 6 00c6 � d 8 + 7 d/d7
EI b 8 8 dlc7 EI a8 9 d/d7 or 6 . . .
� aS + 7 dlbs EI a7 8 dlb6 (A. Tro­
itzky, 1 9 36).
No. ss 1 r1 b7 h6 2 E! b6 hS 3 r1 bs
h4 4 bl b4 h3 S r1 b l � c2 6 bl b 3 +
dl any 7 r1 x f3 (L. Kubbel, 1 934).
321
No. 56 1 � g 1 \t>g3 2 � c6 ! r:bg2 3
�d4 h2 4 � f6 h 1 (®) s �b2 ® hs
6 � d4 ®as + 7 \t>b 1 ® a4 8 � a 1
( G . Nadareishvili, 1 954).
No. 57 1 �b6 + ! ( 1 � cS ? b3 2
\t>d2 � X g6 3 \t>c 1 � h6 + 4 d/d 1
� c2 + 5 dl any � g7 and Black
wins) 1 . . . dld7 2 � aS b3 3 r:bd2
� x g6 4 d/c1 � h6 + 5 � d2 � g7
6 �e 3 , etc. (L. Kubbel, 1 93 4 ).
N o. 5 8 1 � d4 ! f3 2 g3 ! h3 3 dlb4
h2 4 ct>a3 h1 (®) s dlb2 dlds 6
�e 3 \t>c6 7 � d4 ct>bs 8 � e 3 ® h 8
9 � d4 ® f 8 1 0 �eS dla4 1 1 � d4
®a 3 + 1 2 \t>b 1 ct>as 1 3 � eS dlbs
14 � f6 ( 1 4 �d4 ? \t>a4 15 � f6
®d6 1 6 � d4 dla3 and Black wins)
1 4 . . . \t>a4 1 5 �d4, etc. (V. Che­
khover, 1 9S4).
No. 59 1 h7 \t>e3 2 h8(®) ® h 2 + 3
\t>g4 ! ® X h8 4 � X h8 d/d4 5 � f7
a3 6 � x gS a2 7 dlg3 (or 7 dlh3)
a t ( ® ) 8 \t>h2 ®b2 + 9 �g2 any 1 0
� e4, etc. (V. Petchenkin, 1 9S S) .
No. 6 0 1 f3 a4 2 \t>£2 ! a 3 3 \t>g3 a2
4 d/ X h3 a 1 (@) S d/ X g2 ®b2 + 6
� f2 \t>gs 7 dlg3 ® e t 8 � a7 !
® f4 + 9 d/g2 ®d2 + 1 0 d/g3 ! ,
etc. (V. Chekhover, 1 94 8 ) .
N o . 6 1 1 Ei f3 + ! dlg2 2 Ei b3 !
322
� ac3 + 3 d/c l ! il x a 1 4 .!db2 +
dl any 5 bl a2 ! ! � x a2 (5
&X x a2 stalemate) 6 d/b2
(A. Herbstman, 1 9 34) .
No. 62 1 \t>g4 ! ®c8 + 2 ct>f3 ! !
®b7 + 3 Ei dS ! ® x dS + 4 �e4 or
3 . . . @ X b 1 4 bl aS + d/b7 S
l:j bS + (G . Kasparyan, 1 969) .
No. 63 1 dle2 ! � g2 2 \t>f3 � h4 +
3 dlg3 � g6 4 � bs ! dlds 5 � d6
followed by 6 � x e8 (L. Kubbel,
1 929) .
No. 64 1 � b 5 ! Ei h6 ( 1 . . . Ei e6 2
� x e8 Ei x e8 3 !l bS + and 4
b;l X hS) 2 b;l d2 + ! d/e6 3 � X e 8
� f6 + 4 \t>g 1 ! � x e8 S Ei h2 !
l:j g6 + 6 l:j g2 Ei f6 7 Ei f2, etc.
(L. Kubbel, 1 9 3 1 ) .
No. 65 1 � f6 f2 2 �d5 + r:bb5 !
(2 . . . dlb6 3 � g2 d2 4 � ds + fol­
lowed by 5 � e3) 3 � c4 + ! et> x c4
4 � g4 ! fl ( � ) S � e S + dlc3 6
� x d3 00 X d3 7 00h6 (V. and
M. Platov, 1 90S).
No. 66 1 dle6 � g6 2 dlfs � h4 + 3
c;t>g4 � g2 4 dlf3 � e 1 + 5 d/e2
� c2 6 � b 2 ! � b4 ! 7 �a3 ! � c2 8
�b2 ! 00c8 9 �c3 ! � g3 + 1 0 dld3
� a3 1 1 � b4 � b S 12 dlc4 (G. Kas­
paryan, 1 963-64).
Index of names Biyiasas 99
Bogolyubov 88 1 39 1 49
Bondarevski 50
Botvinnik 4 8 1 4 8 2 509
Boudy 328
Bran 82 1 1 8
This index contains the names of Bronstein 56 444
those players and composers whose Buchman 50 1
games or studies are dealt with in Burn 8 7
the instructional part of the book Byrne, R . 1 29
(for complete games confer Index
of games) . Capablanca 79 1 39 445
The numbers refer to diagram Casper 1 42
numbers. Centurini 1 96
Chekhover 1 8 72 1 3 8 1 44
Afanasyev 433 1 47 1 94 234 S O S
Aisenstadt 1 76 Cheron, A. 1 56
Alatortsev 8 5 9 1 Chistyakov 3 1 4
Alekhine 2 1 7 9 1 05 1 52 1 95 Cook 405
322 327 4 1 6 473 487 Cozio 463
Alster 1 40 Czerniak 67
Amelung 460
Anderssen 2 3 1 Damiano 25 26 27 28 29 30
Antoshin 1 36 31 32 33 34
Aronin 57 Dawson 97
Averbakh 24 50 1 09 1 2 8 1 43 Defosse 5 1 2
3 2 8 478 4 8 0 4 8 4 502 523 Dufresne 4 8 6
5 2 5 526 Duras 1 5 3 456 52 1
Dtirsch 1 62
Bagdasaryan 452
Balashov 99 Eliskases 1 34
Barcza 56 Engels 232
Bat6czky 470 Euwe 2 1 65 92 1 07 322 327
Bauer 5 1 9 Evans 490
Bebtchuk 4 8 4
Beilin 524 Fedorov 320
Belenki 1 55 Fine 4 1 5
Berger 1 26 Fischer 527
Bernstein 1 1 2 Flohr 503
Betak 1 40 Forintos 13 3
Betbeder 202 Friedstein 57
Bething 1 59 Fritz 1 1 1
Birnov 343 4 1 4 440 499 Furman 46

323
Galic 200 Kasparyan 3 6 64 14 7 2 1 3
Gaprindashvili 1 77 344 442 450
Geller 1 09 492 Kazhdan 45
Gel ler, K. 1 1 3 Keres 96
Gendel 7 5 Kikovic 1 3 3
Gofstein 5 2 2 Kipping 446
Gogolev 474 Kling 1 0 1 204
Gollner 5 1 9 Kloza 5 1 5
Golombek 8 0 Koblenz 5 8
Gorgiev 6 0 70 Konig 76
Greco 5 0 8 Konstantinopolski 8 5
Grigoryev 1 5 8 Kopayev 478
Guldin 525 Korolkov 71 20 8 2 1 5 432
Gunsberg 49 Kortchmar 4 1 2 475
Gurvitch 69 78 44 1 45 1 Koskinen 1 6 1
Kotov 3 4 1 2
Halumbirek 4 55 Kretschmar 1 67
Harrwit z 4 8 6 Kubbel, L. 40 95 1 2 3 1 2 4
Hasek 1 63 506 507 1 75 207 465
Havel 209 Kupper 1 3 5
Healey 466 Kupreitchik 23
Heemsoth 1 66
Heissenb utter 1 66 Larsen 5 27
Herbstman 47 439 Lasker 5 1 65 1 30 1 99 4 8 7
Hewitt 5 1 1 Lasker, E d . 1 06
Hoch 5 3 Laue 1 67
Hohler 67 Lawson, T. 97
Holzhausen 4 89 Liburkin 1 7
Horberg 1 2 8 Lilienthal 5 0 3
Horowitz 45 Lisitsin 1 4 8
Huhner 5 Lj ubenski 1 37
Ljubojevic 476
Lolli 394
Ilivitsky 502 Lowenfisch 1 27 1 4 1
Ilyin 495
Isakov 44 Mac Clure 469
Isenegger 1 1 0 Maciewski 2 4
Mackenzie 5 1 3 5 1 7
Kaiyev 66 Madler 6 1
Kakovin 68 1 02 Maiselis 1 90
Kaminer 1 03 1 1 9 Makogonov 1 8 1 5 1
Kan 72 1 27 1 34 4 1 1 Mandler 76
Kasanen 1 6 1 Manov 468

324
Mar eo 4 1 3 Ragozin 5 8
Margolis 1 76 Rasuvayev 2 3
Mar6czy 8 8 1 6 8 202 2 3 2 Reisch 467
Marshall 87 1 06 445 469 Reissman 5 1 6
Martynov 522 Reshevsky 1 07 490
Mason 5 1 3 Reti 94 1 22 1 8 8 4 3 8
Mattison 1 64 1 9 8 Richter, K . 90
Mees 52 Richter, W. 1 42
Menchik 48 1 Rinck, H. 37 42 1 2 1 1 5 4 206
Mikenas 1 5 1 393 443 448 454 457
Miszto 5 1 5 Romanovski 472
Moto·r 494 Romi 1 6 8
Rossolimo 5 1 6
N adareishvili 447 500 Rowner 4 8 3
Nenarokov 1 5 8 Rubinstein 4 1 6
Nestor 1 52 Rudolph 395
Neustadt 59 3 2 1 Ryumin 1 4 1
Nikitin 44
Nikolayevski 392 Saavedra 9 3 427
Novotelnov 4 8 3 Sackmann 1 56 1 97
Nunn 3 2 5 Sairab 3 1 6
O'Kelly 5 1 2 Salkind 1 46
Olafsson 1 3 5 Schmid 1 26
Orinos 470 Sehwers 453
Otten 459 Simagin 8 3 3 1 4 524
Sliva 96
Parr 479 Smeikal 476
Perelman 5 1 0 Smyslov 3 46 1 0 8 1 1 2 1 50
Petrosyan 5 488
Philidor 1 92 Soderborg 477
Pillsbury 340 Sokolski 1 44
Pirc 1 29 Soloviev 480
Pirogov 1 5 5 Stamma 3 5
Platov, W. u nd M. 1 7 1 493 Stein 5 20
Podgayets 1 20 Steinitz 5 1 1
Pogosyants 1 32 1 70 1 74 3 1 8 Stepaniec 1 37
Polugayevsky 4 8 5 5 1 4 Stepanov 4 8 2
Polyak 475 Stolyar 1 43
Popiel 4 1 3 Sultan Khan 1 49
Portisch 520 Surachov 5 0 1
Sushkevitch 75
Rabinovitsh 1 3 8 Szilagyi 5 1 4
Radulov 477
Taimanov 1 69 392 492
Tal 523
Tarrasch 2 2 1 62
Thomas 92 4 1 5
Tolush 1 36 49 1
Torre 5 1
Trifunovic 80
Troitzky 20 4 1 43 1 1 5 1 1 6
1 1 7 1 45 1 60 205 2 1 0 233
3 1 7 3 96 406 423 497
Troyanski 467
Tshigorin 22 49 8 1
Uhlmann 6 1
Varshavski 4 74
Vasyukov 4 8 8
Vecsey 1 1 4
Velimirovic 325
Verlinski 1 OS
Vladimirov 1 20
Vliet 203
Index of games
Adams-Torre 2 8 0
Aronin-Kantorovich 94
Averbakh-Aronin 298
Averbakh-Estrin 1 27
Averbakh-Furman 293
Averbakh-Panno 29 1
Averbakh-Platonov 1 2 8
Averbakh-Taimanov 1 2 1
Byrne, D.-Fischer 1 25
Byrne, R.-Fischer 296
Capablanca-Dus- Khotimirski
1 19
Hon-Zheyandino 98
326
Voorema 1 8 0
Vukcevic 200
Weltmander 4 8 5
Wheatcroft 4 79
White 54
Winz 90
Wotawa 449
Yates 473
Yudovitch 4 1 1
Zagoryansky 8 3 49 1
Zakhodyakin 3 1 9 339 404
4 1 7 498
Zamikhovski 91
Zarytshev 1 00 342
Zeletski 1 73 2 1 1
Zhivayera 1 77
Zhukhovitski 1 69
Zimkhovitch 496 504
Zita 108 444 526
Karpov-Spasski 303
Kotkov-Akopyan 99
Lasker-Bauer 272
Lehmann-Teschner 96
Mikenas-Averbakh 268
Mohrlock-Kramer 96
Portisch-Karpov 1 3 1
Ravinski-Panov 2 8 6
Reti-Alekhine 2 8 3
Rotlevi-Rubinstein 277
Rubinstein-Snosko-Borovski 1 1 7
Rusakov-Verlinski 97
Spasski-Petrosyan 3 0 1
Steinitz-Bardeleben 273
Suetin-Kholmov 97
Thomas-Shapiro 93
Uhlmann-Averbakh 1 23
Voorema-Zakharyan 95
Zukenort-Blackburne 268
Contents

Introduction . . . . . . . . . . . . . . . . . . . . 5
Compilation of technical terms used in the book 8
Part 1 : The double attack . . . . . . . . . . . . 11
The interrelations between strategy and tactics 11
Three stages of attack . . 14
Methods of defence . . . . 18
The second assault wave . 24
An excursion into the past 26
The fork . . . . . . . . 29
The discovered check . . . 32
The double check . . . . . 37
The two-fold attack on a defended piece 39
The two-fold attack on two targets . . . 42
The two-fold attack in conjunction with a pin . 46
The mutu al two-fold attack . . . 50
The double attack . . . . . . . . . 54
Defence against the double attack 57
Forcing the double attack . 64
1 . The attack . . 64
2 . Zugzwang . . 67
3. The exchange 69
The sacrifice . . . . 70
Decoy attack . . . 77
The two-fold double attack . 80
Particulary strong moves . . 85
The double attack in the opening . 93
The double attack in the end-game 99
Studies of double attacks . . . . . . 1 09
Sample games . . . . . . . . . . . . 1 17
Summarizing remarks on the double attack 1 33
Exercises for independent study 1 37
Solutions . . . . . . . . . . . 151
Part 2 : The combination . . . . 157
Co-ordination of piece moves . 1 57
. rhe mechanism of the mating attack . 1 64
Individual contacts 0 0 0 0 0 171
1 0 The attacking contact 0 1 72
2 0 The restricting contact 0 1 72
3 0 The queening threat 0 0 1 72
40 The protective contact 0 1 74
5 0 The interposing contact 1 74
60 Refuting the queening threat 1 74
The co-ordinated attack 0 1 78
1 0 The cornered king 0 0
0 0 0 0 1 79
2 0 The king on the edge 182
Contacts and combinations in the double attack 0 1 88
1 0 One piece attacks or threatens to attack two objects 0 1 89
20 Two pieces attack or threaten to attack an object 1 93
30 Two pieces attack two objects 0 0 0 0 0 0 0 1 95
Contacts and combinations with passed pawns 1 99
What is a combination ? 0 0 0 0 0 206
Various means of applying force 212
1 0 The attack 0 0 212
20 The threat 0 0 215
3 0 The exchange 217
40 Zugzwang 0 0 219
Kinds of sacrifice 0 22 1
CLissification of combinations 230
Winning combina�ions 0 0 0 0 235
1 0 Combinations against the king 235
20 Combinations against pieces and pawns 238
30 Queening combinations 239
Drawing combinations 240
1 0 Perpetual check 24 1
20 The stalemate 0 0 242
3 0 The-blockade 0 0 0 245
40 Perpetual pursuit 0 247
5� Building fortifications 249
60 Balance of forces leading to a draw 253
The attack on the king 0 0 0 0 0 0 0 0 0 255
The importance of concrete calculation of variants 262
Sample games 0 0 0 0 0 0 268
Exercises for home study 305
Solutions 318
Index of names 323
Index of games 326