Povah, Nigel - Chess Training-Faber and Faber (1981)

<I; CHESS TRAINING
t/f Nigel Povah
FABER AND FABER
London Boston
First published in 1981
by Faber and Faber Limited
3 Queen Square London WClN 3A U
Phototypeset by Western Printing Services Ltd, Bristol
Printed in Great Britain by
Whitstable Litho Ltd Whitstable
All rights reserved
©Nigel Povah 1981
This book is sold subject to the condition that it

CONDITIONS OF SALE
shall not, by way of trade or otherwise, be lent,

re-sold, hired out or otherwise circulated without
the publisher's prior consent in any form of
binding or cover other than that in which it is
published and without a similar condition
including this condition being imposed on the
subsequent purchaser
British Library Cataloguing in Publication Data
Povah, Nigel
Chess training.
1. Chess - Handbooks, manuals, etc.
I. Title
794. 1'2 GV1445
ISBN 0-571-11 604-3

ISBN 0-571-11 608-6 Pbk
To
my family and friends
Author's Note
I should like to express my deepest gratitude to Doris Hall for the

conscientious way in which she typed my manuscript.
Contents
Introduction page 11
Algebraic Notation 13
Preliminary General Advice 16
1. Basic Endings 17
2. Openings in Theory and Practice 35

3. Basic Middle Game Features 56
4. Pawn Structures and Pawn Weaknesses 72
5. Tactical Complexities of the Middle Game 89

6. More Complex Endings 111
7. General Advice 141
8. Chess Unlimited 171
Introduction
The bookshops are full of chess books for beginners, children and the
young enthusiast . . . so why write another?
The author hopes that Chess Training will not be regarded as 'just
another chess book' for keen players wishing to learn the rudiments
of the royal game. The inevitable and continuous improvements in
standards of play throughout the chess population, make it impera
tive for the young ambitious player to have a book which will take
him beyond a basic mastery of the rules and moves.
For this reason this book is primarily aimed at players in the age
range of ten to fifteen years and/or the BCF (British Chess Federation)
grading range 100-1 60. Such a player will often experience a period
of several years (usually in the early teens) during which rapid
progress is made. It is the author's object to guide players through
this interesting phase by showing them which areas of chess are
most relevant and worthy of study. It is also hoped that the young
enthusiast will adopt a methodical approach to his studying, to save
time and effort while improving results.
However, it should be noted that there is no reason why players
outside the defined categories should not also derive benefit from
this book. For instance, chess teachers and coaches may find the
contents and instruction techniques helpful in their own teaching
situations.
The author hopes that with the use of this book an enthusiastic
player will be able to take another step towards realizing his true
potential.
N. E. POVAH
JULY 1979
Algebraic Notation
Many hardened fans of English descriptive notation may regret the

exclusive use of algebraic notation, but the author, among others,
believes that algebraic notation is far easier to use once the player
overcomes the initial problems.
Furthermore, FIDE (Federation Internationale des Echecs-the
World Chess Federation) has ruled that algebraic notation will be the
only recognized form of notation in international tournaments as
from 1981; and since this book is attempting to train young players to
aspire to reach the top, they should be prepared for every eventual
ity!
Algebraic notation uses a very simple cross-referencing system (as
in map readings) in which each square on the board is given one, and
only one, reference.
Diagram i
The references are determined by labelling the eight files (col

umns) with a different letter, from a-h, and the eight ranks (rows)
with a different number, from 1-8 , and then combining these two
labels to form a separate reference for each square.
The board is always considered from White's point of view, and
his pieces occupy ranks 1 and 2 at the beginning of the game. Hence,
White will always be playing 'up the diagram' throughout the
book.
14 ALG E B R A I C N OTAT I O N
The pieces are designated by capital letters a s follows:

KING K BISHOP B
QUEEN Q KNIGHT N
ROOK R PAWN (P)
It is more common to omit P for pawn, s"o the
absence of a capital letter implies a pawn move.
NOMENCLATURE
x captures or takes
0-0 castles on the king side
0-0-0 castles on the queen side
+ check
e.p. en passant
good move
!! very good move
? bad move
?? very bad move
!? interesting move
?! dubious move
± advantage to White
+- decisive advantage to White
advantage to Black
- + -- decisive advantage to Black
the position is equal
6 with the idea of
H draw
1-0 White wins
0-1 Black wins
zugzwang move-bound, forced into making a bad move
tempo 1 move (tempi, 2 or more moves)
fianchetto when a bishop on knight's squares, b2 , d2 , b7 or g7,
is surrounded by pawns
isolani isolated pawn (IQP)
The following moves, written first in extended algebraic and second

in descriptive, should give rise to the position in Diagram ii.
Play through the algebraic moves if you are unfamiliar with this
form of notation.
ALG EBRAI C N OTAT I ON 15
ALGEBRAIC DESCRIPTIVE
White Black White Black
1 e2-e4 e7-e5 P-K4 P-K4
2 Ngt-f3 Nb8-c6 N-KB3 N-QB3
3 Bft-b5 a7-a6 B-N5 P-QR3
4 Bb5Xc6 d7xc6 BxN QPxN
5 �o f7-f6 �o P-B3
6 d2-d4 e5xd4 P-Q4 PxP
7 Nf3xd4 c6-c5 NxP P-QB4
8 Nd4-b3 Qd&xdt N-N3 QxQ
9 Rf1xd1 Bc8-d7 RxQ B-Q2
10. Bct-f4 �o B-B4 �o
Diagram ii
As can be seen extended algebraic notation is much more precise

and once mastered it is easy to use.
Throughout the rest of this book normal algebraic notation will be
used, which simply involves a capital letter to designate the piece
moved and a reference for the square of arrival. Using this method
the above sequence of moves would be:
White Black White Black

1 e4 e5 6 d4 exd4
2 Nf3 Nc6 7 Nxd4 c5
3 Bb5 a6 8 Nb3 Qxdt
4 Bxc6 dxc6 9 Rxdl Bd7
5 � f6 10 Bf4 ��o
Note the marked decrease in the number of dashes, etc.

16 ALG EB RA I C N OTATION
It is important that the reader makes sure that he or she is familiar

with the above method of notation before continuing.
Preliminary General Advice
Although the penultimate chapter of the book also gives general

advice, it is very important that certain habits and ideas are put into
practice as soon as possible--hence the reason for putting this sec
tion here.
First, the keen aspirant should look round for a good local chess
club to join; unfortunately, a few chess clubs do not adopt the policy
of welcoming and encouraging younger members, but there are
many clubs which provide plenty of competitive matches and a
congenial atmosphere in which the student can improve his game.
Second , chess can and , to some extent, should be considered just
like any other subject studied at school or college. The keen student
should try to set aside a regular time for study and should try to
maintain a steady routine, never forgetting that chess is a game
which should always bring enjoyment.
Having established that regular study sessions are important, the
student can then use this book as a source of study material and
suggested study techniques. The conscientious student should get a
maximum return for his efforts, and an important piece of advice,
which is evident throughout the book, is to study as many Master
games and to practise as much as possible for, as the saying goes,
'Practice makes perfect'!
Basic Endings
Why should any player study endings? Such a question should not
need to be asked; but unfortunately aspiring or established players
too often tend to neglect this phase of the game. Many club players
complete 30 or 36 moves in League chess, only to reach an ending
which neither player knows how to assess-let alone play-and the
usual remark is 'Oh well, the adjudicator can sort that one out! ' This
may be acceptable to a casual player, but the keen and ambitious
player must know more.
There are two basic reasons why at least a fundamental knowledge
of the endgame is necessary. First, during the course of a game
various opportunities may arise to exchange pieces and it is neces
sary to know which endings are won or drawn in order to make the
correct decision. Second, in games that are played to a finish the
player should know the winning procedure in many different
endgame positions: it is one thing to know that one has a won ending
and another to be able to demonstrate the win.
An illustration of a player's faili11g to realize this first point was
seen in the following example:
• • •••
-�t•
·· · . "
rwt�t
�- -
•t•..t.• •
• • • •
-�p
d�U •
d •
. . .{). Diagram 1
ft B. ft •�H ft�:: G. D. Lee-A. Bernfield
London
� ; :�.§· • .§ 1 975
18 BAS I C E N D I N G S
The diagrammed position was reached after White's 1 1th move in

the game-G. D. Lee v A. Bernfi.eld, London League, 1975. White has
excellent winning chances in an ending, since Black's doubled
f-pawns make it virtually impossible for him to create a king-side
passed pawn, while White's effective majority on the queen side
should allow him to create a passed pawn with excellent winning
prospects. At the moment, however, Black has some compensation
in the form of the two bishops in a reasonably open position. Clearly
Black should avoid exchanges, but unfortunately he failed to
appreciate the problems that awaited him in the ending and con
tinued 11 . . . Bd5 ?! 12 Qxe7 Bxe7 13 Bxd5 Rxd5 14 Rhe1 Re8? This self
pin is clearly wrong, but Black is in difficulties anyway 1 5 Re2 Rdd8
16 Rde1 Kd7 17 c4. White prepares for the minor-piece ending, since
Black cannot avoid the exchange of rooks if he is to activate his
pieces-1 7 . . . Bd6 18 Rxe8 Rxe8 19 Rxe8 Kxe8 and White has
obtained a clearly superior endgame, which he duly won on move 57.
Thus we can see the obvious dangers of totally neglecting the
endgame. Some players avoid the disaster of our first illustration by
learning which positions are won or drawn, but fail to study the
techniques that are required to achieve the desired result. For
instance, the Swedish Grandmaster Ulf Andersson has twice man
aged to win the theoretically drawn King , Rook and Bishop versus
KiD.g and Rook ending, against opponents who certainly knew that
the ending was drawn, but who failed to demonstrate the correct
defensive technique.
A reasonable knowledge of endings has always formed an integral
part of the Master's armoury. Today, even the average club player
must spend some time studying this phase of the game; since leagues
such as the London League (the strongest league in the country)
encourage games to be played to a finish, and will not adjudicate
until at least 60 moves have been played. Most weekend tourna
ments have also abandoned adjudications in favour of 'the blitz
finish' , which entails adding an additional amount of time (say, ten
or fifteen minutes) to each player's clock once the first time control
has been reached, when the players must then complete the game
before one of them loses on time. Obviously in such a situation
knowledge of endings will be beneficial. Having established that we
cannot neglect endings, we can begin to familiarize ourselves with
some of the fundamentals; let us start by examining what we shall
call the elementary mates.
BAS IC E N D I N G S 19
ELEMENTARY MATES
The starting-point for endings is the extra material (i.e. the bare
minimum) required to force mate against a lone king, and for this
purpose we shall assume that there are no pawns on the board.
1. King and Queen versus King
Diagram 2
This ending usually arises after the promotion of a pawn, when the
winning procedure is very simple. The queen should be used to
confine the enemy king to a small area of the board, such as the five
squares ( d8, e8, f8, g8 and h8) seen in Diagram 2. Then the white king
is brought up to support the queen until we reach a position similar
to that of the diagram, when the queen can deliver checkmate with
four different moves. An important point to remember in this posi
tion is the danger of stalemating the lone king-a disaster w�ich can
easily be averted by always allowing the lone king· at least two
squares.
Consider the position-White: Ket Qdt , Black: Ke5 . Play could
continue-1 Qa4 Kd5 2 Kd2 Ke5 3 Ke3 Kd5 4 Qb5 + Kd6 5 Ke4 Ke6
6 Qc6 + Ke7 7 Kf5 Kd8 8 Qb7 ! (Not 8 Ke6 ? stalemate!) 8 . . . Ke8 9 Ke6
Kf8 10 Qf7 mate.
2. King and Two Rooks versus King

An even easier win than with a queen, since the two rooks mate on
their own without the help of the king. This helps to show that two
co-operating rooks are slightly better than a queen.
Consider the following position-White: K e t Rat Rht , Black:

Ke5 . Play might continue-- t Rh4 confining the king t . . . Kf5 the
king rushes to attack the rook, since rooks operate best at a distance
2 Ra5 + Kg6 3 Rb4! maintaining maximum distance from the king
without interfering with the other rook 3 . . . Kf6 4 Rb6 + Ke7 5 Ra7 +
Kd8 6 Rb8 mate.
3. King and Rook versus King
Diagram 3
Here White must use waiting tactics in order to drive the black king
back by checking from the side, when the kings are opposite. This
can only be achieved by the kings being a knight's move apart with
the black king being forced to move opposite, when the white rook
would give check. Eventually a mating set-up similar to those in
Diagram 3 will be achieved. In the top half of the diagram, Black
must play . . . Kg8, since his king is restricted by the edge of the
board when White mates with Rb8; whereas in the bottom half of the
diagram White must play Kdt (Since his king is restricted by the
rook) allowing Rfl mate.
Consider the position-White: Ket Rht , Black: Ke5. Play might
continue--t Ke2 Kf4 2 Kd3 (inviting 2 . . . Kf3 when 3 Rfl + Kg2
4 Rf4 would leave the black king severely restricted, e.g. 4 . . . Kg3
5 Ke3 Kg2 6 Rg4 + Kft 7 Rg3 ! waiting for 7 . . . Ket 8 Rgt mate) 2 . . .
Ke5 3 Rft Kd5 (3 . . . Ke6 4 Kd4 loses ground without a fight) 4 Rf5 +
Ke6 5 Ke4 Kd6 6 Rh5 Kc6 7 Kd4 Kb6 8 Kc4 Ka6 ? (8 . . . Kc6 9 Rh6 +
Kd7 to Kd5 would allow Black to prolong the agony) 9 Rb5! There
are four edges to the chess-board equally effective in assisting with
the mating process! 9 . . . Ka7 1 0 Kc5 Ka6 1 1 Kc6 Ka7 t2 Kc7 Ka6
t3 Rc5 Ka7 t4 Ra5 mate.
BAS I C E N D I N G S 21
4 . King and Two Bishops versus King
Diagram 4
Due to the fine co-ordination of the two bishops on an open board,
the black king is soon driven into a corner and one waiting move is
all that is needed to administer mate. In the position-White: Kel ,
Bel , Bfl , Black: Ke5 . Play could continue-1 Bd3 Kd4 2 Kd2 Kd5
3 Kc3 Ke5 4 Kc4-notice the impenetrable barrier formed by the two
bishops along the b1-h7 and c1-h6 diagonals-4 . . . Kd6 5 Bf4 +
Kc6 6 Bf5 Kb6 7 Bd7 Ka5 8 Kc5 Ka6 9 Bc7 Kb7 10 Bb6 Kb8 (not 10 . . .
Ka6 ? allowing 1 1 Bc8 mate) 1 1 Kc6 Ka8 1 2 Bgl (not 12 Kc7? stale
mate! ) Kb8 13 Kb6 Ka8 14 Bh3 the ' waiting move'-since the black
king must be 'checked into the corner' for the knockout blow-14 . . .
Kb8 1 5 Bh2 + Ka8 16 Bg2 mate. It is worth noting that only three
checks were given and, in fact, . the mate could have been adminis
tered with only the last two checks, which are always necessary.
5. King, Bishop and Knight versus King
Diagram 5
22 BASIC ENDINGS
Here the black king must be driven into a corner of the same colour as
that on which the bishop operates, to get a final mating position
similar to that given in the top half of Diagram 5. Driving the king
back is usually a two-stage process: (i) forcing the king to the edge of
the board and normally the wrong-coloured corner, since the king
will naturally avoid the 'correct' corner; (ii) driving the king along
the edge of the board from one corner to the other.The first stage,
although not as easy as with two bishops, is none the less not too
difficult.
The second stage can be quite tricky and a useful tip worth
remembering is the pattern of knight moves indicated in the bottom
half of Diagram 5.Play from this position continues-t Bc4 waiting
for the black king to move towards the correct corner-ht.t ...Kct
2 Ba2 Kdt 3 Nd4 Ket (3 ...Ket 4 Ne2+ is similar) 4 Kd3 Kf2 (4 ...
Kdt 5 Ne2 is again similar) 5 Ne2 Kg2 6 Be6! Kf3 7 Bd7 winning a
tempo 7 ...Kf2 (7 ...Kg2 8 Ke3 is no better) 8 Bc6 Ket 9 Ke3 Kdt
10Ba4+Ket ttNf4Kft t2Bb3-another waitingmove-Ket t3Ng2+
Kft t4 Kf3 Kgt t5 Kg3 Kft t6 Bc4+ Kgt t7 Nf4 Kht t8 Bd3-this
final waiting move allows White to give a knight check, driving the
king into the corner, followed by the decisive bishop check ... t8
... Kgt t9 Nh3+ Kht 20 Be4 mate.
The above method is not the quickest way to mate, but it has the
advantage that a player should readily be able to recognize positions
which arise in the mating procedure.
Diagram 6 shows a position from a game of Nilsen-Lee, Post Office
Team Tournament, t977:
Diagram 6
Nilsen-Lee
Post Office
Team Tournament
1977
Play continued-5 t Ke5 Nc5 52 Kf6 Kd4 53 Kf5 Be4 + 54 Kf6 Kd5
55 Kf7 Ke5 5 6 Ke7 Bd5 57 Kea Ke6 (57 . . . Kf6 ! ) 5a Kfa Kf6 59 Kea
Be4 (preparing to start from the beginning of the above analysis , e.g.
60 Kfa Nd7 + 6 t Kga Ne5 62 Kha Nf7 + 63 Kga Bf5 , etc.-however 59
. . . Bc4 ! was quicker) 60 Kda Ke6 6 t Kc7 Nd7 when White finds
himself in a mating net similar to that reached after a Bc6 in the
above example; White resigned on move 7 2 .
6 . King and Two Knights versus King
Diagram 7
Here there is no way to force mate. In the top half of Diagram 7 Black
simply avoids t . . . Kfa ?? 2 Ne6 mate by playing either t . . . Kg7 or t
. . . Kha. Similarly in the bottom half of the diagram after t Nc3 +
Black should play t . . . Ket with no problems, but not t . . . Kat ? ? 2
Nc2 mate. Basically the defending king will always have a choice of
at least two squares and the right choice should always lead to safety.
These last three minor-piece situations have demonstrated the
superiority of a bishop over a knight on an open board-mate with
two bishops is fairly easy, while bishop and knight is fairly tricky,
whereas with two knights it cannot be forced.
This concludes the survey of the elementary mates and we are now
ready to consider king and pawn endings.
24 BAS I C E N D I NGS
KING AND PAWN ENDINGS
We consider king and pawn endings in this chapter because they are
among the easiest endings to play, but that is not to say that these
endings are without hidden resources.
The objective of all king and pawn endings is to 'queen' a pawn
when the win is usually straightforward, so an extra pawn quite
often proves to be decisive. But we must consider a variety of differ
ent positions, based on the relative position of the two kings, in order
to come to terms with the basic principles which apply in such
endings.
The Square
Diagram 8
First we consider a straight race between the pawn and the enemy
king. Play from Diagram 8 could continue--1 h6 Kf6 2 h7 Kg7 when
the pawn is lost. It can quickly be seen if the king can stop the
pawn in this way by imagining a square as shown in the diagram: the
king must be within the square; otherwise the pawn queens.
Two special features need to be noted:
(i) the square of a pawn on the second rank is the same as that of a
pawn on the third, due to the pawn's right to move two squares on its
first move.
(ii) the presence of other pawns on the board may impede the king's
entrance into the 'square' of the passed pawn in question. Troitsky
(1913) composed a beautiful study to illustrate this theme:
Diagram 9
Troitsky
1 91 3
White has a potential passed pawn on the a-file, but first he must
capture the pawn on g2 otherwise Black mates with . . . Kg3 followed
by . . . h5-h4-h3-h2 mate. However, the immediate 1 Kxg2 ? allows 1
. . . Kg5 2 a4 bxa3 e.p. 3 bxa3 Kf6 (not 3 . . . Kxf5 ? 4 a4 Ke5 5 d6! cxd6
6 c6! dxc6 7 a5 Kd5-the only move to keep in the 'square' but after 8
a6 Black cannot play 8 . . . Kc6, and the pawn queens) 4 a4 Ke7 5 a5
Kd8 when Black stops the a-pawn and goes on to win. Thus , the
correct winning procedure is-1 f6! gxf6 when the h�8 diagonal is
blocked 2 Kxg2 Kg5 3 a4 bxa3 e.p. 4 bxa3 Kf5 5 a4 Ke5 6 d6! (6 c6 also
wins) 6 . . . cxd6 7c6! dxc6 8 a5 and the pawn cannot be stopped.
Another example where the principle of the square is relevant can
be seen in the following case: White-Kh3 Pa3 , Black-Ka2 Pf3 .
White to play appears to be winning after 1 a4 since the black king
cannot catch the white pawn, while the white king is already in the
square of the black pawn. But this situation is illusory and with
correct play the result is a draw-1 . . . Kb3 2 a5 Kc4 ! 3 a6 (3 Kg3
Kb5 = ) 3 . . . Kd3 ! 4 a7 (4 Kg3 Ke3 5 a7 f2 6 a8 = Q f1 = Q with a draw) 4
. . . f2 and both sides promote.
Black managed to draw by threatening to enter the square of the
a-pawn while simultaneously rushing over to support the advance of
his f-pawn.
Pawn on the Sixth Rank

We logically come to the conclusion that in many positions the pawn
requires the support of its own king if it is to promote. Support from
the king , however, is not a guarantee of victory; consider the follow
ing diagram:
26 BAS l"C E N D I N G S
Diagram 1 0
The result of this position, a s with many king and pawn endings, is
determined by whose turn it is to move. With White to play, the
result is a draw since 1 d7 + Kd8 2 Kd6 is stalemate, and king retreats
do not help: 1 Ke5 Kd7 2 Kd5 Kd8 ! (other moves lose, such as 2 . . .
Ke8 ? 3 Ke6 or 3 Kc6, winning in either case after 4 d7) 3 Ke6 (3 Kc6
Kc8 is just the same) 3 . . . Ke8 and White has made no progress.
However, with Black to play, White wins after i . . . Kd8 2 d7 Kc7
3 Ke7 since White has achieved a major objective of controlling the
'queening' square. White should make sure that he reaches Diagram
10 with Black to move.
The Opposition
White has a better chance of achieving the aforementioned objec
tives if he remembers to keep his king in front of his pawn, while
making use of a principle called 'the opposition'.
To understand the opposition it will help to consider the position
on the left-hand side of Diagram 1 1 :
_;,1 •
�- /. .7,
• • • •
��
•: - . ffi
�

-
• a R ftR
• • • •
• • • •
• • • •
<it>� • • Diagram 1 1
It may come as a surprise to discover that the white king can force his
way through to either aa or ca by using only the a, b and c files.
Thus-1 Kh2 Kb7 2 Kb3 Kb6 3 Kb4-White has obtained the opposi
tion since the kings are directly opposed and neither can advance;
Black, however, is to move and must either retreat voluntarily or step
aside, e.g. 3 . . . Ka6 4 Kc5-Black has stepped aside so White can
bypass 4 . . . Kb7 (4 . . . Ka7 5 Kc6 and White's king marches on) 5
Kb5 White obtains the opposition again, since it is Black to play, who
must retreat. 5 . . . Kc7 6 Ka6 Kha 7 Kb6 opposition! 7 . . . Kea a Ka7
Kc7 9 Kaa.
To understand the opposition more clearly, look again at Diagram
11 and consider the following sequence where Black does not
advance immediately-1 Kh2 Kaa 2 Kb3 Kb7 3 Kb4 Kb6 when Black
has obtained the opposition (since it is White to move) and White
can make no progress, e.g. 4 Ka4 Ka6 5 Kb4 Kb6 6 Kc4 K.c6. Hence a
player is said to have the opposition when the kings are opposing
and it is his opponent's turn to move.
With correct play from the diagram, White can take and hold the
opposition throughout, if he remembers two points:
(a) when Black's king moves to a square on the centre file (in this case
the b-file) , th,en White moves on to a square of thesame colour on the
same file.
(b) If Black goes sideways to the a or c file, then White should
advance to the other file. This is known as bypassing. Therefore, 1
Kh2 Kaa 2 Kc3 bypassing 2 . . . Kb7 3 Kb3 !-obtaining the distant
opposition-3 . . . Kea 4 Ka4 still bypassing; 4 . . . Kc7 5 Ka5 Kb7
6 Kb5 Kha 7 Kh6 when White reaches aa or ca.
To see why the opposition is so important, consider the right-hand
side of Diagram 1 1 . If it is Black's tum to move (i.e. White has the
opposition), then the win is easy-1 . . . Kea 2 Kg7 or 1 . . . Kga 2 Ke7
when in both cases the pawn is shepherded home by the white king.
With White to move (Black having the opposition) play might con
tinue--1 Ke6 Kea (other moves allow 2 Ke7 + ) 2 f6 winning since
-
White has now gained the opposition and has a position similar to
that of Diagram 10-2 . . . Kfa 3 f7 Kg7 4 Ke7 wins.
In order to reach this winning position in the right-hand side of
Diagram 1 1 , White must first use the opposition to push back the
black king and, only then, advance his pawn; otherwise the pawn
will interfere with the white king and Black will gain the opposition.
Consider, for instance, the two situations in Diagram 1 2 :
Diagram 1 2
On the left-hand side o f the diagram White's pawn i s already on the
fourth rank and the white king has no chance of getting in front of the
pawn in order to win the opposition. Play might continue-1 Kb2
Kb7 2 Kb3 gaining the distant opposition 2 . . . Kb6 (now White
cannot keep the opposition since he cannot play Kb4)' 3 Kc4
Kc6-Black has gained the opposition and White must give up all
hope of getting his king in front of the pawn-4 bs+ Kb6 5 Kb4 Kb7 !
regaining the opposition 6 Ka5 Ka7 7 b6+ Kb7 a Kbs Kba!-other
moves lose-9 Kc6 (9 Ka6 Kaa also draws) 9 . . . Kea when Black has
achieved a similar position to Diagram 10 with White to move, hence
a draw. In this example we saw the interference of the pawn with
White's attempts at maintaining the opposition: if we now consider
the right-hand side of Diagram 12 we get a different picture-1 Kf2
Kfa-maintaining distant opposition 2 Kg3 Kg7 3 Kg4 Kg6 when
Black has kept the opposition, but White has a pawn! 4 g 3 ! Now the
boot is on the other foot and Black must concede ground 4 . . . Kg7
5 Kg5 Kf7 6 Kh6 ! (bypassing, but not 6 g4? Kg7 when Black has
regained the opposition and would draw) 6 . . . Kga (6 . . . Kf6 7 g4
Kf7 transposes) 7 g4 Kf7 (7 . . . Kha a g5 Kga 9 Kg6 transposes) a g5
Kga (a . . . Kfa 9 Kh7 wins) 9 Kg6 , when White has achieved the
winning position shown on the right-hand side of Diagram 1 1 .
Having considered those two examples we should note that, i n the
first case, White could not get his king in front of his pawn and had to
be content with a draw, whereas, in the second case, he did manage
to get his king in front of the pawn but found that he had lost the
opposition. Fortunately he had a pawn move with which to regain
the opposition. Logically we should consider the following position
with the pawn on the third rank-White: Kel Pe3 , Black: Kea. Elay
would continue-1 Ke2 Ke7 (gaining the opposition since White

cannot play 2 Ke3) 2 Kf3 Kf7 ! 3 Kf4 Kf6 4 Ke4 Ke6 when Black has the
opposition and White has no pawn move with which to obtain the
opposition, therefore White can no longer keep his king in front of
the pawn if he is to make progress, e.g. 5 Kd4 Kd6 6 e4 Ke6 7 e5 Ke7
a Kd5 Kd7 9 e6 + Ke7 10 Ke5 Kea ! with a draw.
SOME GENERAL THEMES
Now we must consider some general cases within king and pawn
endings , in order to complete our initial study of these instructive
endings.
1. King and Rook's Pawn versus King

Rook's pawns are often regarded as a curse in many endings , since
they frequently prove impossible to promote. The reason for this is
that the inferior side can use the defensive resource of stalemate.
Consider the case-White: Kg6 Ph6 , Black: Kha . White to move
cannot force the black king out of the corner, while 1 h7 stalemate's
Black immediately. Black to move would simply play 1 . . . Kga
2 h7 + Kha when White must either give stalemate with 3 Kh6 or lose
his pawn.
Therefore, in order to draw against a rook's pawn, the defending
king simply needs to get (or be able to get) on to the queening square.
Whenever possible the white king should try to prevent th� black
king from getting into the corner. Consider the position in ..Diagram
13:
Diagram 1 3
30 BAS IC E N D I N G S
Black to move plays-1 . . . Ke7 but White must not play 2 h4? Kfa
3 Kh7 (3 h5 Kga draws easily since 4 . . . Kha cannot be prevented)
3 . . . Kf7 4 h5 Kfa 5 h6 (White must not allow . . . Kga followed by . . .
Kha) 5 . . . Kf7 6 Kha Kfa 7 h7 (7 Kh7 Kf7 draws by repetition) 7 . . . Kf7
and Black has stalemated White! Instead of 2 h4 ? White wins easily
with 2 Kg7 ! Ke6 3 h4 Kf5 4 h5 Kg5 5 h6 and the pawn goes through
safely.
2. The 'Outside' Passed Pawn

The outside passed pawn normally confers a substantial advantage
in king and pawn endings. By the term 'outside' we mean that the
pawn in question is away from the main body of pawns, quite often
on the other side of the board.
Diagram 1 4
For instance, in Diagram 1 4 White has an outside passed pawn on

b2 . Although material is level White's advantage is clear, since he
can use his outside passed pawn as a decoy to distract the black king
from his defensive duties-1 b4 Kd6 2 Kf5 Ke7 3 b5 Kd6 (3 . . . Kf7 ?
would take the black king out of 'the square' of White's passed pawn)
4 Kxf6 Kc5 5 Kxg5 Kxb5 when the white b-pawn,' although lost, has
done his job of distracting the black king, leaving White with a
simple win. '
Another example of the power of the outside passed pawn can be
seen in Diagram 1 5 :
Diagram 1 5
White has an extra pawn, but at the moment no passed pawns;

which move should he play-1 h4, 1 f4 or 1 Kg2 ? In fact the best
move is 1 f4 ! sacrificing the extra pawn to obtain an outside passed
pawn, e.g. 1 . . . gxf4 + 2 Kf3 Kg5 3 h4 + Kxh4 with an easy win, for
White's king is closer to the remaining pawns. The immediate 1 Kg2
doesn't seem to achieve anything after 1 . . . Ke5 , while 1 h4 ? is
positively suicidal after 1 . . . gxh4 + 2 Kxh4 Kf4 3 Kh5 Kxf3 4 Kg5
Ke4 when it is the black king who reaches the queen-side pawns
first.
3. The 'Protected' Passed Pawn

A passed pawn is always a useful asset, so much so that Nimzovich
in his excellent bookMy System remarked that, 'the passed pawn is a
criminal, who should be kept under lock and key. Mild measures ,
such as police surveillance, are not sufficient.' By this he meant that
one of the defending pieces needs to attend ('blockade') the impos
ing pawn. In a king and pawn ending, only the king can fulfil this
task; consequently a passed pawn grows in stature as exch!lllg es
occur. The advantage of the passed pawn, however, is less effective if
it requires the defensive support of its own king, who would much
rather be able to roam the board in pursuit of vulnerable pawns.
The protected passed pawn who enjoys the support of one of
his fellow pawns is an even greater asset. For instance, consider
Diagram 16:
32 BAS I C END INGS
Diagram 1 6
White has a protected passed pawn on f6 which ties the black king
down to a purely defensive role, while the white king is able to attack
and dissolve Black's queen-side pawn majority thus 1 a4 Ke6 2 axb5
(not 2 c4 ? b4 when both sides would have protected passed pawns
and neither side could do anything, hence a draw) 2 . . cxb5 3 c4
.
bxc4 + (if 3 . . . b4 then 4 c5 leaves Black the impossible task of

stopping both passed pawns, e.g. 4 . . . Kd5 5f7 wins) 4 Kxc4 Kf7
(Black cannot attack the supporting g�pawn because the f-pa·wn
would be free to run-4 . . . -Kf5? 5 f7 wins) 5 Kb5 a4 6 Kxa4 Ke6 ;
having demolished the queen-side pawns White must now make use
of his extra pawn on the king side. ,
Normally this is straightforward , with the white king supporting
the advance of the passed pawn, or alternatively winning Black's
remaining pawn. The resulting position from our example
-although winning-is rather tricky:
Diagram 1 7
BAS IC END INGS 33
7 Kb5 Kd7 8 Kc5 Ke6 9 Kc6 Kf7 10 Kd7 Kf8 1 1 Ke6 ( 1 1 Kd6 Kf7
doesn't help) 1 1 . . . Ke8 1 2 f7 + Kf8 1 3 Kd6! (not 1 3 Kf6 stalemating) ;
White sacrifices the advanced f-pawn to gain the opposition, win the
black g-pawn and then to escort his own pawn home by achieving
the winning position on the right-hand side of Diagram 1 1 ; 13 . . .
Kxf7 14 Kd7 ! 'horizontal opposition' 14 . . . Kf8 1 5 Ke6 Kg7 1 6 Ke7
Kg8 1 7 Kf6 Kh7 18 Kf7 Kh8 19 Kxg6 Kg8 2 0 Kh6 Kha 21 g6 Kg8 22 g7
Kf7 23 Kh7 wins.
4. Effective versus Ineffective Pawn Majorities

Quite often positions arise when the players have an equal number of
pawns, but the balance is different; if, for instance one player has a
majority on one wing and the other player has a majority on the other
wing. Sometimes one of these majorities will be more effective than
the other, and this can be enough to win a game. Let us consider
Diagram 18 which could arise from the following sequence of
moves-1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Bxc6 dxc6 5 0-0 f6 6 d4 Bg4
7 dxe5 Qxdl 8 Rxdl fxe5 9 Rd3 Bxf3 10 Rxf3 Nf6 1 1 Nc3 Bb4 12 Bg5
Bxc3 1 3 bxc3 Rf8 (if 1 3 . . . Nxe4 14 Re1 !±) 14 Bxf6Rxf6 1 5 Rxf6 gxf6
16 Rdl . So far, Fischer-Smyslov, Monaco 1967, when Smyslov
continued 16 . . . Ke7 and after a long hard struggle managed to
draw. Many players might challenge White's control of the d-file by
16 . . . Rd8 when 1 7 Rxd8 + Kxd8 gives our diagrammed position:
Diagram 1 8
Although material is level, this position is a simple win for White
who has an effective 4 versus 3 pawn majority on the king side. Black
on the other hand, has an ineffective 4 v 3 majority on the queen side,
despite the fact that White has doubled c-pawns. Play might con-
34 BAS I C E N D I N GS
tinue-18 f4 ! exf4 ( 18 . . . Ke7 19 fxe5 fxe5 when White has a poten

tial outside passed pawn which should prove to be decisive) 19 Kf2
Ke7 20 Kf3 Ke6 21 Kxf4 b5 (21 . . . h5 22 h4 intending g4 creating a
passed h-pawn) 22 g4 c5 23 h4 b4 24 c4 (24 e5 also wins, but not 24
cxb4 ? undoubling Black's c-pawns and giving him the chance to
create a queen-side passed pawn) 24 . . . as-notice the pawn on c7
is useless. White now makes a temporary pawn sacrifice to create an
outside passed pawn. 2 5 e5! fxe5 + (if 2 5 . . . c6 26 exf6 Kxf6 2 7 g 5 +
Kg6 2 8 Kg4 Kf7-not 28 . . . h5 + 29 Kf4-with a decisive protected
passed pawn-29 Kf5 Kg7 30 h5 Kf7 3 1 Ke5 Ke7 32 g6 hxg6 3 3 hxg6
a4 34 g7 Kf7 3 5 Kd6 and White demolishes Black's remaining
pawns) 26 Ke4 a4 27 g5 (not 2 7 h5 ? h6 when White's majority is
crippled! ) 2 7 . . . c6 28 h5 a3 2 9 g6 hxg6 30 hxg6 (if Black's king were
on d6 then White could win without recapturing, i.e. 30 h6 Ke6 3 1 h7
wins) 30 . . . Kf6 3 1 g7 Kxg7 32 Kxe5 Kf7 3 3 Kd6 and the black king
can only look on as White mops up the queen side. This elementary,
although prolonged, win was never in doubt once the last pair of
rooks were exchanged, due to Black's inability to create a queen-side
passed pawn.
This example concludes our examination of basic endings, but the
reader is urged to continue his studies of such endings: it was such a
knowledge of the resulting king and pawn ending that enabled
Smyslov to avert the disaste� of our last example by avoiding the
exchange of rooks. For this purpose the reader is advised to refer to
more specialized books on endings , since it has been possible to give
only brief coverage of a highly detailed and important area.
Openings in Theory
and Practice
Chess is a game which can conveniently be divided into three stages:

1. the opening
2. the middle game
3. the endgame
Most beginners and novices are aware of this subdivision but they
are not always able to identify the borders between the stages. Nor
mally we cons�der the opening phase as the period of the game when
both players tcy to develop all or most of their minor pieces (knights
and bishops) and perhaps some of their major pieces (queens and
rooks) . Castling to find a safe haven for the king is another important
priority in the opening. Ideally these objectives should be carried out
with as few pawn moves as possible; hence Lasker's maxim, 'In the
opening one or two pawn moves not more'. This quote must, how
ever, be taken loosely, but not too loosely!
Before examining opening classification� and the underlying
principles, we should consider ill.e question 'Why do so many play
ers spend much time in studying openings ?' The answer is very
simple--every time you sit down to play a game you will naturally
get an opening, but not necessarily a middle game and even less
likely, an endgame. Nevertheless, this does not imply that one
should neglect the other phases of the game: what is the point of
playing a superb opening and gaining a useful initiative if one does
not know how to capitalize on this in either the middle game or
endgame ?
So now we tum our attention to the basic principles of the opening
which are:
1. development
2. central control
3 . king safety
36 O P E N I N G S I N TH EORY AN D P RACT I C E
I t should be noted that these three concepts ar e frequently related,

since a lead in development will often mean greater central control
and may well provide greater protection to your own king, while
presenting a threat to the enemy king.
Our first game illustrates the dangers of treating the opening
principles with contempt.
Game 1
White Black
1 e4 es
2 Nf3 Nc6
3 Bc4
The 'Giuoco Piano' which is Italian for 'quiet game'--an exciting
opening which usually leads to anything but a quiet game!
3 . . . h6?
Black breaks an opening principle, by playing an unnecessary pawn
move which does not aid his development. More natural was either
3 . . . Bc5 or 3 . . . Nf6 (the Two Knights defence) .
4 �
Notice the economy of White's development; in four moves he has

moved two pieces and castled, making only one, very constructive,
pawn move.
4 . . . d6?!
A reasonable move but ill-timed. Black should play 4 . . . Bc5 and
then 5 . . . d6, so as not to trap his black-squared bishop behind the
pawns.
5 Nc3 a6?
Another totally unprovoked pawn move which wastes valuable
time. 5 . . . Nf6 was much better followed by . . . Be7 and . . . 0-0 .
6 Rel
This move may appear to be rather strange, but it is quite logical.
White envisages a position where the centre pawns will be
exchanged off thereby exposing the black king who has yet to escape
from the centre----consequently the rook is lined up against the king.
6 . . . b5?
Again wasting time, but also committing the sin of permanently
O P E N I N GS I N TH EORY A N D P RACTI C E 37
weakening the queen side, and to no good effect. Remember, 'PAWN

MOVES MUST BE CAREFULLY THOUGHT O uT , FOR THEY CAN NEVER BE
'
RETRACTED.
7 Bb3
Diagram 1 9
7 Bd5 is also good, e.g. 7 . . . Bb7 8 d4 ! When your opponent leaves
his king in the centre, it is sensible to try to exchange off central
pawns to open up the position thus-8 . . . exd4 9 Nxd4 Qd7 (Not 9
. . . Nxd4 10 Bxb7 when White wins material) 10 e5! consistent with
the aforementioned· plan 10 . . . dxe5 (10 . . . 0-0-0 1 1 ·e6 ! is deci
sive for White) 1 1 Nxc6 Bxc6 12 Rxe5+ Be7 13 Qf3 Bxd5 14 Nxd5
when Black has problems completing his development, as the fol
lowing variations demonstrate:
(a) 14 . . . 0-0-0 ? 1 5 Nxe7+ ! Nxe7 ( 1 5 . . . Kb8 1 6 Nc6 + + - ) 1 6 Qa8
mate.
(b) 14 . . . Nf6 ? 15 Nxe7 + -
(c) 14 . . . c6? 1 5 Nb6 Qc7 1 6 Rxe7 + ! Nxe7 (16 . . . Qxe7 1 7 Qxc6 +
wins) 1 7 Nxaa Qb7 18 Qg3 ! Qxa8 1 9 Qxg7 Rg8 20 Qxh6 c5 threaten
ing mate on g2 2 1 Qh3 with a clear cut material advantage. Although
in this variation White appears to have neglected the development of
his bishop and rook, it was well justified and is easily remedied by
Bf4 followed by Rel with an excellent game.
(d) 14 . . . Rd8 1 5 Bd2 ! intending Bb4 and Rael with overwhelming
pressure against e7, e.g. 1 5 . . . f6 16 Qh5 + Kf8 1 7 Nf4! threatening
both Ne6 + and Ng6 + )'1 7 . . . fxe5 18 Ng6 + Kf7 (18 . . . Kea? 19
Nxe5 + g6 20 Qxg6 + Kfa 21 Qf7 mate) 19 Nxe5 + Ke6 20 Nxd7 with
an easy win because of White's material advantage of queen and
pawn versus rook and knight as well as Black's exposed king.
38 O P E N I NGS I N -TH EORY A N D P R ACT I C E
7 . . . Na5?
Black breaks yet another opening principle by moving a piece (in
this case his only developed piece! ) for a second time; this would be
an acceptable idea if Black had completed his development but in
this position it is unforgivable!
8 d4 Nxb3
By making this capture Black loses a move, for the white bishop had
only moved twice while the black knight had moved three
times-thereby giving White an extra tempo (move) . On the plus
side the black knight did little on a5 and Black now has the advan
tage of the two bishops, but in this position White's lead in develop
ment and control of the centre are far more decisive factors.
9 axb3
There is a useful rule of thumb worth remembering:
Capture towards the centre when you have a choice because this
enhances central control and often rids one of annoying rook's
pawns which can prove to be a handicap in the ending, but more of
that later.
9 exd4
. . .
10 Qxd4
White can safely develop the queen on this strong central square
since his development is nearly complete and she is difficult to
attack; 10 . . c5 is, for instance, the only move to threaten the queen,
.
and this.would be anti-positioned, for it would create a weak square

on d5 as well as creating a backward pawn on d6.
Diagram 20
O P ENING S IN TH EORY AND P RACT I C E 39
In the diagrammed position White enjoys a substantial lead in

development, control of the centre and more space in general. He
also has the tactical threat of 11 Nxb5 winning a pawn, since the
black pawn on a6 is pinned to the undefended rook on a6 by White's
rook on al.
This example illustrates two extremes-White's economical and
efficient development as opposed to Black's non-existent develop
ment, due to his failure to comply with the opening principles.
In fact Black managed to break almost every opening principle
in only 10 moves! First he neglected t�e development of his
pieces, second he failed to control or contest the centre; third he
controls less s pace since he has fewer open lines for his pieces than
White; finally he has failed to castle, consequently his king is in
jeopardy.
On the other hand, White's development is almost complete and
we can regard the opening phase as virtually over. Now White must
plan his middle game strategy, and this will be considered in a later
chapter. .
We shall now examine the different opening concepts one by one
with appropriate examples.
DEVELOPMENT
Chess is a game which relates time (in terms of moves) with space
(the 64 squares) . Consequently it is criminal to waste time in the
opening because each move should be seen as an opportunity to
control more squares.
Efficient development should adhere to the following guidelines:
1. Develop all your pieces quickly and effectively; do not attempt to
attack with just one or two pieces developed.
2. Do not move a piece more than once in the opening until you have
completed your development, unless of course, you stand to gain a
material or positional advantage, without compromising your posi
tion.
3. Do not make unnecessary pawn moves in the opening unless,
again, it is to your advantage.
4. Be wary of snatching pawns in the opening if, as a result, your
development is hindered while your opponent's development is
enhanced.
40 O P E N I N G S I N TH EORY AN D P RACT I C E
The next example illustrates the dangers o f neglecting one's

development by making too many unnecessary pawn moves.
Game 2
White: Dr E. N. Bramley (Berkshire) Inter-county Championship,
Black: N. E. Povah (Surrey) 1977
1 c4
The English opening, which is normally regarded as a quiet method
of opening the game, White intends to play on the flanks and nor
mally leaves his centre pawns (e2 and d2) unmoved, so the need for
early castling is not so critical.
1 . e5
. .
2 Nc3 Nf6
3 g3 Bb4
4 Bg2 0-0
Both players have developed sensibly so far, but now White plays a
dubious pawn move which slows his development and also weakens
the white squares. 5 Nf3 is regarded as best.
5 e3?! Bxc3!
A logical capture doubling White's pawns before Nge2 (connecting

the knights) can be played. Furthermore, as Black wishes to attack on
the white squares, it is logical to exchange a piece which moves
exclusively on black squares for one which can defend the white
squares.
6 bxc3 e4!
Immediately exploiting the white square weakens with the intention

Nb8-c6-e5-d3 or f3 .
7 f3?!
White continues to neglect his development and, instead of moving

only three pawns, he now has only three unmoved! Alternatives are
difficult to find, however, e.g. 7 d4 exd3 e.p. leaves White with
doubled isolated c-pawns; 7 Nge2 Nc6 8 C>--0 improves White's
development, but does little to solve the problem of the weak white
squares after 8 . . . Ne5 .
O P E N I N G S I N T H EORY A N D P RACT I C E 41
7 . ReB
. .
8 Qc2?
White should play 8 fxe4 Nxe4 9 Nf3 followed by 0-0 with
approximate equality. The move played puts the queen on a vulner
able square and allows Black to accelerate his development with
tactical tricks.
8 . d5!
. .
9 cxds
White has undoubled his pawns, but Black does not recapture
immediately.
9 . Bfs!
. .
10 f'Xe4 Nxe4
Black threatened 10 . . . exf3 1 1 Qxf5 fxg2 winning. Now the threat is

11 . . . Nxg3 winning the exchange
11 d3 Nc5
Diagram 21
Dr E. N . Bramley-Povah
Inter-county Champion
ship 1 977
Notice the activity of Black's pieces compared with White's, which

are reduced to defensive duties. White now loses his extra pawn and
the black queen enters the game with deadly effect. 12 Bf1 is met by
12 . . . Qxd5+.
12 e4 Qxd5!
13 Kf1 Qxd3 +
14 Qxd3 Nxd3
15 Ba3
42 O P E N I N G S I N T H EORY A N D PRACT I C E
At last White tries t o develop his pieces , but it i s too late; Black's
extra pawn and better position make the win simple. The bishop on
f5 is immune since 1 5 exf5 ?? Rel is mate.
15 . . . Be6
16 Ke2 Bc4
17 Ke3 Nd7
18 Rd1 N7e5
19 Nh3?! Rada
Black's last piece moves in, ready for the kill!
20 Bfl? Ng4 +
0-1
White must los_e further material after 2 1 Kf3 Nde5 + and 22 . . .
Rxdl .
CENTRAL CONTROL
The advantages of central control are easily illustrated when one con
siders the efficiency and mobility of the pieces in different positions.
Consider, for example, the following table which demonstrates the
number of squares controlled by a knight occupying certain squares:
Square occupied Number of squares
by knight controlled
al 2
az 3
a3 4
b3 6
c3 8
It is not difficult to appreciate that the knight grows in stature as it
approaches a more centralized square; in fact the knight on c3 may
be said to be four times as powerful as a knight on at . Of course this
generalization assumes that control of all squares is of equal value,
so when we consider that the knight on c3 controls important central
squares such as d5 and e4 as well as b5, a4 , aZ, b1 , dl and eZ, we soon
realize that this knight is probably more than four times as effective
as a knight on al . From this simple illustration we can conclude that
it is important to deploy one's pieces efficiently when developing, in
order to maximize their mobility and effectiveness. Control of the
centre offers increased mobility, since one's pieces can move readily
to either flank as the need arises. Furthermore, central control can
have the effect of splitting the opponent's forces as the next example
·
illustrates.
Game 3
White Black Vienna Opening
1 e4 e5
2 Nc3 Nf6
3 f4 exf4?
Althouglithis capture is playable in the King's gambit (1 e4 e5 2 f4
exf4 , etc.) it loses precious time here because the knight on f6 is
forced to move again. 3 . . . d5 was better.
4 e5
White takes the opportunity to chase the black knight back to base
and to establish a dominant pawn centre.
4 . . . NgB
5 Nf3
White continues developing and prevents the powerful 5 . . . Qh4.
5 . . . b6
Black intends to fianchetto his light-squared bishop, but he is neg

lecting the centre. 5 . . . d5 was more logical.
6 d4.Bb7
7 Bxf4
Diagram 22
44 O P E N I N G S I N T H EORY A N D P RACT I C E
White has regained his pawn and controls the centre.

7 • Bb4
. .
8 Bd3
With the likelihood that Black will castle short, White prepares an
aSS!i-Ult on the vulnerable h7 square, since the pawn on e5 denies the
black knight the defensive square f6.
8 . Ne7
. .
9 0-0 Bxc3
If Black castles immediately then his dark-squared bishop runs into
trouble, e.g. 9 . . . 0-0 10 Ne4 ! threatening 11 c3 and b4 as well as a
king-side attack with Nfg5, Qh5, Nf6, etc.
10 bxc3 0-0
Diagram 23
White does not mind his doubled pawns because his centre now
enjoys further support. Notice how the centrai wedge of white
pawns tends to split Black's forces into two separate units with the
black king isolated on the king's side. It should not come as a
surprise to find that White has a decisive sacrificial breakthrough!
11 Bxh7 + !
This sacrifice is well known in such positions and is referred to as
the 'Greek Gift' ; the basic features are:
1. Black's castled king behind three unmoved pawns without a
knight on f6.
2. White has a bishop on the bl-h7 diagonal, a knight able to get to
g5, and a queen able to get to h5.
O P E N I N G S I N TH EORY A N D P RACTI C E 45
3. An open or half-open file for a rook (usually the e- or f-file) is an

added advantage.
11 . . . Kxh7
The alternative 11 . . . Kh8 just leaves Black a pawn down with a bad
position.
12 Ng5 + Kg6 .
Not: (a) 12 . . . Kha ? 1 3 Qhs+ Kg8 14 Qh7 mate.
(b) 12 . . . Kh6 ? 13 Nxf7 + + - .
(c) 12 . . . Kg8 13 Qhs Res (the only move) 14 Qxf7 + Kha 1 5
Qh5 + Kg8 1 6 Qh7 + Kf81 7 Qh8 + Ng8 18 Be3 + Ke7 19 Qxg7
mate.
13 Qg4
Threatening 14 Ne6 + winning.
13 . . . f5
14 exf6 e.p.
Then Black could resign since he must lose his queen or be mated.
The reader is advised to verify this for himself.
KING SAFETY
Early castling is invariably advisable in the opening in order to

secure a safe haven for your king. It is true that Reti has said, 'you
should castle when there is no better move' ; but many players tend to
castle sooner than this remark suggests. More often than not they
were justified for, if you delay castling in order to launch a prema
ture advance or attack, then you may suffer a similar fate as that in
the next example.
Game 4
White: G. D. Lee London North Circular League.
Black: A. J. Wheatley 197 ·5
1 e4 c5
2 Nc3
The Closed Sicilian; White does not intend to open the centre with
d4 as in the open variations of the Sicilian.
46 O P E N I N G S I N T H E O RY A N D PRACT I C E
2 . . . Nc6
After this move it would be difficult for White to play d4 in any case.
3 f4 g6
4 Nf3 Bg7
5 g3
Also possible was 5 Bc4 or 5 Bb5 .
5 . . . d6
6 Bg2 RbB?!
A reasonable move but badly timed; Black should concentrate on
castling, e.g. 6 . . . e6 7 . . Nge7 8 . . . 0-0 , and then he can attempt
_.
queen-side expansion with . . . Rb8 and . . . b5, etc.

7 0-0 b5
Again Black should get on with his king-side development.
8 a3 a5?
A very bad move, Black is intent upon playing . . . b4 and could be
accused of criminal negligence of his king side.
9 a4 b4
10 Nb5 Ba6
White is quite happy to sacrifice a pawn, especially as he will gain
tremendous control of the white squares on the queen side.
11 c4! bxc3 e.p.
12 dxc3 Bxb5
13 axb5 Rxb5
Diagram 24
G . D . Le&-A. J. Wheatley
London 1 975
O P E N I N G S I N T H EORY A N D P RACTI C E 47
Black has won a pawn, but his retarded development and weak
white squares offer White more than enough compensation.
14 e5!
Logically opening up the long white diagonal.
14 .dxe5
. •
15 Qa4 Qb6
(not 1 5 . . . Rb6 1 6 Nxe5 + - or 1 5 . . . Na7 1 6 c4 + - )
1 6 Nxe5 Nxe5
17 fxe5 c4 +
Black prevents 18 c4 and does so with a gain of tempo.
18 Kh1 e6?
Diagram 25
Lee-Wheatley
Black's last hope to get his king out of the centre was 18 . . . Nh6 ! (not
18 . . . Bxe5 ? 19 Qxc4! threatening both Qxf7 + and Bc6 + ) 19 Be3 !
Qa6 20 Qxc4 (20 Bxh6 Bxh6 2 1 Qxc4 0-0 and now 22 Rxf7 ? fails to
22 . . . Rxf7 23 Bd5 e6! 24 Bxe6 Qb7 - + ) 20 . . . 0-0 21 Bc6 Rb6 22
Qxa6 Rxa6 2 3 Bb7 Re6 24 Rxa5 Ng4 2 5 Bd4 Bxe5 with an advantage to
White, but at least Black is still in the game!
19 Be3 ! Qa6
(Obviously not 19 . . . Qxe3 20 Qxb5 + + - ) .
2 0 b3 !
48 O P E N I N G S I N TH EORY A N D P RACT I C E
This clever move forces the black king t o stay i n the centre, e.g. 20 . . .
cxb3 loses to 2 1 c4.
20 . . . Kf8
21 Qxc4
Having lost the pin on the a4-e8 diagonal, White now establishes
another pin.
21 . . . Ne7
22 Rad1 Bxe5
23 Rd7 Nf5
24 Rxf5!
White feels justified in sacrificing the exchange because of Black's
vulnerable king and poor piece co-ordination. If 24 . . . gxf5 25 Ra7
forces the queen to desert the pinned rook, e.g. 25 . . . Qd6 26
Qxb5 + - :
24 . . . exf5?
25 Qxf7 mate.
In this example Black was guilty of neglecting his king-side
development, and he was duly punished for leaving his king in the
centre.
Finally, an example from the London Under-14 Championship,
1974, in which Black was guilty of leaving his king in the centre, but
this time for no apparent reason-an even greater sin!
Game 5
White: Taiman London U14 Championships
Black: Wooley 1974
1 e4 c5
2 Nf3 d6
3 d4 cxd4
4 Nxd4 Nf6
5 Nc3
Diagram 26
Taiman-Wooley
London Under-1 4
Championship
1 974
This is a very common position in the Sicilian Defence from which

Black can opt for several different variations: 5 . . . g6 the Dragon
variation, 5 . . . e6 the Scheveningen variation, 5 . . . a6 the Najdorf
variation, and 5 . . . Nc6 the Four Knights variation. The move
played in the game is a poor alternative since it weakens the d5
square and creates a weak backward pawn on d6, which will require
protection from Black's pieces.
5 . . . e5?!
6 Nf3 Be6
7 Bg5 Nbd7
8 Be2 Be7
9 Qd2 a6
10 �o
Both players have developed reasonably well so far, and if Black

were to continue with 10 . . � then his disadvantage due to
.
weaknesses on the d-file would be minimal. However, he now

undertakes a dangerous pawn snatch on the queen side.
10 . . . Qb6?!
11 Rfd1
White baits the trap!
11 . . . Qxbz ?!
Castles short was more prudent.
12 Rab1 Qa3
13 Rxb7
50 O P E N I N G S I N T H E O R Y A N D P RACT I C E
Diagram 27
Taiman-Wooley
White has regained his pawn and now has his rook menacingly
placed on the seventh rank.
13 . . . Nc5?
Black suicidally neglects his king's safety. 13 . . . 0-0 would have
minimized White's advantage.
14 Rxe7 + ! KfB
If 14 . . . Kxe7 1 5 Qxd6 + Kea 16 Qc6 + Ke7 17 Nd5 + Bxd5 1a Rxds
Rhea 19 Qd6 + Kea 20 Rxe5 + Ne6 21 Qxa3 + - .
15 Qxd6 Ne8?
16 Rxe8 + !
( a nice finish)
16 . . . Kxe8
17 Qe7 mate.
Black could have spared himself this disaster if he had castled at
move 10, 11 or 1 3 . The moral to this game is simple-no NOT NEGLECT
THE SAFE'IY OF YOUR KING.
OPENING CLASSIFICATION
Having examined the opening principles, we will now consider the

different types of opening positions to which they apply. Chess
openings are frequently classified as follows:
1. open games
2. semi-open games
3. closed games.
This classification is based almost entirely upon the type of pawn
structure that is evident. Open games usually arise from symmetrical
king pawn openings where early pawn captures provide open lines
for the pieces; the Danish gambit is a classic example-1 e4 e5 2 d4
exd4 3 c3 dxc3 4 Bc4 cxb2 5 Bxb2 with an extremely open position.
Further examples of open games are the King's gambit (1 e4 e5 2 f4
exf4, etc.) ; the Giuoco Piano (1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 c3 Nf6
5 d4 exd 4 6 cxd4, etc.); the Scotch gambit ( l e4 e5 2 Nf3 Nc6 3 d4 exd4
4 c3 dxc3 5 Nxc3 , etc.) , and the Open Ruy Lopez (1 e4 e5 2 Nf3 Nc6
3 Bb5 a6 4 Ba4 Nf6 5 0-0 Nxe4 6 d4 , etc.).
On the other hand, semi-open games generally arise from asym
metrical pawn structures with only a few (one or two) pawn
exchanges. Examples of semi-open games are the Sicilian Defence ( 1
e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 , etc.) , the Caro-Kann (1 e4
c6 2 d4 d5, etc.) ; the Gruenfeld (1 d4 Nf6 2 c4 g6 3 Nc3 d5. 4 cxd5 Nxd5
5 e4 Nxc3 6 bxc3 Bg7, etc.) and the Modern Benoni (1 d4 Nf6 2 c4 c5
3 d5 e6 4 Nc3 exd5 5 cxd5 d6 6 e4 g6, etc.).
Finally, the closed games rarely have any early pawn captures at
all, consequently strategical manoeuvring is the order of the day.
Examples include the Reti (1 Nf3 Nf6 2 g3 g6 3 Bg2 Bg7 4 0-0 0-0
.
,
etc.); The English (1 c4 ) ; the Closed Ruy Lopez (1 e4 e5 2 Nf3 Nc6 3

Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Rel b5 7 Bb3 d6 8 c3 0-0 etc.) ; the
,
Czech Benoni (1 d4 Nf6 2 c4 c5 3 d5 e5 4 Nc3 d6 5 e4 Be7, etc.) and

certain variations of the Pirc and French defences , etc.
It should be noted that this classification is only a rough guide
based on the author's impression of various openings; it cannot be
denied that some variations of these openings , such as the Pirc, can
give rise to reasonably open positions. We can now compare the
different opening principles according to the type of opening in
consideration, as illustrated in the following table:
52 O P E N I N G S I N T H EORY A N D P RACT I C E
OPENING CLASSIFICATION
Opening
Principle Open Semi-Open Closed
Development Speedy devel- Speedy devel- Speed is less

opment is very opment is still important-
important. important. manoeuvring to
occupy key
squares is more
essential.
Central Less important More essential Still important

Control due to flexibility due to danger of because this
of position and expanding might result in a
lack of centre pawn centres. spatial advantage
pawns. for one side. But
static pawns
imply little
danger from
expanding pawn
centre.
King Early castling Still important Less important

Safety is normally but castling can because of
very important. often be delayed blocked pawn
for a few more centre and early
moves. play on the
flanks.
It is hoped that this table will provide the reader with a rough guide
to the different types of openings and their relationship with the
basic opening principles.
Now we turn our attention to the problems of selecting, learning
and understanding an opening.
DEVELOPING AN OPENING REPERTOIRE
How should a player decide which openings to play? This problem

can be handled in three methodical stages :
1 . The student should study his own games in order to determine
the type of position that he favours: does he prefer open tactical or
closed strategical positions. It is simply a matter of being objectively
self-critical in order to identify one's style of play.
2. Next the student should examine the spectrum of chess openings
and select those which suit his mode of play. He should prepare lines
for each opening he expects to meet; for instance, as B lack he must
have defences against 1 e4, 1 d4, 1 c4 and 1 Nf3 , and he also needs
defences against various obscure openings, such as 1 b 3 , 1 g3 , 1 b4
and 1 f4 which are occasionally encountered.
3. The third stage in selecting an opening repertoire is based on the
individual's attitude. Some players prefer to play highly fashionable
openings which require a great deal of learning and continual up
dating, otherwise they run the risk of being surprised by the latest
theoretical wrinkles. Other players might prefer to play a less popu
lar opening which, with some home study, can prove to be a very
effective surprise weapon.
The author, being a basically lazy character, decided to adopt the
latter approach. In 1970 he was well known among his peers as an
advocate of the Sicilian Dragon. So, in preparation for the British
Junior Championship, he sought a new reply to 1 e4. He had decided
to stick with the Sicilian Defence, but which of the many variations
should he play ? He could not face the mass of complicated variations
that arose from the fashionable Najdorf variation, neither could he
understand the ramifications of the more subtle Taimanov and
Scheveningen variations. Having browsed through his small library,
he stumbled across a rather obscure variation known as the Labour
donnais-Lowenthal (1 e4 c5 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 e5) , which
had the merit of being relatively unanalysed with very few recent
Master games. The Lowenthal offered two advantages : first, home
analysis was necessary to understand the opening, which meant that
most opponents would be unprepared; second, the lack of recent
Master games meant that the variation would come as an unpleasant
surprise to many adversaries.
54 O P E N I N G S I N TH EORY AN D P RACTI C E
As a consequence, the author enjoyed a great deal of success with

this variation and by documenting his results (name of opponent,
opponent's grade and variation played) , he was able to determine
which sub-variations needed more analysis, etc. In four and a half
years with the Lowenthal he had the following record:
played - 4 5
won - 2 3
drew - 1 5
lost - 7
against opponents in the following grade ranges:
Grade Frequency Won Drew Lost

200 + 11 2 6 3
190-1 99 8 3 3 2
1 80-189 12 5 6 1
1 70-1 79 7 6 1
below 169 7 7
45
This data suggests that the Lowenthal was a very effective weapon in
the 'black armoury' of the author, particularly when you consider
that his own grade was in the 180s for the first three years of this
period. Why did the Lowenthal serve him so well? The answer is
simply that the methodical study of the opening gave him a fuller
appreciation and understanding of the opening than most of his
opponents.
It is worth detailing the method of study adopted so that the reader
may apply some of the techniques to the formulation of his own
opening repertoire:
1. It is necessary to research through as much chess literature as
possible, e.g. relevant books and magazines. The FIDE publication
Chess Informant is particularly useful. Make use of your local library!
2. A loose-leaf or card-index system should be established,
listing critical games, positions and personal analyses. Both systems
have the advantage of being flexible and allowing for regular up
dating, restructuring, etc.
3 . It is very important to familiarize yourself with an opening before
actually playing it, and there are several ways of doing this:
(a) discuss the ideas and any critical games or positions with chess
friends and club colleagues; save yourself a lot of misery, and

make sure you are aware of any traps and pitfalls that await
discovery!
(b) Practise the opening in a series of 'blitz games' with a suitably
prepared colleague. The author, for instance, used to play count
less five-minute games with a friend in the Sicilian Dragon; after
about 1 7 moves of established theory, each player still had
approximately four minutes in which to investigate a theoretically
critical position. The procedure can be repeated with colours
reversed.
These training techniques enable a player to go to the board
confidently armed with the knowledge that he has a reasonable feel
for some of the positions that result from the opening under study.
Nevertheless, it must be admitted that this rather mechanical
approach may not appeal to everyone, but there are many players
who should derive some benefit from the methods expounded here.
Basic Middle G ame
Featu res
So far we have covered the opening and ending phases of the game,
now comes the difficult task of bridging the gap.
The middle game is undoubtedly the most complex phase of the
game, and it is the stage when the players can allow their true ability
and style of play to come' to the fore.
Unfortunately there is a tendency for players to neglect the middle
game, with most study time being devoted to either the ending or
more commonly the opening. The complexity of the middle game,
with its wealth of different strategies, is well worth study, however,
since various basic principles can be learnt and applied with little
effort and the dividends are often large!
In this chapter and the next we shall consider some of the basic
themes of the middle game, and students should study them with a
view to establishing which type of middle game is best suited to their
style of play. For instance Korchnoi is renowned for his liking of
positions with an isolated queen's pawn (IQP) , because such a posi
tion allows active play for his pieces. On the other hand some players
fanatically believe in the power of 'the two bishops' while others
enjoy playing 'minority attacks'.
We shall now attempt to give a brief breakdown and explanation of
some of the basic middle game themes, but it should be emphasized
that the keen enthusiast ought to refer to one of the many classic
middle game texts for a more thorough coverage.
Open Files and the Seventh and Eighth Ranks

Placing rooks on open files (files with no pawns) is one of the most
important principles in the middle game, because control of an open
file will often lead to penetration into the enemy position-namely
the seventh and eighth ranks. However it should be noted that
control of an open file does not guarantee a win; normally other
BAS I C M I D D L E G A M E F EATU R E S 57
advantages are required to add to the opponent's downfall.

Consider, for instance, the following strategical gem from Steinitz:
Diagram 28
Steinitz-Sellman
Baltimore
1 885
Steinitz holds all the trump cards, in that he has the knights (to be
favoured in closed positions) ; Black has very weak dark squares and
a dreadful white-squared bishop. Steinitz now hammers the last nail
into Black's coffin by taking absolute control of the open c-file. Play
continued-24 Rxc8 Qxc8 25 Rel Qb8 26 Qc2 ! Bd8 (Black attempts
to prevent penetration to the seventh) 2 7 Nac6 (Steinitz wants to
eliminate Black's ' good' bishop on dB when the dark squares will be
there for the taking; in so doing, he allows Black the chance to get rid
of his 'bad' bishop on a8, but White's infiltration on the c-file would
prove decisive, e.g. 2 7 . . . Bxc6 28 Qxc6 Nf8 29 Nxe6 ! Nxe6 30 Qd7 +
Be7 31 Rc6 ! Nf8 32 Rf6 + and Black is finished) 2 7 . . . Qb7 28 Nxd8 +
Rxd8 29 Qc7 ! Qb8 (Black avoids the exchange of queens which
would bring a rook to the seventh) 30 Bf2 Qb6 (to prevent 3 1 Bh4) 3 1
Nf3 Qxc7 32 Rxc7: the end i s very near now since Black hardly has a
plausible move. 3 Z . . . Ke8 (to free the black rook and to break the pin
along the seventh) 33 Ng5 Nf8 34 Bc5 Nd7 (to prevent 35 Re7 mate)
35 Bd6! resigns. Black is in zugzwang since all legal moves result in
drastic material losses. Steinitz managed to force victory in this
game due to several advantages , one of which was his domination of
the open c-file which eventually resulted in the penetration of one of
his rooks to the seventh rank. A rook on the seventh or eighth rank
will invariably cause havoc because of its ability to attack and harass
the enemy pawns .
Another example of the importance of controlling an open file is
58 BAS I C M I D D L E G A M E F E AT U R E S
Diagram 29
Alekhine-Yates
London
1 928
Alekhine to move, has several advantages in his favour (1) control of
the open file, (2) a well-placed knight which can attack on the dark
squares against a hapless bishop and (3) Black has vulnerable queen
side pawns on a4 and b4. Now watch Alekhine translate these
advantages into a full point! 20 Rfct Ba6 21 Ne5 ! (Black was hoping
to solve some of his problems by exchanging the rooks on the c-file
with 21 . . . Rac8, but this now loses a pawn to 22 Rxc8 Rxc8 2 3
Rxc8 + Bxc8 24 Nc6 ! when White threatens both 2 5 Nxb4 and 2 5
Ne7 + winning the bishop , and Black has n o effective way o f meeting
both threats) 21 . . . Reba 22 f3 ! (White cannot improve the position
of his rooks or knight at the moment, so he prepares to activate his
last piece-his king!) 22 . . . b3 23 a3 (since White controls the only
open file it is logical to keep all other files closed-that is 23 ab ab
would give Black chances of counterplay down the a-file) 23 . . . h6
(23 . . . Rc8 is still unplayable because of 24 Rxc8 + Rxc8 2 5 Rxc8 +
Bxc8 26 Kf2 Kf8 27 Ket Ke7 28 Kd2 Kd6 29 Kc3 Kc7 30 Kb4 Bd7
31 Nxd7 Kxd7 32 Kxa4 winning) 24 Kf2 Kh7 2 5 h4 Rf8 26 Kg3 Rfb8
(Black can do nothing but wait!) 2 7 Rc7-at last White's rooks make
their inevitable penetration to the seventh rank-2 7 . . . Bb5
28 Rt c5! (attacking the bishop to tie down the black rooks) 28 . . . Ba6
29 R5c6 Re8 30 Kf4 Kg8 31 h5 (White is in no hurry to double rooks on
the seventh since Black is helplessly tied up at the moment) 31 . . .
Bft 32 g3 Ba6 3 3 Rf7 Kh7 34 Rcc7 Rg8 3 5 Nd7 ! (White's threats along
the seventh rank now bring the game to a rapid conclusion-White
threatens 36 Nf6 + winning the exchange) 3 5 . . . Kh8 36 Nf6!
Rgf8-obviously not 36 . . . gxf6 ? ? 3 7 Rh7 mate-3 7 Rxg 7 ! ! Rxf6
(forced since White threatened 38 Rh7 mate. Alekhine now finds the
killer move!) 3 8 Ke5 ! resigns. Black cannot save his rook because if
BAS I C M I D D L E G A M E F E ATU R ES 59
either rook goes to fa there follows 39 Rh7 + Kg8 40 Rcg 7 mate.

It should be noted that an open file is sometimes of little signifi
cance if both players are capable of contesting the file, and no other
advantages are evident. The diagrammed position illustrates the
point well:
Diagram 30
Capablanca-Maroczy
Lake Hopatcong
1 926
The position arose from an exchange variation of the French
Defence, and this virtually symmetrical pattern petered out to
dead-pan equality as follows-1 5 Rxe8 + Rxe8 1 6 Bxg6 hxg6 1 7 Rel
when the exchange of the last pair of rooks renders the open file
useless. The game continued 17 . . . Rxe1 + 1 8 Nxe1 Ne8 19 Nd3 Nd6
20 Qb3 a6 21 Kh !-!.
The maxim 'Rooks should be placed on open files' also applies to
semi-(half) open files. These are files with only one pawn on them
and in many cases the pawn in question becomes an object of attack
for the enemy rooks. Consider the following example:
Diagram 31
Povah-Nielsen
Jersey
1 977
White enjoys pressure o n the half-open d-file against Black's weak

backward pawn. By intensifying this pressure White is able to tie
Black up and then switch his attention to Black's vulnerable queen
side pawns-2 3 Rd5 Rd8 24 Rfdt Qe6 (White threatened 2 5 Rxe5!
dxe5 ? 26 Nxe7 + Rxe7 27 Rxd8 mating) 25 Qd3 Rad7 (an amusing
position with all eight pieces concentrating on the d6 square, but
now White can strike on the queen side) 26 a4 ! g6 (26 . . . bxa4 2 7
Qxa6 wins a pawn, since the black pawn o n a4 must fall; attempts at
defending the b-pawn such as 26 . . . Rb8 leave the d-pawn inade
quately defended) 2 7 Nxe7 + Qxe7 28 axb5 axb5 29 Rxb5 with a clear
advantage. White's advantage in this position was due to his perma
nent pressure on the half-open d-file and the fact that Black had a
vulnerable backward pawn which required support from his pieces.
However, there are some half-open files which prove to be of little
use to either player, consider Diagram 3 2 :
Diagram 32
Tartakover-Sultan Khan
1 2th Match Game
1 931
White controls the half-open g-file, although it is of little practical
value since he cannot increase the pressure against g7, nor can he
open the. g-file. On the other hand, the semi-open d-file offers Black
good attacking prospects since he can increase pressure against dz
with the doubling of his major pieces on the d-file and a timely . . .
Bf4. Furthermore Black controls useful squares .on the file such as d3
and d4, and he can also open up the adjacent c-file with . . . b5. The
game continued 1 . . . Bd4 2 Bdt Qe5 3 Ba5 Rd e8 4 Bc3 b5 5 cxb5 axb5
6 f3 e3 ! 7 Bxd4 cxd4 8 Kbt Nd5 when Black had a clear advantage.
Knight Outposts and the 'Good' Knight versus the 'Bad' Bishop
Although when first starting to play chess we are told that a knight
BAS I C M I D D L E G A M E F E ATU R E S 61
and a bishop are both worth approximately three pawns each, we

soon learn that there are certain positions when we can favour one
piece or the other. As a general rule we can say that the bishop
prefers open positions with few pawns on the board, while the
knight tends to prefer closed positions with many pawns on the
board.
Knights are particularly useful in crowded middle game positions
if they can reach squares which are unassailable from enemy
pawns-such squares we refer to as ' outposts'. The diagrammed
position shows White with two well-placed knights in the centre of
the board. The knight on e5 can only be removed by one of the black
minor pieces , while the knight on f4 can be driven away by . . . g5;
but this would leave Black's king-side permanently weakened:
Diagram 33
Korchnoi-Bellin
Hastings
1 975/6
As we know from the previous section White enjoys yet another
advantage in this position-his control of the only open file. In
addition to these advantages , Black's pieces are rather cramped and
poorly co-ordinated, consequently it should come as no surprise to
find that White has a powerful combination-Korchnoi played 19
Qxc7 ! Bx d 20 Rxc7 Rfd8 (perhaps Black should try 20 . . . Bc8 when
White would have the interesting 2 1 Nxd5! exd5 2 2 Bxd5 + Kh8 2 2
Bxa8 leaving White with rook, bishop and two pawns , and an active
position for the queen; alternatively 2 1 . . . Qd6 2 2 Rc6 Qb8 23 Rxc8!
24 Nxd7 winning masses of material. However, 2 1 . . . Qd8! appears
relatively best for Black, e.g. 22 Nxd7 Bxd7 23 Rxd7 Qxd7 24 Nb6 Qa7
25 Nxa8 Rxa8 26 Bxa8 Qxa8 when Black should not be losing) 2 1
Rxb7 Rac8 2 2 Nfd3 (White reinforces his knight on e 5 , and threatens
to bring the other knight in on c5-another strong outpost) 2 2 . . . a3
62 BAS I C M I D D L E G AM E F E AT U R ES
23 Bel Qd6 ? 24 Nxd 7 Rxd7 25 Bxa3 ! Qxa3 26 Rxd7 Rc2 2 7 Bfl g5

28 Rb7 h5 29 Rxb5 h4 30 Rc5 Rd2 3 1 Rae l hxg3 32 hxg3 Kg 7 3 3 Rc 7 +
Kh6 3 4 Rlc6 1-0.
Yet another example of a knight occupying a strong outpost was
Diagram 34
Gaprindashvili-Grinfeld
Tbilisi
1 976
The women's ex-world champion has a very powerful knight posted
on c5, and she now embarks upon a manoeuvre to win absolute
control of the a-file---3 1 Rxa7 Rxa7 32 Nd3 Rb7 (forced in view of the
threat of 33 Bxd5 and 34 Nxb4) 33 Rat Bd7 34 Nc5-the knight now
returns to its strong outpost-34 . . . Rc7 35 Ra8 + Ke7 36 Bxd5!
(White eliminates the well-placed knight in order to render the
b-pawn defenceless) 36 . . . exd5 3 7 Rb8 Bg4 38 Na6 Ra7 39 Nxb4 Kd6
40 Rea (controlling the important open e-file) 40 . . . c5 41 dxc 5 +
Kxc5 42 Nd3 + Kd4 1-0 Black loses further material after 43 Ne5 B d 7
4 4 R e 7 + - . This example demonstrated the superiority o f a 'good '
knight over a 'bad' bisho�the knight could occupy several key
squares such as c5, d3 and e5 without fear of being disturbed,
while the bishop was hemmed in by his own pawns for most of
the game, and even when he did escape he never actually achieved
anything.
There are many examples throughout chess literature of the 'good '
knight versus the 'bad' bishop, but it is not within the scope of this
book to give more than a few examples. The keen student would be
very well advised to study Master games which can be found in
countless chess books in order to identify the skilful way in which
the Master conducts an attack or utilizes some positional advantage,
such as the 'good knight versus bad bishop' situation.
BAS I C M I D D L E G A M E F EATU R E S 63
The 'Two Bishops' and the 'Good' Bishop versus the 'Bad' Knight
In the last section we considered the superiority of a knight over a
·
bishop in a closed position, but as positions open up (i. e. more
pawns are exchanged) the power of the bishops tends to
increase--hence the reason why a bishop is frequently preferred to a
knight in an ending with pawns on both sides of the board. The
bishop being a long-range piece is capable of monitoring both sides
of the board from one square, while a knight cannot effectively
support play on both wings at the same time. Two bishops working
together can offer a potent advantage especially in the middle game,
assuming of course that the bishops can locate useful diagonals.
Our first example is the classic game Rosenthal-Steinitz, Vienna
1873, which is thought to be one of the earliest examples illustrating
the power of the two bishops :
Diagram 35
Rosenthal-Steinitz
Vienna
1 873
Steinitz has just played 14 . . . Nc4 and Rosenthal played 1 5 Bxc4

Qxc4 ceding the bishop pair. Black's advantage is not decisive as yet
since the position is only semi-open and the white minor pieces have
support points (outposts) at d4 and e5. Steinitz now embarks upon a
series of pawn moves to gain control of these squares and so enhance
the scope of his own bishops. However, it should be noted that
accurate judgement is needed to ensure that the pawn moves, which
are frequently weakening, are adequately compensated for by the
increased scope of the bishops and the restricted activity of the
enemy pieces. Play continued 16 Qf2 c5! 1 7 Nf3 b6 (Black has driven
the white knight back and now he prepares to activate his powerful
white-squared bishop) 18 Ne5 Qe6 19 Qf3 Ba6 20 Rfet f6 another key
64 BAS I C M I D D L E G A M E F EAT U R E S
outpost i s rendered uninhabitable! 2 1 Ng4 h5-and yet again the

knight is pushed back. Such · a move might well be criticized as
weakening, but it is justified because of the resulting immobility of
the white pieces. 22 Nf2 Qf7 (preparing to seize the long white
diagonal with . . . Bb7) 23 f5 g5 24 Radl Bb7 25 Qg3 Rd5 ! Black
ignores the a-pawn but goes for the vulnerable f-pawn 26 Rxd5 Qxd5
27 Rdl (27 Qh3 fails to 27 . . . g4 when the f-pawn falls) 27 . . . Qxf5
28 Qc7 Bds 29 b3 Re8 30 c4 Bf7 3 1 Bel Re2 !-Black occupies the
seventh rank with deadly threats-3 2 Rfl Qc2 (threatening 33 . . .
Rxf2 34 Rxf2 Qxcl + when the two bishops will overwhelm the rook)
33 Qg3 Qxa2 0-1 .
Another example demonstrating the superiority of the two
bishops is seen in Diagram 36:
Diagram 36
Tscheshkovsky-Svesh
nikov USSR
1 978
White threatens mate but Sveshnikov was able to parry this threat
and offer the exchange of queens since the tw!) bishops would give
Black the better ending. The game continued 23 . . . Qc5 24 Qxa6 ?!
(White snatches a pawn since 24 . . . Bxd5 is answered by 2 5 Rxd5
Qxd5 26 Qxc8 + with the advantage. Black, however, takes the
opportunity to castle.) 24 . . . 0-0 2 5 Nce3 Rfd8 26 Qb7 Rb8 27 Qc7
Qxc7 28 Nxc7 Rxdl + 29 Nxdl c3 ! + (Black opens up the position for
both his rook and white-squared bishop. 30 Nxe6 c2 is decisive as
Black queens; while 30 be Bc4 + 3 1 Kel Rc8 wins the white knight
which is trapped by the white-squared bishop) 30 Nxc3 Bc4 + 3 1 Kgl
Rxb2 (Black's rook has penetrated to the seventh rank and his two
bishops, plus White's poorly co-ordinated pieces, ensure a clear
advantage) 32 g3 Rc2 33 N3d5 Bd3 34 Ne8 Rel + (not 34 . . . Be4?
BAS I C M I D D L E GAM E F E ATU R E S 65
3 5 Nef6 + + -) 3 5 Kg2 Be4 + 36 f3 Rc2 + 3 7 Kh3 Bxd5 0-1 Black's extra

piece is decisive.
The power of the two bishops is further illustrated in the next
example where Black loses material, but finds that his bishops on
adjacent diagonals (often referred to as 'Horwitz Bishops') , coupled
with a powerful advanced passed pawn, provide adequate counter
play.
Diagram 37
D. J. Sewell-Povah
Surrey Under-1 6
Championship
1 968
White bar: just played 2 5 Bc5 attacking the black rook, and Black
answers with 25 . . . Rea ? (25 . . . Rd8 would avoid the loss of the
black queen) 26 Rd6 ! (the black queen is amazingly trapped in the
middle of the board!) 26 . . . Be6! (the white-squared bishop takes his
opportunity to come into the game on the useful long diagonal from
ht-a8) 2 7 Rxd5 Bxd5 28 Ra3 ?! The rook is poorly placed on this
square--it was better to keep the rook on the b-file. 28 . . . exf4 29
Ne2 ? (It is understandable that White should want to activate his
knight, but he should neutralize the power of Black's two bishops
with 29 Bd4 first, and then play Nf3 blockading the f-pawn.
The immediate 29 Nf3 would be met by 29 . . . Rac8 when Black
would still have dangerous threats for his material deficit) 29 . . .
f3 30 Ngt ? (30 Ng3 was better but Black still has dangerous counter
play, e.g. 30 . . . Be5 31 Kgt Rac8 32 d4 Bf4-intending 33 . . . b6
34 Bxb6 Rel + 35 Kh2 Re2 - +-therefore 33 Rc3 b6 34 Bb4 Rxc3
35 Bxc3 Be3 - + ; alternatively, 31 Bd4 Bf4 32 Kgt-to stop 32 . . .
Re2 - +-32 . . . Rac8 3 3 Rc3 ? Rxc3 34 Bxc3 Be3 - + Notice how
Black's bishops and rooks co-operate superbly) 30 . . . Be5 + 3 1 Kht
Bg3 ! !
Diagram 38
Sewell-Povah
This bishop sacrifice decides the game in Black's favour. 32 Qd4 ! (If
32 Qxg3 f2 + 33 Nf3 fl = Q + followed by 34 . . . Qxf3 leaves Black a
rook up , while 33 Kh2 ? fl =N! ! mate! ) 32 . . . f2 + 3 3 Qxds fl = Q with
an easy win for Black who has rook and two pawns for a knight.
We now conclude this section with a complete game from one of
the strategical masters, ex-world champion Vassily Smyslov. The
whole game is given, because the study of Master games is one of the
best methods of improving your game. Most of the contemporary
Masters and Grandmasters of today have studied the games and
playing styles of the former greats . The keen student should cover up
the page and try to guess Smyslov's moves-in this fashion the game
will be more instructive.
I. Bondarevsky-V. Smyslov, Moscow, 1946
1 e4 es 2 Nf3 Nc6 3 Bbs a6 4 Ba4 Nf6 5 0-0 Be7 6 Bxc6 . This
delayed-exchange variation of the Ruy Lopez gives Black the two
bisho ps as compensation for his doubled pawns . 6 . . . dxc6 7Re1
Nd7 8 d4 exd4 9 Qxd4 0-0 10 Bf4.
Diagram 39
Bondarevsky-Smyslov
Moscow
1 946
BAS I C M I D D L E G A M E FEATU R ES 67
Now cover up the page and try to guess Black's moves and see if you
can play as well as Smyslov!
10 . . . Nc5
Black does not mind the exchange of queens since his two bishops
will be very powerful in the ending.
11 ' QxdB
11 . . . BxdB
Naturally not 11 . . . Rxd8 which loses a pawn to 12 Bxc7:

12 Nc3
12 . . . f5!
Srnyslov is trying to open up the position for his pieces, 13 exf5 Bxf5
would activate both his white-squared bishop and the rook on
fB.
13 es
This reduces the scope of White's pieces , Srnyslov suggested 13 Bg5
Bxg5 i4 Nxg5 h6 1 5 b4 Nxe4 1 6 Ngxe4 fxe4 1 7 Nxe4 Bf5 with
approximate equality.
13 . . . Ne6!
As Nimzovich says 'the passed pawn is a criminal who should be

kept under lock and key'. Knights are very good at blockading
because they still exert control over their normal range of squares.
14 Bd2
14 . . . g5!
This pawn advance threateping to win the white knight by . . . g4 is
characteristic of Steinitz's policy-to use the pawns to harass the
enemy pieces and to enhance the scope of the two bishops.
1 5 Ne2
Intending to meet 1 5 . . . g4 with 16 Nfd4 dislodging Black's power
ful knight on e6.
15 . . . c5
Smyslov is logically using his doubled c-pawns in order to renew the
threat of . . . g4.
68 BAS I C M I D D L E GAM E F E ATU R ES
16 Bc3
Preparing a flight square for the knight on d2.
16 . . . b5
Another powerful pawn advance, with the threat of t 7 . . . b4 forcing
18 Bd2 when 18 . . . g4 wins the embarrassed knight on f3 .
17 b3
17 . . . Bb7
18 Ng3
18 . . . g4
19 Nd2
19 . . . Be7
Smyslov prepares to centralize his king and bring his rooks into
play. 19 . . Bg5 threatening . . . b4 also looks promising.
.
20 Nh5
White intends to use the f6 square, which has been weakened by the
advance of Black's g-pawn.
20 . . . Kf7
21 Nft
21 . . . Kg6!
Black encourages the invasion on f6.
22 Nf6
The alternative 2 2 Ng3 would be met by 2 2 . . . hs followed by . . h4
.
when White's minor pieces have been overwhelmed by Black's

advancing hordes.
22 . . . Rad8
Occupying the open d-file and ensuring that White's active knight
on f6 has no retreat.
23 Radl
White challenges for control of the d-file.
23 . . . Rxdl
24 Rxdl
24 . . . Rd8
BAS I C M I D D L E GAM E F E ATU R E S 69
White must not be allowed to dominate the file.
25 Rxd8 Bxd8
White now has problems since he must guard against 26 . . . Nf4

threatening both 2 7 . . . Ne2 + and 2 7 . . . Nxg2.
26 Ne3
Black forcefully refutes this move.
Diagram 40
Bondarevsky-Smyslov
26 . . . f4!
27 Ndt
The tactical justification is that 2 7 Nexg4 fails to 2 7 . . . hs winning

the helpless knight, while 2 7 Ned5 c6 is no better.
27 . . . Bxf6 !
Smyslov gives up one of his bishops in order that he might win a

pawn.
28 exf6
28 • Be4
• •
The point of Black's previous move is that the knight no longer

guards this square, and the white c-pawn is indefensible.
29 Bb2
29 . . . b4!
70 BAS I C M I D D L E GAM E F E AT U R E S
Although Smyslov could snatch a pawn with 2 9 . . . Bxc2 he notic.es

that the white knight is decidedly short of squares-now . . . Bxc2 is
an even more potent threat since the white knight is virtually
immobilized.
30 f3
30 . . . Bxc2
31 Nf2 gxf3
32 gxf3
32 . . . Bb1
Black's 29 . . . b4 immobilized not only the white knight, but also

White's queen-side pawns and White must lose yet another pawn
since 33 a3 bxa3 34 Bxa3 loses the f6 pawn.
33 Ne4 Bxa2
34 Nd2
With .this move White imprisons the black bishop, but Smyslov
intends to get his bishop out on bail!
34 . . . a5!
Intending to free the bishop at the appropriate moment with . . . a4 .

35 Kf2
35 . . . Nd4
36 Bxd4 cxd4
White had little choice but to_ capture on d4 , otherwise he would
have lost his b-pawn, e.g. 36 Ne4 Nxb3 and if 37 Ng5 (threatening 3 8
f7 ! winning) then Black simply plays 3 7 . . . Nd4 38 Ne4 Ne6 when
White is finished.
37 Ke2
37 . . Kxf6
.
38 Kd3
38 . . . Ke5
39 Kc2
39 Kc4 a4 40 Kxb4 Bxb3 is no better for White.
39 . . . a4
Smyslov chooses to free his bishop while he still can-White

threatened 40 Kb2 .
BAS I C M I D D L E G A M E F E ATU R E S 71
40 bxa4
40 . . . c5
41 a5
41 . . c4•
42 a6
42 . . d3 +
.
0-1
The three connected passed pawns plus the advantage of bishop
versus knight decide the issue quickly. The finish could have been
43 Kb2 c3 + 44 Kxa2 cxdz 45 a7 dt = Q 46 a8 = Q b3 + 47 Kb2 Qc2 + 48
Ka3 (48 Kat Qct is mate) 48 . . . Qa2 + winning the white queen and
the game.
While we are on the subject of studying Master games, it should be
mentioned that a lot can be gained by discovering a Master or
Grandmaster whose style of play you admire and then trying to
emulate it. Such trends in style of play are evident throughout the
history 'of chess. The ex-world champion Bobby Fischer greatly
admired Morphy and Capablanca and his own style of play was often
likened to theirs. Similarly the American Grandmaster Walter
Browne adopted a similar opening repertoire to Fischer and tried to
emulate his style of play.
Because the middle game is that stage of chess when a player can
be most creative, it follows that this is the period when his style of
play will be most evident. Consequently, studying complete Master
games should reveal how top-class players handle certain types of
position, and the average player should get an insight into a variety
of middle game policies , such as how to play against certain pawn
configurations, or how to make the best use of a bishop versus a
knight, etc. The middle game in chess is full of different positions
with new and exciting problems, and an effective way of meeting
this never-ending challenge is the continual study of Master games!
Pawn Structu res and
Pawn Weaknesses
'Pawns are the soul of Chess' wrote Philidor, and he was right!
Although the pawn is the lowliest unit on the chess-board, its influ
ence upon the course of a game is far greater than might at first be
imagined.
Many people think that Philidor's famous quote refers to the
ability of the pawns to promote-a form of reincarnation, in which
the analogy with the soul is clear. His analogy is more subtle than
this, however, for, if we consider the soul as being the invisible
underlying influence upon a human being, then we can draw a
comparison with the pawn on the chess-board. Collectively the
pawns are capable of exerting a powerful influence on the course of a
game, which is not always immediately obvious to the lesser player.
The reason for this influence is that pawn structures tend to be static
by nature-that is, they do not change very quickly, so weak squares
and weak pawns are very much in evidence.
We shall now consider some of the special features in pawn
structures, paying particular attention to pawn weaknesses.
DOUBLED PAWNS
The theory of doubled pawns is very complicated, for one must

consider: are they isolated, mobile or static, part of a pawn majority
or minQrity,, etc?
Generally, doubled pawns confer some form of disadvantage , with
varying degrees of seriousness. The main advantage gained from
having doubled pawns, is the open or half-open file that is inevitably
created for the rooks.
We now consider a few examples of doubled pawns in middle
game positions; it is worth mentioning that doubled pawns are
PAWN STR U CTU R E S AN D PAWN W E AK N E S S E S 73
sometimes an important feature of certain opening variations. For

instance, consider the Ruy Lopez exchange variation given on page
33; White gives up the bishop pair in order to double Black's
c-pawns; the position tends to be roughly balanced with Black's two
bishops providing compensation. Another opening variation,
featuring doubled pawns , occurs in the Caro-Kann after the moves 1
e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Nxf6 when Black can capture
with either his e- or g-pawn. The capture 5 . . . exf6 gives the diag
rammed· position:
Diagram 41
The doubled pawn has the effect of providing White with a queen
side pawn majority (with a potential passed d-pawn) , while Black
seems to have very little in the way of compensation, except for free
development for his pieces.
Our first middle game example illustrates the danger of doubled
pawns in the vicinity of the king :
Diagram 42
Nunn-Mestel
Hastings
1 977/8
74 PAWN STRU CTU R E S A N D PAWN WEA K N E S S E S
The diagrammed position was reached i n a Sicilian Defence, Dragon

variation, between two of England's leading players. Jonathan Mes
tel now played a standard exchange sacrifice (rook for minor piece)
in order to give White doubled isolated pawns on the c-file, and to
denude the white king. The game continued t2 . . . Rxc3 ! t3 bxc3
Qb6 + t4 Kat QcS tS Qd2 (the queen comes back to defend the
shattered queen side, but this allows Black to castle and mobilize his
remaining rook. A better try was ts Kb2 when Black can play for a
dra\f if he wishes by t S . . . Qb6 + t6 Kat-self-pinning with t 6
Nb3 ? would invite t 6 . . . as followed b y . . . a4-t6 . . . Qcs t 7 Kb2
Qb6 + repeating moves) . ts . . . 0-0 Black has definite compensa
tion for his material deficit, because of his long-term prospects
against White's permanently vulnerable queen side. 16 Rbt b6 t 7
Bbs Rea t a Bxd7 Nfxd7 t9 f4 Nc4 the knight occupies a powerful
outpost created by the exchange sacrifice, since the absence of
White's b- and .d-pawns make the c4 square unassailable. 20 Qct a6
2 t h4 es Black dislodges White's centrally placed knight 22 fxes
dxes 23 Ne2 Qf2 24 Qdt (24 Qft Qxft 2 S Rbxft Ncs gives Black a
good ending) 24 . . . NcS 2S hs White is hoping to create threats
down the h-file 2S . . . Ne3 26 Qd2 Nxe4 2 7 Qd3 Nxc3 ! 28 hxg6 (28
Nxc3 fails to 28 . . . Nxc2 + 29 Kb2 Nb4 + 30 Qe2 Nd3 + 3 t Kat Qd4
- + ) 28 . . hxg6 29 Rxb6 e4 (White was threatening 30 Rxg6 + fxg6
·
3t Qxg6 + with a mating attack.) 30 Qxa6 (30 Rxg6 + fxg6 3 t Qd7

Nxc2 + 32 Kb2 Na4 + is hopeless for White; Mestel now finishes the
game with a neat combination 30 . . . Nxc2 + 3t Kb2 Qxb6 + ! 0-t (32
Qxb6 is answered by 3 2 . . . Na4 + when Black emerges with an extra
piece and pawn) .
The next exaipple comes from the Catalan opening, and illustrates
the serious problems posed by the weakness of isolated doubled
pawns on a· semi-open file:
Diagram 43
Lombard-Rogoff
Biel lnterzonal
1 976
PAWN STR UCTU R E S A N D PAWN W EAK N E S S E S 75
Rogoff played 6 . . . Ne4 applying pressure to the c3 square. White

cannot avoid doubled c-pawns now. The game continued 7 Qd3 Qa5
(adding to the pressure against c3, and intending to answer 8 Qxe4
with Bxc3 + 9 bxc3 Qxc3 + winning the rook on al) 8 Nc2 Bxc3 + 9
bxc3 Nc5! (as in the previous example, the knight occupies a strong
outpost, blockading the c-pawns) 10 Qd2 b6 1 1 Bg2 Bb7 12 Bxb7
Nxb7 13 Ba3 Nc6 14 CHJ d6 ! 1 5 Rab1 (if 1 5 Bxd6 then 1 5 . . . �!
wins the pinned bishop) 15 . . . Qa4 16 Rfd1. (The c-pawn is inde
fensible since 16 Ne3 ? loses the bishop on a3 and 16 Qd3 is met by
. . . Rd8 and . . . Nc5 driving the queen away) 16 . . . CHJ 1 7 Rb5 Qxc4
18 Rg5 f6! (White intended to play-19 Rxg7 + Kxg7 20 Qg 5 + Kh8
21 Qf6 + , etc., with a draw by perpetual check) 19 Rh5 Qxa2-yet
another pawn falls and Black had little trouble in winning in a few
more moves. The weak c-pawns always promised to be a handicap
for White.
The reader should search through Master games for examples of
how to handle or how to play against doubled pawns. The following
game illustrates some of the problems that arise from having a
doubled pawn.
Nisman-Karpov: Soviet Team Championship 1 968

1 c4 Nf6 2 Nc3 e6 3 d4 Bb4 4 a3 Bxc3 + 5 bxc3 c5 6 e3 Nc6 7 Ne2 b6 8
Ng3.
Now cover the page and try to guess Black's moves:
Diagram 44
Nisman-Karpov
Karpov has blockaded the double pawn in standard fashion and now
he uses a common method of attacking the pawns.
76 PAWN STR U CT U R E S A N D PAWN W E AK N E S S E S
8 . . . Ba6
9 Bd3
9 . . . Na5
10 Qe2
White defends the pawn adequately but his position is very passive.
10 . . . d6
Black prepares to set up a central pawn wedge.

11 Bb2
11 . Qd7!
. .
The Queen is well placed on this square to add to the pressure

against c4 with a timely . . . Qa4.
12 e4
White threatens 13 e5 vacating the e4 square for his pieces.
12 . . . 0-0-0!
A clever defence to 13 e5 which simply loses a pawn to 13 . . . dxe5.

The alternative 12 . . . e5 would allow the invasion 1 3 Nf5 .
13 a4
White rules out a future . . . Qa4.

13 . . . h5
Karpov wishes to play . . . e5 fixing the centre, without having to
worry about Nf5 , so first he plans to dislodge the white knight.
14 0-0 h4
15 Nh1 e5
16 £4
16 . . . Nh5!
The f4 square is vulnerable especially as 1 7 g3 is very unappealing.
17 Bc1 loses a pawn to 17 . . . Nb3 18 Rb1 Nxc1 19 Rbxc1 Nxf4 .
17 fxe5 dxe5
18 d5 Nf4
White has a protected passed pawn, but it is effectively supervised,

and the white central pawn mass is a stationary target. Queen moves
PAWN STRU CTU R ES A N D PAW N W EAK N E S S E S 77
would now lose the c-pawn, so White gives up the exchange in order
to activate his pawns, but the weakness of the doubled c-pawns is a
great handicap!
19 Rxf4 exf4
20 e5
20 • h3!
• •
Karpov wishes to prise open the white king position.

21 g4 KbB
White threatened 22 Bf5 , and this move sets a cunning trap.
22 Bet?
Nisman tries to activate this poorly placed bishop with 2 3 Bxf4
which looks very appealing; while 22 . . . g5 invites the knight
manoeuvre 23 Nf2 - e4 - f6. The move played allows a decisive
combination:
Diagram 45
Nisman-Karpov
22 . . . Bxc4!
23 Bxc4 Nxc4
24 Qxc4 Qxg4 +
The white queen has been deflected from the defence of the g-pawn,
and the point of the combination is that after 2 5 Kf2 Qg2 + wins the
knight, which is no longer defended because of the Bet move.
Therefore . . .
25 Ng3 Rh4!
78 PAWN S T R U C TU R E S A N D PAWN WEAK N E S S E S
The only move t o ensure the win of the piece.

26 Qxf4 Qxf4
2 7 Bxf4 Rxf4
with an easy win for Black.
ISOLATED PAWNS
The isolated pawn-a pawn without a pawn of its own colour on a

neighbouring file, is normally considered to be a weakness, simply
because it cannot enjoy the protection of one of its colleagues. There
are numerous examples in chess literature of an isolated pawn fall
ing by the wayside and resulting in the loss of the game. Consider the
following example:
Diagram 46
D. J. King-Povah
Surrey Trophy
1 978
Although Black's four isolated pawns are adequately guarded at the
moment, the strain soon proved to be too much for the author, who
had the misfortune to have this position against one of his pupils,
whom he had evidently taught too well! White played 22 Qd3 Raes
23 Bf2 ! (the 'bad' bishop comes to life with the annoying threat of 24
Bh4 and 25 Bxf6 undermining Black's defence of the f-pawn) 23 . . .
Ng6 24 Bg3 h5 (Black tries to lash out before White can play 2 5 Ne3)
25 Rbft KfB 2 6 Ne3 h4 27 Bet Bh6 28 Ne2 Black is now devoid of
counterplay and his position falls apart at the seams. 2 8 . . . Qa6 Nxa6
30 Rg2 White does not rush to capture the hapless f-pawn, but
guards his second rank 30 . . . Ke8 31 Nxf5 h3 32 Rgf2 Bf8 33 Ne3
1-0.
PAWN S T R U C T U R E S A N D PAWN W E AK N E S S E S 79
Our next example is an impressive Master game between two

contemporary giants of the chess-board.
Portisch-Spassky, Candidates Match, Geneva 1977

1 d4 d5 2 c4 dxc4 (the Queen's Gambit Accepted, which is normally
overshadowed by the more popular declined variations, which start
with 2 . . . e6) 3 Nf3 Nf6 4 e3 Bg4 5 Bxc4 (White regains his pawn with
a slight initiative, since he has a two-one central pawn majority, and
he should be able to increase his central control with an inevitable
e4) 5 . . . e6 6 Nc3 Nbd7 7 0-0 Bd6 8 h3 Bh5 9 e4 e5 10 Be2 ! 0-0
11 dxe5 Nxe5 12 Nd4! Bg6 13 Bg5 Rea 14 Ndb5 Nc6 (not 14 . . . Be7?
1 5 f4 winning a piece because ofthe threat 16 f5) 1 5 Nxd6 cxd6 16 f3 ! ± .
Diagram 47
Portisch-Spassky
Candidates' Match
Geneva
1 977
Portisch has inflicted an IQP (Isolated Queen's Pawn) upon his

opponent, and he now increases the pressure so that Spassky has to
make further concessions if he is to maintain counter-chances. It
should be noted that White has two other advantages in the dia
grammed position; first he has 'the two bishops' and second Black's
bishop is a very poor piece indeed. Play continued-16 . . . h6 1 7
Bh4 Bh7 ( 1 7 . . . Qb6 + fails t o 18 Rf2 Qxb2 ? 1 9 Bxf6 gxf6 20 Nd5 + -)
18 Rf2 g5 (Black breaks the annoying pin on the h4-d8 diagonal, but
weakens his king side in the process) . 19 Bg3 Qb6 20 Bc4! (White
does not rush to take the d-pawn, but protects his b-pawn while
activating his white-squared bishop) 20 . . . Ne5 2 1 Bb3 Rac8 22 Qf1 !
(threatening 23 f4) 22 . . . Kg7 2 3 Qb5 ! + - (Black cannot exchange
queens , since 2 3 . . . Qxb5 24 Nxb5 forks two pawns; neither can he
allow 24 Qxb6 inflicting further pawn weaknesses) 23 . . . Qc7 24
Rd1 Nh5 2 5 Bh2 (not 2 5 Bxe5 dxe5 ridding Black of his IQP) 2 5 . . . b6
80 PAWN STRU CTU R E S A N D PAWN WEAK N E S S E S
26 Rfd2 Portisch piles u p o n the weakness 2 6 . . . Qc5 + 2 7 Khl Nf4

28 Rxd6 Qxb5 29 Nxb5 Re7 30 Nc3 White with an extra pawn and the
better position consolidates, before rapidly ·concluding the game 30
. . . Nfg6 3 1 Nd5 Rees 32 Ne3 heading for the powerful outpost on f5
32 . . . Nh4 33 Bxe5 Rxe5 34 Rd7 Bg6 35 Nc4! Rec5 36 Nd6 RfB 37 Rxa7
another pawn falls and the end is in sight 37 . . . h5 38 Kgl g4 39 f4
Rd8 1-0 Spassky resigned before Portisch could play 40 f5 driving
the bishop away from his defensive duties.
Notice in this game how the permanent weakness of the d-pawn
forced Black to seek counterplay with his king-side pawn advances ,
which only increased his problems. Portisch never allowed Black
any counter-chances, and captured the IQP at his leisure. It would be
fair to say that the result was never in doubt from move 16 onward.
The isolated pawn and particularly the IQP however, need not
always be a disadvantage. It is true that the isolani will become
weaker as the game progresses towards an ending: certain middle
game positions, however, are enhanced by the presence of an IQP
who can control important central squares. Consider the following
example from one of the IQP's leading exponents Victor Korchnoi:
Diagram 48
Korchnoi-A. Zaitsev
30th USSR Champion
ships 1 962
Zaitsev played 1 1 . . . Nfd5 in order to blockade the IQP before

applying pressure to the weakness. However, Korchnoi demons
trates in masterly fashion how the IQP facilitates central control and
offers good king-side attacking prospects. Another instructive game
from this position which the reader might wish to refer to, is Botvin
nik-Vidmar, Nottingham, 1936.
Korchnoi continued 12 Bxe7 Qxe7 1 3 Rel Rd8 14 Rel notice how
PAWN STR U CTU R E S A N D PAWN WEAK N E S S E S 81
White's rooks have occupied the open and semi-open files either
side of the IQP 14 . . . Nf6 1 5 Qe2 Qb4 16 Ne4 Korchnoi is trying to
remove Black's solitary king-side defensive piece--the knight on
f6-16 . . . Nbd5 1 7 Nc5 Qa5 (Attempts to displace White's dominant
knight from c5 with 1 7 . . . b6 are extremely dangerous, e.g. 18 Rc4
Qb5 19 Ba4 Qa5 20 Bc6 threatening 21 Ra4, etc.) 18 Ne5 :
Diagram 49
Korchnoi-Zaitsev
Almost an ideal position with an IQP ! Although Black has the pawn
firmly blockaded, White's pieces are occupying such dominant
posts that further resistance is short-lived. 18 . . . Qb6 19 Qf3 Rf8
20 g4 White's position is so overwhelming that he can afford to weaken
his king side in order to gain more space on that flank 20 . . . Qd8 2 1
g5 Ne8 2 2 h4 f6 2 3 g 6 ! a neat tactic which forces further weaknesses
in Black's position 23 . . . h6 (23 . . . fxe5 fails to 24 gxh7 Kxh7 25 Qxf8
+ - ; while 2 3 . . . hxg6 24 Nxg6 Rf7 2 5 Qh5 is no better) 24 Nf7 Rxf7
(the alternatives are unappealing, e.g. 24 . . . Qe7 ? 2 5 Bxd5 + - or 24
. . . Qb6 2 5 Nxe6 Bxe6 26 Rxe6 Qxe6 2 7 Bxd5 with a decisive attack)
25 gxf7 + Kxf7 26 Nxe6 ! Bxe6 2 7 Rxe6 Kxe6 28 Qe4 + Kf7 29 Bxd5 +
1-0. After 29 . . . Kf8 30 Qe6 Nd6 3 1 Rel White has the unstoppable
threat of Qg8 mate.
To summarize the relative merits of an IQP, we can say that in an
endgame it will generally prove to be a weakness, while in a middle
game it can often confer certain advantages, such as allowing greater
piece mobility and providing support points at e5 and c5. Therefore
it should come as no surprise to find that many players willingly opt
for the IQP, for as Tarrasch has said: 'Before the endgame the gods
have placed the middle game ! '
82 PAWN ST R U CTU R E S AN D PAWN W EAKN E S S E S
BACKWARD PAWNS
The 'backward' pawn is a pawn which, despite having one or two

colleagues on the adjacent files, cannot enjoy their protection , sim
ply because he has been left behind. It is the square in front of the
backward pawn that causes most of the problems because it tends to
fall under enemy control. In this respect the backward pawn is
similar to the isolated pawn, except it is invariably considered a
weakness without offering any counterbalancing advantages.
Our first example of a backward pawn is a fairly typical case
arising from the Sicilian Defence:
Diagram 50
Fischer--Gadia
Mar del Plata
1 960
A quick assessment reveals that White enjoys a spatial advantage

and a slight lead in development. But there is another very important
factor in this position, namely Black's backward pawn on d6 . This
weak pawn not only restricts the mobility of Black's dark-squared
bishop, but it is going to require piece-protection for the rest of the
game. If Black could liquidate his backward d-pawn with . . . d5 then
he will have solved most of his problems, hence White concentrates
his forces on the weak d5 square which promises to be an excellent
outpost for White's pieces. Notice also that White has a backward
pawn on e4 ; this pawn is not as vulnerable as his co�nterpart on d6,
because he is not on a semi-open file and the e5 square is not
available as an outpost to Black's pieces.
Fischer exploits his positional advantage ( ± ) very effectively. Play
continued-1 3 Bg5! (the exchange Bxf6 will enhance White's con
trol of the d5 square) 13 . . . Qb6 + 14 Khl � 1 5 Bxf6 Bxf6 16 Bd5!
PAWN S T R U C T U R E S A N D PAWN WEAK N E S S E S 83
The exchange of light-squared bishops will leave White with a

'good' knight versus 'bad' bishop situation-the resulting position
is remarkably similar to the Povah-Nielsen position considered on
page 59-and the winning strategy was more or less the same. 16 . . .
Rac8 1 7 Bxc6 Rxc6 1 8 Rad1 (since this rook eventually returns to al ,
there might be a case for playing 18 Rfd1 intending Nd5 , c3 and a4 as
in the game) 1 8 . . . Rfc8 1 9 Nd5 Qd8 20 c3 Be7 Black's weak d-pawn
and the white knight on d5 give Fischer an overwhelming positional
advantage, and he now switches his attention to Black's vulnerable
queen-side pawns. Black is so tied up it is difficult to find an active
plan for him. 2 1 Ral ! f6 22 a4 ! White intends to give Black an isolated
pawn on the queen side: yet another weakness to worry about. Gadia
now blunders, but his position is disintegrating anyway. 22 . . . Rb8 ?
23 Nxe7 + resigns. Black loses a rook after 23 . . . Qxe7 24 Qd5 +-an
instructive minature from Fischer.
Our next example features what might be called a 'semi-backward'
pawn:
Diagram 51
Larsen-Donner
BBC TV Master Game
1 978
White has a lead in development which is a temporary advantage,
but more significantly he has the permanent advantage of the better
pawn structure. Larsen has no pawn weaknesses at all, whilst Don
ner has a weak pawn on c6; this pawn is ' semi-backward' because it
does not enjoy the protection of a colleague and the only way of
achieving thi s aim . . . c5, is very difficult to play. The game con
tinued 13 . . . Bb7 (the immediate 13 . . . c5 gives Donner an isolated
c-pawn after 14 dxc5 bxc5) 14 Ne5 ! (Larsen occupies a powerful
central outpost and increases the pressure against c6, thereby ruling
out . . . Nd7) 14 . . . Qd6 1 5 Qf3 (the pin on the long white diagonal
84 PAWN STR U CT U R E S A N D PAWN W E AK N E S S E S
now makes . . . c5 impossible; Donner's position i s at breaking-point

since he cannot defend against the pile-up on the weak c6 pawn, e.g.
1 5 . . . Rea ? fails to 1 6 Qxf7 + and other moves are answered by Be4
and Rfc1 when c6 falls) 1 5 . . . f6 ?! Donner attempts to dislodge the
powerful knight on e5, but this move just causes further white
squared weaknesses in his king side. It is quite common for a player
under pressure to have to make further positional concessions in
order to stay in the game. One might even say, ' weaknesses breed
weaknesses'. 16 Qe4 ! fxe5 ? (A terrible blunder, but the alternative 1 6
. . . f5 leaves the white knight i n complete control, and creates a
further weakness at e6) 1 7 dxe5! (Donner was relying on 1 7 Qh7 +
Kf7 18 Bg6 + when he can use the f6 square, but this interpolation
rules out that possibility) 1 7 . . . Qd8 ? (the last try was 1 7 . . . Qxd3 1 8
Qxd3 Ba6 but 19 R d 1 Bxc4 2 0 Qxc4 i s hopeless for Black) 1 8 Qh7 +
Kf7 19 Bg6 + Ke7 20 Qxg 7 + Rf7 2 1 Qxf7 mate.
Our final example in this section shows what may be termed a
'screened backward' pawn, which often arises in certain variations
of the Sicilian Defence.
Sicilian Defence, Lasker-Pelikan variation

1 e4 c5 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 e5 6 Ndb5 d6 7 Nd5
Nxd5 8 exd5 Nb8 9 c4 Be7 1 0 Bd3 0-0 1 1 0-0 Nd7 1 2 Be3 a6 1 3 Nc3
Diagram 52
Lein-Bilek
Rome
1 976
Black has a backward pawn on d6, but it clearly is not such a problem
as in our previous examples, and the main issue in this position is
White's queen-side majority against Black's king-side majority. Play
continued-1 3 . . . f5 14 f3 ! (with this move White stems the tide of
Black's intended king-side expansion, since 14 . . . f4 ? 1 5 Bf2 leaves
White in control of the e4 square) 14 . . . Bg5 Black strives to

exchange off his 'bad' bishop, so White naturally declines. 15 Bf2 ! g6
16 b4. Now Lein starts his own build-up on the queen side; an
advance which is much more difficult to thwart. 16 . . . b6 1 7 Rbt Rf7
ta Na4 Kha 1 9 Rel Bh4 20 g3 Bf6 21 c5! after patient manoeuvring
White is ready to 'break' on the queen side in order to create a passed
d-pawn 21 . . . bxc5 22 bxc5 Nxc5 23 Nxc5 dxc5 24 Bc4 ! ± Lein
secures his newly formed passed pawn, and the activity of his pieces
easily outweighs his pawn deficit. 24 . . . Rd7 25 Qb3 (not 25 Bxc5 ?
Qc7 winning one of the bishops) 2 5 . . . Bb7 26 Redt Rea 2 7 Rb2 Rcc7
(If 2 7 . . . e4 2a d6 ! ! gives White a potent attack, e.g. 2a . . . Bxb2 29
Qxb2 + Rg7 30 d7 Rba 3 1 Qe5-threatening Qxba !-3 1 . . . Raa 3 2
Bxc5 when White i s i n complete control; alternatively 30 . . . Rc7 3 1
Qf6! Qxf6 3 2 da = Q + wins; or if 2 a . . . Rxd6 29Rxd6 Qxd6 30 Qxb7
Qdt + 31 Bft + - ) 2a Re2 Kg7 29 Be t ! White now switches his
pressure to Black's e-pawn 29 . . . Rea 30 Bc3 Qc7 31 Qb2 notice the
pile-up technique 31 . . . Rfa (Attempts to hold the e-pawn are easily
refuted , e.g. 31 . . . Re7 ? 32 d6+ - or 31 . . . Rea 32 fa+ - ) 32 Bxe5
Qda 3 3 f4 + - Lein has regained his pawn with a dominating position
and he soon wins. 33 . . . h5 (33 . . . Bxd 5 ? loses to 34 Bxd5 Rxd5 3 5
Qb7 + + -) 3 4 Qc3 Bc6 (34 . . . Bxd5 still fails t o 3 5 Bxd5 Rxd5 36 Rxd5
Qxd5 3 7 Bxf6 + Rxf6 3a Re7 + winning) 3 5 Qe3 Bb5 (this time 3 5 . . .
Bxd5 fails to 36 Red2 Bc6 3 7 .Rxd 7 + Bxd7 3a Bxf6 + Rxf6 39 Qd3
winning the pinned bishop) 36 Bxb5 axb5 37 Qxc5 Qa5 3a Kg2 Rff7
39 Qc6 Ra7 40 d6 Qa4 41 Rd4 resigns. White's d-pawn proves to be
too much.
THE MINORITY ATTACK
The Minority Attack is a special feature applicable to certain pawn

structures. It is most commonly seen in the Queen's Gambit
Declined, but it is not exclusive to this opening. However, the QGD
frequently gives rise to positions where the minority attack is
White's most logical plan, hence it provides us with most of our
examples.
The basic objective of a minority attack is to weaken the oppo
nent's pawn structure by creating either an isolated pawn, or a
backward pawn on a semi-open file.
Our first example comes from the Exchange variation of the
Queen's Gambit Declined:
86 PAWN S T R U C T U R E S A N D PAWN WEAK N E S S E S
Diagram 53
This is a standard position which has been reached in literally

thousands of games. White has several moves here, but initiating the
minority attack with 1 1 Rahl is the most logical continuation. Basi
cally White intends to play b4 , b5 (supported by a4 if necessary) and
bxc6 to create a backward pawn on c6. If Black captures on b5 then he
will have an IQP, and White will have pressure against Black's
b-pawn; should Black play . . . b6, then another minority attack
ensues with a4-a5.
In Buslayev-Gedevanishvili, USSR, 1970, play continued from
the diagrammed position with-1 1 Rahl Ne4 Black cannot afford to
be idle, he must seek active counterplay against White's straight
forward but effective plan. 12 Bxe7 Qxe7 13 b4 Bg4? (Black should
try to slow down White's ambitions with 13 . . . a6! although 14 a4
Ng6 1 5 b5 axb5 16 axb5 Nxc3 1 7 Qxc3 Bg4 1 8 Nd2 still leaves White
slightly better) 14 Ne5 Nxc3 15 Qxc3 Bh5 1 6 Bf5 (not 16 b5 f6 1 7 Nf3
Bf3 1 8 gxf3 when Black will always have the prospect of counterplay
on the king side) 16 . . . Qg5 ? (16 . . . a6 was still advisable) 1 7 Bh3 f6
1 8 Nd3 Black's queen is now cut off from the queen side and the
black c-pawn is beginning to look very shaky indeed 18 . . . Re4 1 9
Nf4 Ng6 ? (Black misses the chance t o bolster the c-pawn with 1 9 . . .
Bea ! ) 20 Nxh5 Qxh5 2 1 b5 ± Raes ( 2 1 . . . Ne7 does not help after 2 2
bxc6 bxc6 2 3 Rb7 when White holds all the trumps!) 2 2 bxc6 bxc6 2 3
Qxc6 White's strategy has triumphed perfectly, for h e has not only
won a pawn, but his major pieces are ready to invade the vacuum
created on the queen side and Black's a- and d-pawns make
prime targets. 23 . . . Nh4 24 Rb5 Qe2 2 5 Qxd5 + Kha 26 Qxe4!
resigns.
It should be noted in Diagram 53 that Black could also launch a

minority attack against White's e-pawn which is also on a semi-open
file. However, it is very cumbersome and dangerous for Black to
carry out such a minority attack, since the advance of the f- and
g-pawns would weaken his king position.
In the last-example Black tried to counter the inevitable minority
attack by initiating action on the king side. Unfortunately though,
his king-side ambitions never really got under way. There are several
other methods of trying to counter the minority attack, but a
fully satisfactory answer has yet to be found. The various plans
are:
1. King-side action using the semi-open e-file and particularly the
e4 square. However, the attack is difficult as White has no weaknes
ses on the king side.
2. The minority attack can be delayed by . . . a5 , but a3 and b4 will
generally come anyway, so this is only a temporary measure.
3. The creation of a passed rook's pawn by . . . a6 , so that a4 and b5
can be answered by . . . cxb5 axb5 and then . . . a5.
4. Prepare to meet b5 with . . . c5.
5. Prevent b5 by playing . . . b5 oneself and then try to use the weak
c4 square.
Although plans 3 and 4 can often work, they are basically passive
·
in nature and rely on White's playing b5. The fifth plan is a direct
attempt to counter White's minority attack, but the creation of a
permanently weak pawn on· c6 necessitates accurate timing if Black
is not to run into even greater troubles. Our next example illustrates
just this problem: the game Donner-Hort, Wijk aan Zee 1 9 7 3 , con
tinued from Diagram 53 with 11 Rahl a5 12 Rfcl Ne4 13 Bxe7 Qxe7
14 a3 Ng6 ? ! (if Black wishes to use the . . . b5 plan, he wants his
knight on d6 so that he can use the c4 square-therefore, 14 . . . Nd6
was more logical) 15 b4 axb4 16 axb4 b5 ? (16 . . . Nd6 threatening . . .
b5 and . . . Nc4 was better, but White would play the immediate 1 7 b5
with the initiative) :
88 PAWN S T R U CTU R E S A N D PAWN WEAK N ES S ES
Diagram 54
Donner-Hort
Wijk aan Zee
1 973
17 Bxe4 ! ± Donner rightfully ascertains that his knights will be able

to dominate Black's weak dark squares, e.g. Nf3-d2-b3-c5 , etc. 1 7
. . . dxe4 1 8 Nd2 Bf5 1 9 Ne2 B d 7 20 Nb3 h 5 (Hort i s positionally lost
on the queen side so he must seek counterplay on the king side; 20
. . . Qxb4? would lose a piece to 2 1 Nc5) 2 1 Nc5 h4 (21 . . . Bea
preserving the bishop for the king-side attack was better) 22 Nxd7
Qxd7 2 3 h3 (not 2 3 Qxc6? Qg4 24 Qc2 h3 2 5 Ng3 Nh4! when Black
has dangerous counterplay) 23 . . . Re6 24 Rat-with Black tied
down to his backward c-pawn White takes control of the open
a-file-24 . . . Rxal 25 Rxal Qd5 (intending . . . Qg5) 26 Ra8 + Kh7 2 7
Qc5 ! Qxc5 2 8 bxc5 f5 29 Kfl Re7 3 0 Ket Black's weak pawn on c6
guarantees White a lasting initiative. 30 . . . Rd7 31 Kd2 Kh6 32 Kc2
Kg5 33 Ra6 Rc7 34 Kb3 f4? (34 . . . Ne7 was possible but 35 Nf4
followed by Kb4 and d5 is better for White) 3 5 Nc3 Kf5 36 d5! cxd5 3 7
Nxb5 Rb7 ( 3 7 . . . Rxc5 ? loses to 38 Nd4 + ! Kg5 39 Ne6 + + - or 38 . . .
Ke5 ? 39 Re6 mate) 3 8 Rb6 Rxb6 3 9 cxb6 and Donner's b-pawn soon
decided matters.
In conclusion the reader should note the very powerful influence
of the prevailing pawn structures in the examples quoted, for, as
stated at the beginning of this chapter, the pawn structure can often
dictate the whole course of a game!
Tactical Complexities of the
Middle Game
I n the previous two chapters w e covered what w e might call the

basic positional elements of the middle game, e.g. 'good' knight
versus 'bad' bishop and vice versa; the two bishops; open files; the
seventh and eighth ranks; weak pawn structures. Now we shall take
a detailed look at the tactical aspects of the middle game, paying
particular attention to the co-ordination of the pieces in combina
tions and sacrifices.
COMBINATIONS
The basic principles of chess combinations can be listed as follows:

1. Pin
2. Skewer
3. Fork
4. Discovered Checks, etc.
5. Overloading
We will now consider these principles in tum, but it should be
noted that a combination may be prefaced by, or incorporate, a
sacrifice. Therefore we shall devote the latter. part of this chapter to
the sacrifice.
1. The Pin
The Pin is one of the simplest but most effective weapons found in
chess, and forms the basis of many combinational finishes. Our first
example illustrates a series of pins forcing an opponent to tie up his
pieces, until one final pin proves to be decisive:
90 TACT I CAL C O M P L EX IT I ES OF TH E M I D D L E GAM E
Diagram 55
M. Harmonist-S. Tar
rasch Nuremberg
1 892
Tarrasch plays 28 . . . Ba4 pinning White's rook to his queen and
forcing White's reply-29 Qb2 + Bf6 30 Rxb4 cxb4 ! The g1-a7
diagonal is opened for the most effective pin of all-a pin against the
king ! 31 Qcl (if 31 Qe2 Re8 threatening to intensify the pressure with
. . . Bd4) 31 . . . Bd4 32 Re2 (32 Qd2 loses material to the skewer 3 2
. . . B c 3 ; while 3 2 Rft is met b y 32 . . . Bb5) 3 2 . . . fxe4 3 3 Bxe4
(recapturing with the pawn exposes the pinned knight to further
pressure down the f-file , e.g. 33 fxe4 ? Rf8 34 Qet-not 34 Qd2 or 34
Qft , both of which are answered by 34 . . . Bb5 - +-34 , . . Bb5 3 5
Rc2 Bd3 36 Rd2 B c 3 winning material) 33 . . . Re8 ! (setting up
another pin on the e-file and threatening . . . d5 when the bishop
would have to move allowing the exchange of rooks, removing one
of the knight's defenders, e.g. 34 . . . d5 3 5 Bd3 Rxe2 36 Bxe2 Bxf2 +
- +) 34 Qd2 (34 Qel is met by 34 . . . Bb5 35 Rc2 d5 winning the
pinned bishop) 34 . . . Bb5 3 5 resigns. White must lose material, e.g.
3 5 Bd3 Rxe2 36 Qxe2 sees the white queen overworked, while 3 5 Rel
fails to 35 . . . d5 36 Bxd5 Bxf2 + 37 Qxf2 Rxel + winning; a standard
pin device !
The power of the pin is well illustrated in another one ofTroitsky's
studies:
TACTI CAL C O M P L E X I T I E S OF TH E M I D D L E GAM E 91
Diagram 56
Study by A. Troitsky
1 926
The winning procedure is as follows-1 c6 b2 ( 1 . . . Kb8 trying to
stop White's c-pawn is no better, e.g. 2 Kd8 b2 3 c7 + Ka7 4 c8 = Q
bl = Q 5 Qc7 + Ka8 6 Bg2 + Be4 when the pin on the a8-hl diagonal
allows White to play 7 Kc8 ! threatening Qa5 or Qb8 mate without
having to fear checks on c4; Black's last try 7 . . . Qc2 fails to 8 Bxe4 +
Qxe4 when White mates) 2 c7 bl = Q 3 c8 = Q + Ka7 4 Qc7 + Ka8 (not 4
. . . Ka6 5 Bc8 mate) 5 Bg2 + Be4 6 Qh7 ! ! A brilliant 'cross-pin' which
either wins the black queen or forces mate thus 6 . . . Kb8 7 Bxe4
attacking the black queen and threatening Qb7 mate; the only effec
tive alternative is 6 . . . Bxg2 which loses the queen to 7 Qxbl , when
White 's task is easy.
Another example of a potent pin was seen in the game between
Nimzovich and Tartakover, Carlsbad 1 9 1 1 :
Diagram 57
Nimzovich-Tartakover
Carlsbad
1 91 1
White 's free development clearly gives him the advantage, although
Tartakover appears to have a reasonably solid position and the
92 TACTI CAL C O M P LEXITI ES O F TH E M I D D L E G A M E
exchange of dark-squared bishops will allow . . . Kg7 and the activa

tion of his king's rook. Unfortunately for Tartakover his bishop on g7
is weak because it is pinned to his king, and Nimzovich now takes
full advantage of this fact as follows-1 Qf6! Rg8 (1 . . . Bxf6 is illegal
and 1 . . . Bxh6 allows 2 Qxha mate) 2 Bxg 7 + Rxg7 3 Bxf7 ! wins. The
bishop is immune since 3 . . . Rxf7 4 Qha is mate and 3 . . . Qxf7 4
Qd8 + mates next move, and there is no way to meet White's numer
ous threats.
2. The Skewer
The skewer is a form of inverted pin, where the more valuable piece
is in the direct line of fire, and when he moves a lesser-valued
colleague comes under attack.
The skewer as a tactical device is less common than the pin, and
particularly so in the middle game when the early development of
the minor pieces gives rise to many potential pins. When a skewer
does arise, however, it can be quite devastating, as in the following
example:
Espeli-Andersen, Oslo 1952

1 e4 c6 2 c4 ds 3 cxds cxds 4 Bb5 + ? (a poor move at this stage, but
quite playable after 4 exds Nf6 5 Bbs + etc.) 4 . . . Bd7 5 Qa4 ?
(compounding the error) 5 . . . dxe4 6 Bxd7+ Nxd7 7 Qxe4 Nc5 8 Qc4
Nd3 9 Ke2 Rc8 ! :
Diagram 58
Espeli-Andersen
Oslo
1 952
This skewer on the c-file leads to a crippling state of affairs for
White-10 Qxd3 ? (White could have spared himself with 10 Qbs +
Qd7 1 1 Qxd 7 + Kxd7 1 2 Nc3) 1 0 . . . Qxd3 + 1 1 Kxd3 Rxcl 1 2 Ke2 g6
TACTI CAL COM P L E X I T I E S OF TH E M I D D L E GAM E 93
13 resigns! The pin along the back rank is so potent that White is
faced with inevitable material losses.
Examples of skewers as tactical devices in their purest form are
rarely seen in the middle game, simply because there are so many
intervening factors. So we shall consider first a simple skewer in an
endgame setting:
� � .
R ft a a a
. . . �
• • ···• •
� fa ·
•
• • • •• •
n
- � •t�
. Diagram 59
Study by P. Stamma
. � . 1 737
This study by Stamma shows a situation with both players on the
verge of promotion; White forces the win with the use of a tactical
resource--the skewer. 1 Bd4! this temporarily prevents the black
g-pawn from queening and invites the black king to occupy a square
on the g 1-a7 diagonal 1 . . . Kxd4 2 b8 = Q g 1 = Q 3 Qb6 + and the
black queen is lost.
As one might expect it is unusual to have skewers against the king
in the middle game (since he is normally tucked safely in the comer)
and it is the queen who is often the target for such tricks as our next
example illustrates:
Diagram 60
Geller-Sveshnikov
46th USSR Champion
ship 1 978
94 TACTI CAL C O M P L E X I T I E S O F TH E M I D D L E G A M E
A complicated position has been reached where White has the

material advantage of rook for bishop ('the exchange') , but his king is
very exposed and the two white rooks are being forked. Play con
tinued-32 Rb1 (White solves the problem by pinning the bishop to
Black's undefended rook on ba) 32 . . . Rea (Black threatens the white
queen and renews the threat of Bxa2) 33 Rxa5 (33 Rxb3 ? fails to 33 . . .
Rxe4 34 Rba + Rea - + ; the move played is a deflection sacrifice
which obviously cannot be accepted) 33 . . . d5 34 Rel ? ? (this 'self
skewer' is a fatal error, the correct defence was 34 Qxea + Qxea 3 5
Rxb3 Qg6 + 36 Kf1 Qd3 + 3 7 Kg2 Qg6 + ) 3 4 . . . Qg6 + ! 3 5 resigns.
=
The check is decisive, e.g. 35 Qxg6 Rxe1 + 36 Kg2 hxg6 providing a

flight square for the black king, and emerging with an extra piece.
This example amusingly (although not for Geller) incorporated a
fork, pin and skewer, and it was the last which eventually decided
the issue.
3. The Fork
The fork is perhaps the most well-known tactical device on the
chess-board. It simply means a simultaneous attack on two or more
of the opponent's pieces by a single piece. For this reason any piece
is capable of forking, but the knight fork is definitely the most
common.
Our first example illustrates a pawn fork; consider the following
variation in the Berlin-Classical defence to the Ruy Lopez: 1 e4 e5 2
Nf3 Nc6 3 Bb5 Bc5 4 0-0 Nf6 5 Nxe5 ! :
Diagram 61
Ruy Lopez
Berlin-Classical
Variation
This 'pseudo-sacrifice' is considered very strong in this position, so

much so that Black normally declines the piece and plays 5 . . . Nxe4
TACT I CAL C O M P L E X I T I E S O F TH E M I D D L E GAM E 95
leading to interesting complications. However, we shall examine the

line continuing with 5 . . . Nxe5 6 d4. This pawn fork regains the
piece and ensures White the initiative, e.g. 6 . . . Bd6 7 f4! Ng6 8 e5
Be7 9 f5 ! ± White will snare one of the knights and in the meantime
he has gained a clear spatial advantage.
Korchnoi provides us with a nice example of a queen fork and a
knight fork in the following important theoretical game:
Diagram 62
Lombard-Korchnoi
Switzerland
1 978
In this position in the Rubenstein variation of the Nimzo-Indian,
Korchnoi comes up with a move which was previously condemned
by 'theory'-9 . . . Nxd4 ! ! 10 Qa4 (this is the reason why Korchnoi's
move was thought unplayable: the white queen is forking the two
black bishops, and 10 . . . Ba5 1 1 b4 is clearly no good, so Black's
reply seems forced) 10 . . . Bxc3 t tbxc3. Now both the black bishop
and the knight on d4 are attacked and it seems as if Black must lose a
piece, but Korchnoi has seen a stage further 1 1 . . . Ne6 ! (This move
solves most of Black's problems, he saves his knight and is able to
answer 12 Qxa6 with 12 . . . Nc5 forking White's queen and bishop
on d3, thereby regaining his piece and remaining a pawn to the
good.) 12 Ba3-this is the only attempt for White to gain the initia
tive, but even this doesn't prove adequate 12 . . Bb7 (obviously not
.
12 . . . Rea or 12 . . . d6 when White does win a piece with 13 Qxa6

since 13 . . . Nc5 is met by 14 Bxc5) 13 Bxfa Qxfa White has won the
exchange but Black's extra pawn � d superior pawn structure are
more than enough compensation, and it did not take Korchnoi long
to win White's doubled c-pawns and then the game.
Our next example shows a forcing sequence of moves which bring
the black king and queen into 'relationship' :
96 TACTICAL COM P L E X IT I E S OF T H E M I D D L E GAM E
Diagram 63
M. Euwe-J . Davidson
Holland
1 925
Euwe played 1 Qd8 + Kg7 2 Qxf6 + ! Kxf6 3 Nxe4 + Ke5 4 Nxc5 bxc5
giving him an extra pawn in an easily won king and pawn endgame.
Another more complicated example of the same theme was seen in
the game Zukertort-Englisch played in London in 188 3 :
Diagram 64
J. H. Zukertort-B. Eng
lisch London
1 883
Zukertort starts off with a very ingenious queen sacrifice-1 Qb5 ! !

Qxb5 (forced in view of 1 . . . Kd7? and 1 . . . Nd4 both being
answered by 2 c8= Q + winning) 2 c8 = Q + Kf7 (2 . . . Nd8 3 Nc7 +
+ - ) 3 Qxe6 + ! Kxe6 4 Nc7 + this knight fork decides the game.
Our final example in this section shows a series of sacrifices
leading to a deadly knight fork:
TACTI CAL C O M P L E X I T I E S OF T H E M I D D L E GAM E 97
Diagram 65
Povah-Granat
London
1 979
One more move was all that was necessary to win this game 18 Rxg7 + !
resigns. The continuation would have been 18 . . . Kxg7 19 Bh6 + ! !
Kxh6 (obviously 19 . . . Kh8 2 0 Nxf7 + and 1 9 . . . Kg8 20 Qg5 + are no
better) 20 Nxf7 + (the decisive knight fork) Kg7 2 1 Nxda Rea (21 . . .
Rxda 2 2 Qxe6 + - ) 2 2 Qg5 + when White gains a clear material
advantage.
4. Discovered checks
A discovered check is a truly deadly weapon because there is a lot to
be said for the anonymous quote in Kotov' s excellent book Think like
a Grandmaster-'When I give check I am afraid of nobody! '
As the enemy king rarely walks voluntarily into the firing line of a
discovered check, he needs to be enticed by sacrificial bait. The next
example illustrates this point very forcibly:
Diagram 66
G. W. Wheeler-Povah
London
1 977
98 TACT I CAL C O M P L EX I T I E S O F TH E M I D D L E GAM E
If White continues 1 8 cxd5 ( 1 8 Bxd 5 Ng4 1 9 Rfe1 Qxh2 + 20 Kf1 Bxd 5.

wins for Black) 1 8 . . . Ng4 1 9 Rfel Qxh2 + 20 Kfl then Black has
regained his pawn with the better prospects. So White decides to
expel the black queen before capturing on d5. He should have
preferred the above line, for the game concluded 1 8 Bg2 ? ? when
Black was able to close proceedings with 1 8 . . . Qxg2 + ! 19 resigns.
White's forced reply 19 Kxg2 would have been answered by the
double discovered check 19 . . . Nf4 + followed by 20 Kgl Nh3
mate.
Another example of such a drastic finish was seen in the following
encounter:
Diagram 67
HaldanErClegg
London
1 978
White came up with the stunning 1 5 Qd8 + ! ! and Black resigned! It's
mate in four!-1 5 . . . Rxd8 ( 1 5 . . . Nxd8 comes to the same) 1 6 Nf6 +
(the queen sacrifice was necessary on two counts-first to vacate the
·f6 square, and second to cause an obstruction on d8) 16 . . . Ke7 1 7
Nxd5 +-this double discovered check i s decisive---1 7 . . . Ke8 1 8
Nbc7 + Nxc7 1 9 Nxc7 mate. And this time a miniature from two of
'the greats' :
R . Reti-Dr Tartakover
1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nf6 5 Qd3 (it is more natural to
capture on f6 when Black can consider either pawn recapture. Tar
takover now tries to refute White's play. 5 . . . Nxe4 intending . . . Nd7
and . . . Nf6 was wiser) 5 . . . e5? 6 dxe5 Qa5 + 7 Bd2 Qxe5 8 0-0-0!
White breaks the pin on the e-file and sets a deadly trap which with 8
. . . Nxe5?? Black falls into (8 . . . Be7 was advisable) :
TACTI CAL C O M P L EX I T I ES OF TH E M I D D L E GAM E 99
Diagram 68
R. Reti-Dr Tartakover
Reti now punishes Tartakover for both his greed and lack of
development with 9 Qda + ! ! Kxda 1 0 Bg5 +-it's double check,
which always must be answered with a king move. 10 . . . Kc7 ( 10 . . .
Kea 1 1 Rda is also mate) 1 1 Bda mate. A beautiful finish.
In case the reader thinks all discovered checks are either double
checks, or end in mate, we give the following very instructive ·
example:
Diagram 69
A. Kupferstich-J . And
reassen Denmark
1 953
White is the exchange and a pawn down, but his pieces are so active
(especially compared to Black's) that he is capable of producing a
superb finale. Play continued-2 9 Rg7 + Kfa 30 Rxb7 + Kga (natur
ally not 30 . . . Kea allowing 31 Nd6 + Kda 32 Bg5 mate) 3 1 Rg 7 +
Kfa. The black king is caught in a series of discovered checks amus
ingly known as a 'see-saw'-the white rook could happily demolish
1 00 TACTICAL C O M P L E X I T I E S O F TH.E M I D D L E GAM E
the whole of the second rank if he so desired. As Nimzovich stated,

'the see-saw can be the cause of frightful devastation'. 32 Rxa7 + Kg8;
unfortunately for White the black rook on a8 is guarded by the
bishop on f3 , so how is he going to convert his positional superior
ity? Perhaps the reader would like to study this position, for White's
winning process is an amazing conception: see if you can find it!
Diagram 70
Kupferstich-Andreassen
After 32 . . . Kg8
Kupferstich amazingly exchanged off his active rook. 33 Rxa8 Bxa8
and continued 34 Nd6! when Black resigned! ! A close examination
of the position reveals that Black is helpless against the following
mating sequence: 34 . . . g5-this is as good as anything else since
White's knight and bishop have the black king completely incarcer
ated-3 5 Kd2 Bd5 this bishop is worse than useless, because with
out it Black might have some hope of stalemating himself. 36 Kc3
Bxa2 37 Kb4 g4 38 Kc5 e4 39 Kc6 White is not interested in Black's
pawns since he is after bigger gam�the black king!-39 . . . B e6 40
Kc7 g3-trying for stalemate possibilities--41 hxg3 e3 42 fxe3 Bd7
43 Kd8 Be6 44 Ke7 Bf5 (44 . . . Bd7 45 Ne4 and Nf6 mate) 45 Ne8
followed by 46 Nf6 mate. Unfortunately for Black he could never
force White to capture his bishop. Surely one of the most unusual
conclusions ever seen on the chess-board!
5. Overloading
Overloading is a concept which can often bring about the downfall
of a player's position. By allowing one or more of his pieces to
become 'overworked' a player provides his opponent with . the
opportunity to exploit the lack of mobility and flexibility that results
in the co-ordination of his forces.
TACTICAL C O M P L EX I T I E S O F TH E M I D D L E GAM E 1 01
Diagram 71
Sicilian Dragon
Yugoslav Variation
After 1 0 . . Rc8
.
The position shown in Diagram 71 is standard in the Yugoslav

variation of the Sicilian Dragon. White normally takes heed of
Black's threats on the c-file with 1 1 Bb3 , but what if he were to
initiate an immediate king-side advance with 1 1 h4 ? The answer is a
simple tactical refutation 1 1 . . . Nxd4-this discovers an attack on
the bishop on c4 (i.e. opens the c-file) , thus forcing White's reply 1 2
Qxd4. The white queen i s now lined u p with the 'Dragon' bishop
allowing the powerful discovery 12 . . . Ng4 ! 13 Qd3 . Now we see
that the white queen is 'overworked' since she is protecting two
pieces and is easily deflected away from one of them 13 . . . Nxe3 14
Qxe3 Rxc4 with a simple win.
Our next example shows a queen being overworked, having to
defend two key squares:
Diagram 72
P. Morphy-T. Worrall
New Orleans
1 857
The tactically brilliant Paul Morphy had no trouble in taking advan
tage of the overtaxed black queen with the following straightforward
1 02 TACTI CAL C O M P L E X I T I E S OF TH E M I D D L E GAM E
sequence 1 Rf8 + Qxf8 2 Rxf8 + Rxf8 3 Qxg6 mate. Basically the queen
was deflected from her defensive duty towards the g6 square. The
idea of deflecting a piece from its defensive duties is a well-known
one, and the famous problem and study composer A. Troitsky pro
duced the following example:
Diagram 73
Study by A. Troitsky
1 898
First White draws the black king into a mating net as follows 1 h4 +
Kg4 (obviously not 1 . . . Kf5 or 1 . . . Kg6 which both lose to 2 N e7 + )
2 Kg2 Qc5 (forced i n order t o stop the threat o f . . . Ne3 mate) 3 Ra5 !
White starts harassing the black queen who must keep an eye on the
e3 square 3 . . . Qa7 4 Rxa6 Qc5 5 Ra5 Qd4 6 c3 Qd2 7 Ra2 Qcl 8 Rc2
finally, the black queen has been hunted to death and must desert her
defence of e3, so White wins.
Our final example in this section is, in the author's opinion, one of
the most brilliant combinations ever seen:
Diagram 74
Dr S. Tarrasch-3
Amateurs Naples
1 91 4
TACTI CAL C O M P LEX I T I E S O F TH E M I D D L E GAM E 1 03
Tarrasch as White has an extra pawn and dangerous attacking pros

pects against the black king. At the moment Black appears to have
adequate defences to White's threats of Qb7 + and Rxc5 + : see if you
can find the amazing move played by Tarrasch. The point of this
position is that Black's defences to the key squares b7 and c5 inter
sect at the c7 square. Tarrasch played the beautiful 3 1 Bc7 ! ! where
upon Black resigned. This fabulous move illustrates the famous
Plachutta interference theme, well known to problemists. Black is
powerless to meet White's dual threats, e.g. 31 . . . Rxc7 is refuted by
32 Qb7 + ! (deflecting the black rook since 32 . . . Kxa5 33 Rat leads to
mate next move) 32 . . . Rxb7 33 Rxc5 mate, and 31 . . . Qxc7 32
Rxc5 + (again deflecting the key defender) Qxc5 3 3 Qb7 + with mate
next move. Tarrasch's brilliant bishop sacrifice caused one of the
two black defenders to become overworked, while the other became
obsolete. This combination must rank as one of the most impressive
displays of an interference theme combining with a deflection theme
to deliver mate.
'Chess is 99 per cent tactics'-Richard Teichmann

We round off this chapter with an examination of further tactical
motifs paying particular attention to the sacrifice. As the reader has
probably come to realize, there are few examples of a pure tactical
theme in the middle game, since most combinations are contami
nated by the interaction of several basic tactical principles.
Our next example basically utilizes a potentially deadly knight
fork:
Diagram 75
S. Levitzky F . J. Marshall
-
Breslau
1 91 2
Marshall has very active pieces and he quickly capitalizes on this
fact with the rather obvious 1 . . . Rxh3: the rook is immune from
1 04 TACTI CAL C O M P L EX I T I E S O F TH E M I D D L E G A M E
capture, because the white g-pawn is needed to guard the f3 square,

which is very attractive to Black's knight. Levitzky tries to confuse
the issue with 2 Rc5 attacking the black queen, but now Marshall
comes up with a truly amazing move which forces immediate resig
nation. What did Marshall play? See if you can spot the star move.
The answer to the position is quite simply to move the black queen
to a square where she is seemingly in even greater danger, namely 2
. . . Qg3 ! ! The threat of mate on h2 forces one of the three captures, all
of which lose: first 3 fxg3 loses to 3 . . . Ne2 + 4 Kh1 Rxf1 mate, and 3
hxg3 Ne2 is mate; finally 3 Qxg3 Ne2 + 4 Kh1 Nxg3 + 5 Kg1 leaves
Black with a clear material advantage in the ending.
One could argue whether the last example employed any sac
rifices at all, since if even the sacrifices were accepted, mate or
material gain would result. Therefore it is convenient to group
sacrificial combinations into two categories:
1. the Pseudo-sacrifice
2. the Speculative-sacrifice.
Sacrifices are basically tactical by definition, but the consequences
can vary from a forced mating sequence at one extreme to long-term
positional difficulties on the other extreme.
1. The Pseudo-sacrifice
By ' pseudo-sacrifice' we mean the type of sacrificial combination
where the consequences can be clearly analysed to a satisfactory
conclusion, such as a forced mate or gain of material. Chess literature
is littered with countless examples of sacrifices denuding an enemy
king who is rapidly slaughtered.
That masterful tactician Tarrasch provides us with a typical
example:
Diagram 76
Dr S. Tarrasch-Satzinger
Munich
1 91 5
TACTI CAL C O M P L EX I T I E S OF T H E M I D D L E GAM E 1 05
The game continued-18 fxg7 Nxg7 and now with so many white
pieces pointing towards the hapless black king it should come as no
surprise to find that White has a killing combination. Tarrasch
played 19 Qxh7 + ! ! Kxh7 20 Rh3 + Kg8 (20 . . . Nh5 21 Rxh5 + Kg8 2 2
Rh8 + ! Nxh8 23 Nh6 mate) 2 1 Nh6 + Kha 22 Nf7 + K g 8 23 Rh8 + ! Nxh8
24 Nh6 mate! Tarrasch is to be commended for finding this beautiful
variation in a position where many players would have settled for a
more mundane win with a move such as 19 Bxg6, etc.
The next example is somewhat similar in that the enemy monarch
is exposed but, unlike the previous example where White could
hardly fail to win, this time the consequences of the sacrifice could
not be readily calculated, although the result was never really in
doubt-only the method of extermination needed to be found.
Diagram 77
Y. Averbakh-A. Kotov
Zurich Candidates'
Tournament
1 953
This battle, between two of Russia's leading Grandmasters of the

day, wa� developing into a heavy positional struggle with White
intending to attack down the open g-file and Black poised to open
the queen side with a timely . . . b5. But it was here that Kotov had a
flash of inspiration which led to what may be referred to as an
'instinctive' or 'intuitive' sacrifice. K �tov played the highly imagina
tive 30 . . . Qxh3 + ! ! 31 Kxh3 Rh6 + 32 Kg4 Nf6 + 33 Kf5-White's
king has 'been drawn into the jaws of death but strangely enough
Black has no immediate forced win. Fortunately for Kotov, White's
pieces are passively placed behind a wall of pawns, and can only
watch the annihilation of their monarch from afar. 33 . . . Nd7
(Stahlberg later suggested 33 . . . Ng4 as being even stronger, but 34
Nxf4 Rfa + 3 5 Kxg4 Rg8 + 36 Ng6 + keeps White kicking for a while)
34 Rg5 staves off the threatened mate . . . Rf8, . . . Rg8 and . . . Rf6,
1 06 TACTI CAL C O M P L E X I T I E S O F T H E M I D D L E G A M E
and w e can b e quite sure that Kotov was not able t o analyse all the
consequences of his brilliant queen sacrifice, but his instinct must
have told him that the day of reckoning would not be far away. 34 . . .
Rfa + 3 5 Kg4 Nf6 + 36 Kf5 Nga + 3 7 Kg4 Nf6 + Black is trying to reach
the time control at move 40. 3a Kf5 Nxd5 + 39 Kg4 Nf6 + . By captur
ing the pawn on d5 Kotov has avoided a three-fold repetition of
position and can safely make the time control without having to play
a committal move-very good technique! 40 Kf5 Nga + 41 Kg4 Nf6 +
42 Kf5 Nga + 43 Kg4 Bxg5 44 Kxg5 Rf7 (threatening mate in two with
. . . Rg 7 + and . . . Rf6 mate) 45 Bh4 ( 45 Nxf4 was no good either 45 . . .
Rg 7 + 46 Ng6 + Rgxg6+ 47 Kf5 Ne7 mate) 45 . . . Rg6 + 46 Kh5 Rfg7
47 Bg5 Rxg 5 + 4a Kh4 Nf6 49 Ng3 Rxg3 50 Qxd6 R3g6 51 Qba + Rga
52 resigns. A delightful victory for intuition!
2. The Speculative-sacrifice
Our first example is taken from one of the author's games in which he
had the opportunity to play an exchange sacrifice which denuded
the enemy king thereby offering long-term attacking prospects:
Diagram 78
N. Povah-S. Aarseth
National Plate Tourna
ment 1 976
Black played the rather provocative 16 . . . Bc5?! and White had no
hesitation in playing 1 7 Rxf6 gxf6. Black's king is very vulnerable
now and his defence will prove difficult; White must avoid too many
exchanges, which explains his next move 1a Khl ! It is not easy to
follow up a sacrifice with a 'quiet' move like this, but it was quite
necessary as will be seen. 1 a . . . Bxe3 19 Qxe3 (now we see the point
of White's king move, for if the king were still on gl Black could
exchange queens with 19 . . . Qc5) 19 . . . Qd6 ? (Black has a very
difficult defensive task particularly on the dark squares round his
TACT ICAL C O M P LEX I T I E S OF T H E M I D D L E GAM E 1 07
king. Relatively best was 19 . . . Kha 20 Rxd6 Nd7 2 1 Nd5 Qda 22

Qh6! with powerful threats of Rd6 , Rxd7 and Nf6 or simply Rf5 and
Rh5) 20 Qh6 + - Kha (If 20 . . . Rc6 then 21 Rf3 is decisive, e.g. 21 . . .
f5 22 Rg3 + Qg6 23 Rxg6 + - ) 2 1 Rxf6 ( 2 1 Rf3 Rga 22 Rh3 Rg7 allows
Black to defend) 21 . . . Qd4 (if 21 . . . Qc5 22 Qh3 ! intending Rh6 ,
etc.) 22 Rxb6 White now has a material advantage as well as his
overwhelming positional superiority 22 . . . Rga 23 Qf6 + Rg7 24
Rxb7 Kga 25 h3 (not 25 Nd5 ? allowing 25 . . . Qxd 3 ! with unneces-
sary complications) Qe3 26 Nd5 Qg3 27 Bfl resigns.
This sacrifice was speculative in that the author was not able to
foresee that he would regain his material in such a forceful fashion,
and his seventeenth move was based more upon judgement than
calculation. It is true that Kotov's queen sacrifice in the previous
section might also fall into this category, especially when one con
siders that his sacrificial investment (queen for pawn) was much
higher, but the extremely exposed state of Averbakh' s king made his
chances of survival highly unlikely.
Now we turn to perhaps the 'king' of the speculative
sacrifice-Mikhail Tal. Tal's games are abundant with weird and
wonderful sacrifices of which the consequences are far from clear.
Our first example is from the World Championship candidates'
semi-final, Bled, 1965:
Diagram 79
Tai-Larsen
World Championship
Candidates
Bled
1 965
A sharp battle has ensued so far, but now Tal plays a well-known
sacrifice in a somewhat different setting 16 Nd5 ! ! This sacrifice is
usually seen in such positions when the black king is still in the
centre a'.nd White has a rook on el to take advantage of that fact 16 . . .
exd5 1 7 exd5. This is a truly speculative sacrifice because for his
1 08 TACTI CAL C O M P L EX I T I E S OF T H E M I D D L E GAM E
piece White has only one pawn and two open lines namely the b1-h7
diagonal and the e-file. It is true that White has some tangible threats
such as 18 Qe4, but surely Larsen should be able to defend? 17 . . .
f5 ?! (Well, if there is a defence, then this is the critical juncture. First,
it should be noted that 17 . . . Re8 and 17 . . . B d8 both fail to the
classic double bishop sacrifice 18 Bxh7 + ! Kxh7 19 Qh5 + Kg8 20
Bxg 7 ! ! Kxg7 21 Qh6 + Kg8 2 2 g6 with a mating attack. General
opinion suggests 1 7 . . . g6 was the best defence, and Larsen even
went as far as to say that this was the refutation of Tal's attack. Such
is the way with speculative-sacrifices, however, that we cannot be
sure that Larsen would have been able to defend this position, given
that the defence is tricky and he was running short of time-factors
which Tal frequently exploits! ) 18 Rde1 Rf7 (Tal gave the following
line after the game 18 . . . Bd8 19 Qh5 Nc5 20 Bxg 7 ! Nxd3 + 21 Kb1 !
Nxe1 22 g6 Kxg7 2 3 Qxh7 + Kf6 24 g7 Rf7 ? 2 5 g8 =N mate. Such is
Tal's power of calculation and imagination that he gets to play and
analyse more sacrifices in one game than many players do in a
lifetime!) 19 h4 Bb7 20 Bxf5 ! (this pseudo-sacrifice proves very
strong, but 20 g6 hxg6 followed by 2 1 h5 opening the king side was
also promising) 20 . . . Rxf5 21 Rxe7 Ne5 ! (Enterprising defence: if
instead 2 1 . . . Rf7 then 22 Rxf7 Kxf7 2 3 g6 + ! hxg6 24 h5! gives White
a deadly attack.) 22 Qe4 Qf8 2 3 fxe5 Rf4 24 Qe3 Rf3 25 Qe2 Qxe7 2 6
Qxf3 dxe5 2 7 R e 1 Rd8 2 8 Rxe5 Qd6 2 9 Qf4! (with the threat of 3 0
Re8 + so Tal maintains his two pawn advantage and positional
superiority) 29 . . . Rf8 30 Qe4 b3 31 axb3 Rf1 + 32 Kd2 Qb4 + 33 c3
Qd6 34 Bc5 ! (the final decisive sacrifice, deflecting the black queen
away from her defence of the e6 square) 34 . . . Qxc5 35 Re8 + Rf8 36
Qe6 + Kha 3 7 Qf7 resigns.
Finally we give a complete game illustrating Tal's ability to make
speculative material sacrifices:
Tal-Vooremaa, Tallinn, 1971

1 e4 c5 2 Nf3 e6 3 d4 cxd4 4 Nxd4 a6 5 Bd3 Nc6 6 Be3 (6 Nxc6 is the
alternative) 6 . . . Nf6 7 0-0 Qc7 8 Nc3 Bd6?! 9 Kh 1 ! (defending the
g
h-pawn without weakenin his king side 9 . . . Bxh2 ! would lose the
bishop to 10 Nxc6 and 11 f4, etc.) 9 . . . h 5 ? Vooremaa is intent
upon initiating his king-side attack. 10 f4 Ng4 11 Qf3 Nxe3 12
Qxe3 Qb6 1 3 Nce2 (Black has the two bishops, but his light-squared
bishop cannot be developed for the moment and his king is
still in the centre. Furthermore Tal threatens to expand with e5
TACT I CAL C O M P L EX I T I E S OF TH E M I D D L E GAM E 1 09
and perhaps f5 , so Black takes preventive measures.) 1 3 . . . e5

14 Qg3 !
Diagram 80
Tal-Vooremaa
Tallinn
1 971
After 14 Qg3!
(A typical Tal speculative sacrifice which avoids the simplification
of 14 Nf5 Qxe3 15 Nxe3 exf4 1 6 Nxf4 Bxf4. It is unlikely that Tal could
have calculated all the consequences of his piece sacrifice, but his
judgement must have told him that Black's defensive task would not
be easy.) 14 . . . exd4 1 5 Qxg7 Rf8 1 6 e5 (So for his sacrificed piece Tal
has one pawn, the more active pieces and a dominant pawn centre.)
16 . . . Be7 17 f5 f6 (Vooremaa rightly fears 18 f6 which would give
him an even more cramped position, but since no immediate win is
in sight it was probably preferable to try to unravel his queen side
rather than further the weakening of the king side) 18 Nf4 ! Rf7 ( 18 . . .
fxe5 allows 19 f6, while 18 . . . Nxe5 is met by 1 9 Qxe7 + Kxe7 20
Nd5 + , etc.):
Diagram 81
Tal-Vooremaa
After 1 8 . . . Rf7
110 TACTI CAL C O M P L E X IT I E S O F T H E M I D D L E GAM E
Tal now produces one of those spark.ling queen sacrifices which is so

common in his play. 19 exf6 ! ! Ne5 (Black declines, for 19 . . . Rxg7 20
fxg7 Kf7 21 Nxh5! threatening both 22 f6 or 22 Bc4 + is too unbear
able. Tal produces yet another surprise, but not the last! ) 20 Bc4 ! !
The queen still hangs , and now the bishop i s offered. 2 0 . . . Nxc4 2 1
Qg8 + BfB (If 2 1 . . . RfB then 22 Qxc4 regains one of the pieces and
leaves Black in difficulties due to the threats of Nd5-c7 and Rae t + ,
etc. The move played is little better, because despite Black's two
extra pieces his king is a sitting target on the open e-file. ) 22 Nxh5
Nd6 (White threatened 23 Raet +) 23 Rae t + Kd8 24 Re7 ! Qb5 !
(Vooremaa takes the opportunity t o threaten mate himself. 24 . . .
Rxe7 would lead to mate thus 2 5 fxe7 + Kxe7 26 Qg5 + Kf7 2 7 Qf6 +
Kg8 28 Qg6 + Kha 29 Nf6, etc.) 25 Rfet Qd5 ? ? (Gufeld suggested 2 5
. . . Qa5 26 Rat ! after the game, but i t i s the difficulty of the defence
that is the hallmark of the speculative-sacrifice.) 26 Nf4 Qxa2 2 7
Ne6 + ! ! (The final surprise--Tal sacrifices a third piece i n order to
mop up Black's unco-ordinated king-side forces. 27 . . . dxe6 would
allow 28 Rxf7 Nxf7 29 Qxfa + Kca 3 o Qxf7 + Kb6 31 fxe6 with an easy
win.) 27 . . . Qxe6 28 fxe6 Rxf6 29 Rf7 resigns. All resistance is
crushed after 29 . . . Nxf7 30 exf7 + , etc.
More Complex Endings
In Chapter 1 we considered basic endings; now, having covered

certain aspects of the opening and middle game, we are ready to take
a more detailed look at the endgame. It should be emphasized that
this book cannot possibly compete with the many specialist books
on the endgame, which the reader is advised to consult. Our objec
tive here is to make the student aware of some important practical
issues which can often arise in the endgame.
FURTHER KING AND PAWN ENDINGS
Although king and pawn endings are considered as being the easiest
to play, it is important that they are played with great care, for one
slip can often prove to be fatal. Consider the following example:
Diagram 82
Re&-Ftacnik
Kiev
1 978
An examination of the diagrammed position reveals that Black
stands better; he has the white king tied down to the defence of the
112 M O R E C O M P L EX E N D I N G S
weak e-pawn and h e has an extra tempo o n the queen side with the
pawn move . . . b5 if he should ever need it. Ftacnik thought his
advantage was so clear that he made the common error of underes
timating his opponent's counter chances. His choice is between 56
. . . f5 and 56 . . . g5, one move, although drawing, offers some
winning chances while the other loses, so which move should he
play? Do not read any further until you have made a thorough
investigation of the position and chosen what you believe to be a
correct and safe course of action.
Complacency in chess can be a real killer and Ftacnik was guilty of
just this because he played the fatal blunder 56 . . . g5? He hopes to be.
able to play 5 7 . . . gxh4 58 gxh4 f5 59 exf5 Kxf5 60 Kf3 when the
tempo move 60 . . . b5 places White in zugzwang since he must lose
his h-pawn and consequently the game. Unfortunately, however,
Ree now found a single move which completely destroys Black's
game 57 g4! !-White wins in every variation .as follows:
(a) 57 . . . gxh4 58 gxh5 h3 59 Kf2 Ke6 (59 . . . f5 . 60 h6 Kf6 61 exf5 is
decisive) 60 Kg3 Kf7 61 Kxh3 Kg7 62 Kg3 Kh6 63 Kf4 Kxh5 64 Kf5
with an easy win.
(b) 57 . . . hxg4 58 h5 the point! The h-pawn is now passed. 58 . . .
Ke6 (58- . . . f5 59 h6 f4 + 60 Kf2 g3 + 6 1 Kg2 Kf6 62 e5 + ! Kg6 63
e6-White's two passed pawns are much more powerful than
Black's 3 ! ) 59 Kf2 Kf7 60 Kg3 Kg7 61 Kxg4 Kh6 62 Kf5 Kxh5 63 Kxf6
g4 64 e5 g3 65 e6 g2 66 e7 gt = Q Black queens first but to no avail. 6 7
e8 = Q+ Kh4 68 Qh8 + K g 3 69 Qg7 + Kf2 70 Qxg1 + Kxg t .
This long forcing variation was i n fact the game continuation, and
Ree was now able to win the race to the queen side 71 Ke5 Kf2 72 Kd5
Ke3 73 Kc6 (not 7 3 Kc4? Kd2 7 4 Kb4 Kc2 75 Kxa4 Kxb2 drawn) 73 . . .
Kd2 7 4 Kxb6 Kc2 75 Ka5 Kxb2 76 Kxa4 White wins by one tempo! 76
. . . Kc3 7 7 Kb5 resigns. The a-pawn reincarnates.
The safer try was 56 . . . f5 when White would have drawn, if he
was careful!
Now that the reader has been warned about some of the dangers in
king and pawn endings, we will consider some special cases.
In Diagram 83 we see a common-3 versus 3-situation where
White can win by the use of a well-known resource:
MORE COMPLEX ENDINGS 113
Diagram 83
White plays 1 g5 ! and answers 1 . . . fxg5 with 2 h5! gxh5 (2 . . . gxf4 3

hxg6 f3 4 g7 f2 5 g8 = Q + + -) 3 f5 and White queens first. On 1 . . .
hxg5 the reply is 2 f5 ! with the same consequences.
A similar 'trick' was seen in a game--
Pomar-Kuadras, Olot, 1974:
Diagram 84
Pomar-Kuadras
Olot
1 974
Black played 63 . . . f4! 64 Kxd5 (the alternatives were no better, e.g.

64 exf4 h4! when 6 5 Kxd5 loses to 65 . . . h3 and 6 5 gxh4 g3! 66 fxg3
e3 - + Naturally 64 . . . d4 would also win, but the above variation is
more attractive. The other alternative 64 gxf4 loses to 64 . . . h4 65
Kxd5 h3) 64 . . . h4! Black is determined to break through at any cost.
65 Kxe4 (65 gxh4 loses to the familiar resource 65 . . . g3 66 fxg3 fxe3
- + , while the other captures allow 65 . . . h3 winning) 65 . . . f3 ! 66
gxf3 h3 and the h-pawn decides the game. A very instructive
1 14 M O R E COM PLEX ENDINGS
example of a well-known theme. It was knowledge of position 83

which presumably enabled Kuadras to engineer his clever break
through. From this the reader should see the importance of studying
basic themes.
We now tum to an interesting example which illustrates the
mighty power of zugzwang in king and pawn endings:
Diagram 85
This position can be considered in two parts. First on the left-hand

side of the diagram we have a rare case of 'mutual zugzwang' where
the first player to move loses! If White moves either pawn then he
loses them both, while if Black moves his king then one of the white
pawns queens, e. g. 1 . . . Kc7 2 a7 or 1 . . . Ka7 2 c7. Therefore we are
able to deduce that the real battle rages on the right-hand side of the
board where each player will be trying to restrain the other so as to
place him in zugzwang. So the question arises-'can a lone king
cope with three connected passed pawns?' Well the answer is
dependent on the positions of the king and pawns: the further
advanced the pawns the better their chances. Hence the diagram
position is an easy win for White, no matter whose move, e.g. 1 . . . f5
2 Kg2 h5 3 Kg3 g6 (3 . . . g5 4 Kg2 ! reaches the position of move
8-one rank back) 4 Kg2 ! g5 (4 . . . f4 5 Kf3 g5 6 Kf2 g4 7 Kg2 f3 + 8
Kg3 h4 + 9 Kf2 ! h3 10 Kg3 and Black is in decisive zugzwang) 5 Kg1
g4 6 Kg2 h4 7 Kh2 f4 8 Kg1 ! One of the basic winning positions 8
. . . f3 (8 .
. g3 9 Kg2 + - ) 9 Kf2 h3 10 Kg3 + - when the black pawns
.
move they are captured, e.g. 10 . . . f2 1 1 Kxf2 h2 12 Kg2 g3 13 Kh1 !

g2 + 14 Kxh2 g1 = Q+ 1 5 Kxg1 and now Black must move his king.
We rely on Troitsky again to provide us with a game-like position
which utilizes the above theme:
MORE COMPLEX E N D I NGS 115
Diagram 86
Troitsky
1 924
White's winning chances lie in his potential passed pawn on the
h-file, but the winning procedure is quite subtle: t c6 ! (not t h4 ?
gxh3 e.p. 2 gxh3 KxcS 3 h4 Kd6 ! 4 hs Ke7 s h6 Kf8 wins, but not S . . .
Kf6 ? 6 fS ! when the zugzwang factor rears its ugly head again) t . . .
dxc6 2 d6 ! This clever pawn sacrifice blocks the black king route to f8
via KcS-d6-e7-f8 . 2 . . . cxd6 3 h4 gxh3 e.p. 4 gxh3 KcS s h4 Kds 6 hs
Ke6. The black king must stay in the •square' (see page 24) 7 h6 Kf6 8 fs !
Diagram 87
We now have the basic position akin to that of Diagram 8 S , Black
cannot move his king as the h-pawn queens, therefore White wins as
follows-8 . . . bs ! 9 Ke2 b4 to Kd3 dS tt Kc2 ! (not 11 Kd4 ? cs + t 2
Kd3 b3 ! wins) t t . . . d 4 t 2 Ket ! cs t 3 Kc2 ! c 4 t4 Ket ! when White
has achieved the standard winning position.
In Chapter t we mentioned the ' opposition', a well-known princi
ple in king and pawn endings. Now we shall take the principle one
stage further by citing a short example of the 'distant opposition' .
116 M O R E COM P L EX E N D I N G S
Diagram 88
Mattison
1 91 8
The problem is White to play and draw. The reader might like to
tackle this one before reading on. Although White will capture
Black's g-pawn, the black king is actively placed to win both white
pawns. However White has one variation which draws-1 hxg 5 +
(not 1 fxg5 + ? Kh5 2 Kg2 Kxh4 3 g 6 fxg6 with a simple win) 1 . . . Kh5
2 g6! fxg6 (2 . . . Kxg6 3 Kg2 Kf5 4 Kf3 draws easily) 3 f5 ! gxfs White's
pawn sacrifices have succeeded in bringing Black's remaining pawn
alongside his king, and now it is difficult for Black to follow the
maxim 'King in front of the pawn' 4 Kgl ! (the only move ! Both 4
Kh2 ? Kh4 ! 5 Kg2 Kg4 6 Kf2 Kf4 7 Ke2 Kg3 - + and 4 Kg2 ? Kg4 !
etc.-see Black with the opposition) 4 . . . Kg5 (Black keeps the
distant opposition. 4 . . . Kg4 5 Kg2 would give White the opposition
ensuring the draw) 5 Kfl ! ! Black can no longer maintain the distant
opposition (5 . . . Kf5 is illegal! ) and White draws easily 5 . . . Kf4 6
Kf2 The Opposition! 6 . . . Ke4 7 Ke2 f4 8 Kf2 , etc, when Black's king
will never be in front of his pawn again.
King and pawn endings are littered with examples of the opposi
tion and the reader is advised to make sure that he understands and
recognizes the importance of this vital principle.
We conclude this section with an example from the last century:
M O R E COM P L EX E N D I N G S 1 17
Diagram 89
H. Pillsbury-I. Gunsberg
Hastings
1 895
It seems as if Pillsbury has problems because his pawn on e6 looks

certain to fall, but he finds a clever resource: 32 e4 ! dxe4 33 d5 + all of
a sudden White has two connected passed pawns ! 33 . . . Kd6 (natur
ally not 33 . . . Kxd5 34 e7 + - ) 34 Ke3 b4 (34 . . . f5 is no better-3 5
gxf5 b4 36 f6 a4 3 7 f7 Ke7 38 d6 + Kf8 39 d7 Ke7 40 d8 = Q + + - ) 3 5
Kxe4 a4 3 6 Kd4 White keeps his king i n the 'square' of the a-pawn 3 6
. . . h 5 (This loses quickly. Gunsberg 's best try was 3 6 . . . Ke7 ! 3 7 Kc4
b3 38 axb3 a3 ! 39 Kc3 f5 ! creating a potential passed pawn. 40 gxf5 h5
41 b4! a2 42 Kb2 al = Q + 43 Kxat . Now when Black promotes his
pawn White will be in check! 43 . . . g4 44 b5 h4 45 b6 g3 46 hxg3
hxg3 47 d6 + ! Kxd6 48 b7 Kc7 49 e7 g2 50 b8 = Q + Kxb8 51 e8 = Q +
and White wins b y one tempo!) 3 7 gxh5 a 3 3 8 Kc4 (otherwise Black
might play . . . b3, e.g. 3 8 h6 ? b3 ! 3 9 h7 bxa2 40 h8 = Q al = Q + with a
draw) 38 . . . f5 39 h6 f4 40 h7 resigns.
QUEEN ENDINGS
It is not the objective of this book to offer a systematic coverage of all

the endings, so we shall only consider a few examples which the
author considers to be instructive.
Queen versus King and Pawn on the 7th Rank

This type of position can often arise at the end of a pawn race where
one side promotes and the other side is a move behind. Consider the
following situation:
118 M O R E COM PLEX E N D I N G S
m� •
�� •
t ·.. t.
� � . �� .
• • • •
� . . .
• • • •
• • • •
•
. •
. m
. t ! .,i '· , /,
•
m •
m •
m Diagram 90
This position would be a clear draw if it were Black to move 1 . . .
g l = Q. But White to play can initiate a series of checks to force the
black king on to the queening square, enabling the white king to
assist in the demise of the black pawn, e.g. 1 Kc6 + Kh3 2 Qa7
(threatening 3 Qgl when White has plenty of time to bring his king
over) 2 . . . Kh2 3 Qf2 Khl 4 Qh4 + Kgl 5 Kd5 Kft 6 Qf4 + Ke2 7 Qg3
forcing Black's reply 7 . . . Kfl 8 Qf3 + ; a standard position to aim for
8 . . . Kgl 9 Ke4 Kh2 10 Qf2 Khl 1 1 Qh4 + Kgl 12 Kf3 Kft 13 Qf2
mate. It was not even necessary to win the pawn! This winning
procedure always applies to pawns on the two centre files ( d- and e-) ,
and the knight files (b- and g-) . Pawns on the rooks (a- and h-) and
bishops (c- and f-); files offer the defending side drawing prospects
due to the stalemate resource, thus:
M O R E COM P L EX E N D I N G S 119
In the left-hand side of Diagram 91 White's winning procedure fails

as follows-1 Qf5 + Kb2 2 Qb5 + Kc2 · 3 Qa4 + Kbl 4 Qb3 + Kal and
now White cannot move his king because Black would be stale
mated. The same goes for the right-hand side of Diagram 9 1 , e.g. 1
Qg7 + Kh2 2 Qh6 + Kg2 3 Qg5 + Kh2 4 Qh4 + Kg2 5 Qg4 + Kh2 6 Qf3
Kgl 7 Qg3 + ; again the standard position is reached, and again the
defending king can go happily into the corner 7 . . . Khl for 8 Qxf2 is
stalemate.
This stalemate resource will generally enable the inferior side to
draw, unless the enemy king is close enough to construct a mating
net like this:
Diagram 92
With the white king inside the outlined area, it is possible to con
struct a mating net as follows-1 Qb2 + Kgl 2 Kf4 hl = Q 3 Kg3 !
when Black with equal material and the move, cannot prevent mate!
If the white king were on e4 instead, then the win is even simpler-1
Qb2 + Kgl 2 Ke3 hl = Q 3 Qf2 mate.
Queen versus Queen and Pawns

Because of the power of the queen, such endings often result in a
draw due to the very real possibility of perpetual check. For our first
example we shall consider a few episodes from a monumental strug
gle between the Icelandic Grandmaster Sigurjonsson and Britain's
top player Tony Miles, played at Hastings in 1976:
1 20 M O R E COM P L E X E N D I N G S
Diagram 93
G. Sigurjonsson-
A. J. Miles Hastings
1 975/6
Miles has just exchanged rooks on cl and now has the arduous task
of defending the resulting queen and pawn ending. Play con
tinued--49 Kh2 h4 this pawn move sets up various perpetual check
threats by controlling the g3 square. 50 Qd6 (Sigurjonsson defends
the b8-h2 diagonal and threatens 51 Qe7 winning the f-pawn) 50 . . .
Qe3 5 1 Qc5 Qg3 + 52 Kgl Qel + 53 Kh2 Qg3 + 54 Khl g5 5 5 Qe7 +
Kg6 56 Qe4 + f5 5 7 Qe6 + K g 7 58 Qxf5 Miles i s not worried about the
loss of his f-pawn because he now picks up the a-pawn 58 . . . Qel +
59 Kh2 Qg3 + 60 Kgl Qel + 61 Qfl Qe3 + 62 Qf2 Qcl + ! Miles is in no
hurry to capture the a-pawn, since he can tie White down first of all
63 Kh2 Qc7 + 64 Khl (not 64 g3 ? hxg3 + 65 Qxg3 Qc2 + with per
petual check in sight) 64 . . . Qcl + 65 Qgl . Now that White's forces
are bunched in the comer Black grabs the a-pawn. 65 . . . Qxa3 66 Qd4 +
Kf7 67 Qd6 Qcl + 68 Kh2 Qc4 69 Qb8 g4 70 hxg4 Qxg4. The exchange of
pawns takes Black closer to his goal. 71 Qc7 + Ke8 72 b5 h3--offering
the h-pawn to expose the white king. 73 Qe5 + Kd 7 74 gxh3 Qc4 .
We jump ahead to move 103 where we note that White has been
able to make very little progress, and even has to ditch his b-pawn:
MORE COMPLEX E N D I N G S 1 21
The game continued-103 Qe6 + Kxb5 1 04 Qe5 + Ka4: keeping the

king out of harm's way! 105 h5 Qf3 + 106 Ke6 Qc6 + 107 Kf5 Qc8 +
108 Kg5 Qg8 + 109 Kf4 Qf7 + 1 1 0 Ke4 Qb7 + 1 1 1 Qd5 Qe7 + 1 1 2 Kd4
White is moving towards the black king so as to be able to interpose a
check with a counter-check 1 1 2 . . . Qg7 + 1 1 3 Kc5 Qf8 + 1 14 Kc4
Qf1 + . We jump ahead again, and note that Miles's patient man
oeuvring has prevented White from making progress:
Diagram 95
Play continued-1 36 . . . Qf4 + 1 3 7 Qd6 Qc4 + 1 3 8 Qc6 + Ka5 ! (A
lovely touch of humour from Miles-1 3 9 Qxc4 is stalemate! ) 1 3 9 h6
Qf7 + 140 Qd7 Qc4 + 141 Kb8 (141 Qc6 Qf7 + repeats moves) 141 . . .
Qb4+ 142 Kc8 Qc5 + 143 Qc7 + Ka6 ! Again! ! 144 h7 Qf8 + 145 Qd8
Qc5 + Draw agreed. 146 Qc7 Qf8 + 147 Kd7 Qf7 + 148 Kd6 Qf4 + 149
Kc6 Qc4+ draws, as does 146 Kd7 Qf5 + winning the h-pawn. So
after nearly 100 moves of patient defence Miles got his half-point
reward. An important principle in all queen and pawn endings, is to
keep the queen actively placed in the centre where she can be most
effective while restraining her counterpart. Miles's play in this
example illustrated the point admirably.
Our final example in this section is taken from one of the author's
games:
1 22 M O R E C O M P L EX E N D I N G S
Diagram 96
G. D. Keane-Povah
London
1 976
White played 26 h3 providing an escape for his king and play
continued-26 . . . Qc5 ! 2 7 Bxd4 (the alternatives 2 7 Bd2 loses the
b-pawn and 2 7 Bg5 h6 makes matters worse) 2 7 . . . exd4! This pawn
recapture, which White had failed to consider, is very strong 28 Qd3
(not the suicidal 28 Kgt ?? d3 + 29 Qf2 d2 30 Qxc5 dt = Q + wins) 2 8
. . . b 5 ! 2 9 Kh2 ( 2 9 b 3 Qc3 i s hopeless, while 2 9 Kgt Qct + 30 Kf2
Qxb2 + is no better) 2 9 · . . . Qc4 30 Qa3 White's only hope lies in
activating his queen so as to set up some perpetual check threats. 30
. . . b4 keeping the white queen out of play: 3 1 Qa4 h6 stopping mate
and providing a safe haven for the king 32 Kg3 ? (32 Qe8 + Kh7 33
Qf8! offered better prospects, because if White can tie Black down
with perpetual threats, then he can safely activate his king to deal
with Black's d-pawn, e.g. 33 . . . Qe6 34 Kgl d3 35 Kf2 with good
drawing chances. But 33 . . . Qc5 ! is much stronger, e.g. 34 Kgt d3 +
35 Kft Qct + 36 Kf2 Qxb2 + 37 Ke3 Qe2 + 38 Kd4 Qxg2 offers good
winning chances) 32 . . . d3 33 Kf4 d2 34 resigns. The d-pawn is too
strong, e.g. 34 Qd t ? Qct wins immediately, as does 34 Ke3 Qct
while 34 Qe8 + Kh7 35 Qf8 is ineffective 35 . . . d t = Q 36 Qf5 + Kh8
37 Qf8 + Qg8 - + .
ROOK ENDINGS
Rook and pawn endings are probably the most common endings in
chess, because the rooks are quite often the last pieces to be
exchanged off. Rook endings are frequently drawn, sometimes even
when one side has an extra pawn, because rooks are extremely
MORE COMPLEX ENDINGS 1 23
efficient at demolishing enemy pawns. We shall consider two of the

most important basic positions in rook endings where one side has
an extra pawn, since many rook endings will simplify into one of
these cases :
Philidor's position
Diagram 97
Philidor
1 777
This position is drawn because Black's king controls the queening

square, and accurate defence will make it impossible for White to
improve his position. Philidor demonstrated the correct defensive
procedure over 200 years agcr--1 . . . Ra6 ! Black prevents the white
king from advancing, and since rook checks can be answered by . . .
Ke7 . . . Ke8 , etc . , White must play 2 e6 (any attempts to drive the
black rook off the 3rd rank fail as follows-2 Rg7 Rb6 3 Rg6 Rxg6! 4
Kxg6 Ke7 5 Kf5 Kf7 with a drawn king and pawn ending) 2 . . . Ra1 !
This move ensures the draw because Black will start a relentless
series of rook checks from behind, thus 3 Kf6 Rf1 + 4 Ke5 Rel + 5 Kd5
Rd1 + 6 Kc4 Rc1 + 7 Kd3 Rel ! 8 Rh6 Ke7 wins the e-pawn.
The drawing principle in this ending may be summed up-'place
the defending rook on the third rank to prevent the enemy king from
advancing. When the pawn goes to his 6th rank the rook is trans
ferred to the bacl( rank to commence checking from behind. '
It i s worth noting that the position in Diagram 97 is still drawn
with White to move-- 1 Kf6 (the best winning try, since 1 e6 Rf1 +
and 1 Ke6 Ra6 + both lead to positions already considered) 1 . . . Rel !
2 Rh8 + ( 2 e 6 Rf1 + is all right for Black) 2 . . . Kd7 and White cannot
advance his e-pawn, so . . . 3 Rh7 + Ke8 = .
1 24 M O R E C O M P LEX E N D I N G S
Lucena's position
Diagram 98
Lucena
1 497
Lucena analysed the diagram position nearly 500 years ago and he
demonstrated two different winning procedures. We shall consider
one of them-1 Rfl + this check drives the black king away so that
the white king can emerge---1 . . . Kg7 (not 1 . . . Ke6 ? 2 Kea when
White queens) 2 Rf4! (this is the star move ; White prepares to bring
his king out of hiding. If he tries the immediate 2 Ke7 then he runs
into trouble as follows-2 . . . Re2 + 3 Kd6 Rd2 + 4 Kc5 Rc2 + 5 Kb5
Rb2 + 6 Kc4 Rd2 wins the d-pawn) 2 . . . Rel what else? 3 Ke7 Rel +
4 Kd6 Rdl + 5 Ke6 Rel + 6 Kd5 Rdl + 7 Rd4 ! The p �int! The white
king shelters behind his rook, and his pawn promotes. This winning
procedure is known as 'the bridge building' method.
We now turn our attention to a few examples of rook endings with
several pawns:
Diagram 99
D. L. Meen-Povah
Hammersmith
1 972
This particular example made an effective impression upon the

author, who overheard a well-known British player saying, 'that
looks drawn'-it was nice to prove him wrong! If we remove the f-, g
and h-pawns from the board, then the position is undoubtedly
drawn; White simply plays Kg2-h2-g2 , etc . , and if Black marches to
b2 to try to free his rook from its guard duty, then White starts
checking with his rook as far away as possible, e.g. Rb8-a8-c8 + etc.
However, with the king-side pawns Black has a winning ploy-52
. . . h5 ! Black sacrifices his h-pawn to create a passed f-pawn. 53
gxh5 + (not 53 Rxh5 when 53 . . . Rbl allows Black to promote) . 53
. . . Kh6 54 Kg2 (54 Ra6 Kxh5 doesn't help White because he cannot
play 55 Rxf6 due to 55 . . . Rbt again) 54 . . . f5 55 Kh2 f4 56 resigns.
The f-pawn is immune to capture because the white rook dare not
leave the a-file, and 56 Kg2 f3 + 57 Kxf3 loses to 57 . . . Rft + and
at = Q or 57 Kf2 Rht ! 58 Rxa2 Rh2 + wins.
Our next example was described by Irving Chemev as 'the most
instructive rook and pawn ending ever played'. He may well be right
as the reader may judge for himself:
Diagram 1 00
S. Tarrasch-E. Thorold
Manchester
1 890
Tarrasch who perpetrated many tactical brilliances is a pawn up in
this tricky ending, and he now treats us to a demonstration of sheer
perfection in endgame technique---2 6 Re3 ! (the tempting 26 Re7 +
Kd6 2 7 Rxg7 Rxd3 2 8 Rxh7 Rdl + 2 9 Kf2 Rd2 + 30 Kf3 Rxb2 leaves
White on top, but Black would have more drawing chances than he
deserves. Tarrasch's move keeps the black rook out of the white
position) 26 . . . Kd7 2 7 Kf2 g6 28 Rh3 ! This forces Black to weaken
his pawns further, and enables the white king to relieve the rook of
' his defensive duties. 28 . . . h5-now the g-pawn is a target. 2 9 Ke3
Rd6 3 0 d 4 Re6 + 3 1 Kd3 Rea 32 Rg3 Re6 Black's rook i s tied down to
defence of the g-pawn. 33 Re3 Rd6: naturally, Black cannot exchange
rooks while being a pawn behind. 34 Re5 Rf6 35 a4! Tarrasch moves
his pawns away from the vulnerability of the second rank. 3 5 . . . Rf2
36 Re2 Rf6 3 7 b4 Rf1 3 a Re5 Rf2 (If Black tries to attack the queen-side
pawns by 3a . . . Ra1 or 3 a . . . Rb1 then White answers 39 a5 or 39 b5)
39 Rg5 Rf6 (not 39 . . . Rb2 40 Kc3 Ra2 41 Rxg6 Rxa4 42 Rxb6 + - ) 40
h3 ! (Tarrasch has played the ending with great skill and patience,
and this move is no exception. He wishes to play 40 d5 but 40 . . . Rf4
is irritating, while 40 Ke4 allows 40 . . . Re6 + 41 Re5 Rc6 with
counter-chances. So he waits for Black to commit himself, and he
does.) 40 . . . Kd6 41 Ke4 Re6 + 42 Re5 Rf6 (42 . . . Rc6 is not possible
now) 43 d5! Kd7 44 Rg5 Kd6 (otherwise 45 Ke5 is embarrassing) 45
Rg3 Ke7 46 Rf3 ! This move drives the black rook off the useful f-file.
46 . . . Rd6 47 Ke5 Rda (White threatened 4a Rf7 + Kxf7 49 Kxd6 Kea
50 Kc7 with a simple win) 4a d6 + ! Kd7 (4a . . . Rxd6 fails to the
aforementioned resource 49 Rf7 + Kxf7 50 Kxd6 when the white king
would rapidly devour Black's queen-side pawns) . 49 Rf7 + Kea 50
Rc7 + Kha 51 Rc2 Rea + 52 Kf6 b5 53 d7 Rha (not 53 . . . Rda 54 Ke7
+ - ) 54 Ke7 Rh7 + 55 Kd6 Rha 56 Re2 resigns. 57 Rea + is imminent.
Tarrasch's technique made this ending look so easily won that it
seemed that White could hardly fail, but anyone who has played
rook and pawn endings will know just how difficult they can be.
Finally, an amusing and instructive problem by Saavedra:
uiagram 1 01
F. Saavedra
1 895
A rook will generally win against a lone pawn un less the king cannot
assist in preventing the pawn from promoting, in which case the
rook will have to be Sacrificed.
M O R E C O M P LEX E N D I N G S 1 27
The diagram position shows one of those rare occasions when it is

not even possible to sacrifice the rook. Play continues-1 c7 Rd6 +
(forced since . . . Rd8 and . . . Rc5 both lose) 2 Kb5 ! (not 2 Kb7 Rd7 =
or 2 Kc5 Rdt followed by . . . Rel) 2 . . . Rd5 + 3 Kb4 Rd4 + 4 Kb3
Rd3 + 5 Kc2 ! Now Black can no longer rely on the skewer on the c-file
and . . . Rd8 is hopeless so all looks lost, but-5 . . . Rd4! (Very
clever! If White promotes to a queen Black draws 6 c8 = Q Rc4 + ! 7
Qxc4 stalemate! ) 6 c8 R! Now we have one of those rare cases where
=
a win is possible in a rook versus rook ending. 6 . . . Ra4 (forced to

prevent 7 Ra8 mate) 7 Kb3 ! and White wins either the rook or forces
mate with 8 Rel . A clever conception!
MINOR PIECE ENDINGS
Minor piece endings are probably the least understood endings of

all, simply because it is hard to draw upon useful generalizations.
We will consider some special cases.
Bishops of opposite colour

These endings are often thought to be ' drawish' because each player
controls one colour complex, making it difficult for progress to be
made. It is worth enumerating some basic themes.
1. Bishop and pawn versus bishop is almost certainly a draw.
2 . With many pawns , one extra pawn is usually not enough to win,
unless there are special circumstances and there is play (pawns) on
both wings (see Diagram 107).
3. Two extra pawns does not always guarantee a win, but winning
prospects are increased the more files there are between the pawns.
(Connected pawns are better, however, than the worst case when the
pawns are separated by only one file.)
4. Two connected pawns will generally only win if they are already
on the fifth rank, unless the inferior side's pieces are poorly placed.
Salvioli (1889) demonstrated the winning procedure for discon
nected pawns at lea.st two files apart:
• • •
. . � -� .
•t• • •
• -�· ft · ·
• • • •
• • • •
• • • • Diagram 1 02
C. Salvioli
• • •• • 1 889
Black has both pawns blockaded at the moment, but White breaks
the blockade by forcing the advance of his f-pawn-1 Bf3 the c-pawn
must be protected to free the king. 1 . . . Bh4 2 Ke6 Kd8 3 f6 Bg5 4 f7
Bh6 5 Kf6 Bf8 6 Kg6 Ke7 7 Kh7 Kd8 (not 7 . . . Kxf7 8 c7 + ) 8 Kg8
-
wins. There is one important exception to this particular ending, if

one of the pawns is 'the wrong-coloured rook pawn'.
Diagram 1 03
wh;te cannot win as in the last example because aft�r-1 Kc5 Bf3 2
d5 Bxd5 3 Kxd5 Kb7 4 Kc5 Ka8 we have the familiar drawn 'bishop
and wrong rook pawn' position, since the black king can never be
evicted from a8. The reader is advised to check this for himself.
Another drawn case exists if the inferior side can blockade the
pawns along a single diagonal. Consider the following case:
MOR E COM PLEX END INGS 1 29
Diagram 1 04
The standard plan of using the king to remove the bishop's blockade
fails as follows-1 Ke4 Kc6: the bishop on f7 holds up White's
c-pawn as well. 2 Kf5 Kd6 3 Kg5 Kd 7 4 Kh6 Ke8 5 Kg7 Bc4 and White
has made no progress. No better is-1 Be3 Bg8 2 Kd4 Bf7 3 Ke5 Kc6 !
with no further progress possible.
We now consider connected pawns. Diagram 105 shows a typical
case where the win is possible due to the poor placement of the
defender's pieces:
Diagram 1 05
White plays-1 Bb5 + (this check controls two important white
squares in front of the pawns c6 and d 7; therefore it is more effective
than Bg4 + ) 1 . . . Ke7 2 Ba4 (this prepares the white king's march to
the queen side. Both 2 d6 + ? Bxd6 and 2 c6 ? Kd6 would draw) 2 . . .
Bf4 3 Kc4 Bg3 4 Kb5 Kd 7 5 Kb6 + Ke7 6 Kc6 Bf4 7 Bb3 Bg3 8 d6 + Kd8
9 Kd5 when 10 c6 will enable White to achieve the winning position
with two pawns on the sixth rank.
1 30 M O R E COM PLEX E N D I N G S
If Black had adopted the correct defensive stance as shown in

Diagram 106 then he could have ensured the draw:
Diagram 1 06
White can make no progress since (a) d6 is always met by the Bishop
sacrifice . . . Bxd6 (b) the white king is tied down to the defence of the
c-pawn and (c) c6 gives up the dark squares for ever. Therefore, all
Black needs to do is 'pass' with Be7-f�7, etc. From this we can
deduce that if White's pawns were on the sixth rank, Black would
not have the 'pass-move' and he would be in zugzwang. Check this
for yourself!
In case you think that 'bishop of opposite colour' (BOC) endings
are easily drawn if you are one pawn behind, take heed from the
following example:
Diagram 1 07
P. Walker-Povah
Inter-county
1 972
Both my opponent and the adjudicator thought this position was
drawn, after all BOC endings are always drawn, aren't they? How-
MORE COM P L EX E N D I N G S 1 31
ever, this one is far from simple. Apart from his extra pawn, Black
has several factors in his favour. (a) White's king is passively trapped
in the corner, (b) White has a weak h-pawn and (c) There is play on
both wings. The reader should try to find Black's unusual winning
strategy, paying particular attention to the aforementioned points ,
for without any one of them the position would probably be drawn:
e.g. if White 's pawn were on h2 instead of h4 the position is drawn.
The winning procedure was as follows: 60 . . . f5 61 Kh2 f4 62 Khl
Kg4 63 Bf6 (White 's bishop is now rooted to this square because he
must protect the weak h-pawn, and he dare not leave the a1-h8
diagonal otherwise . . . c3 wins). 63 . . . Kh3 64 Kgl f3 65 Khl Bfl 66
Kgl Bg2. The point! White is in zugzwang! Play might continue 67
Bc3 (not 67 Bg5 c3! 68 Bf6 cxb2 69 Bxb2 Kxh4 - + ) 67 . . . Kxh4 68
Kh2 Kg4 69 Bd4 h4 70 Bc3 Kf4 71 Bf6 Ke4 72 Bc3 (note that White's
king is imprisoned on gl and h2 , and 72 Bxh4 c3 is fatal) 72 . . . Kd3
73 Bf6 Ke2 74 Kgl (74 Bd4 Kfl would be zugzwang again! ) 74 . . .
Bh3 (74 . . . h3 ? 75 Bd4 ! Kel 76 Bc3 + Ke2 77 Bd4 gets Black
nowhere) 75 Bd4 Bf5 76 Bf6 (76 Kh2 Kfl , 77 Khl Bh3 78 Kh2 Bg2
zugzwang! ) 76 . . . h3 77 Bd4 Be6 zugzwang again! 78 Kh2 (78 Bf6
h2 + ! 79 Kxh2 Kxf2 80 Bd4 + Ke2 81 Kg3 Bd5 zugzwang, since Black
will force either . . . f2 or . . . c3) 78 . . . c 3 ! 79 Bxc3 ! (naturally not
79 bxc3 b2 winning) 79 . . . Kxf2 80 Bd4 + Ke2 81 Kg3 h2 82 Kxh2 f2
83 Bxf2 Kxf2 84 Khl Ke2 and Black's b-pawn decides the game. As
can be seen from the above analysis , the win was quite involved,
but it was there to be found.
BISHOP VERSUS KNIGHT ENDINGS
It is a question of eternal debate which is better, the knight or the

bishop? Obviously there is no hard and fast answer, since the posi
tion will always dictate which is to be preferred. In general we may
observe that the bishop prefers ' open' positions with few pawns,
while the knight has a preference for blocked positions with many
pawns.
Our first position shows a knight triumphing over a bishop
because Black's bishop was severely restricted by his own pawns :
Diagram 1 08
R. Emerson-G. D. Lee
National Club
Championship
1 979
White realizing the superiority of his knight over Black's bishop
offers the exchange of rooks. 30 Rel Bc6: Black decides to avoid the
exchange and White now gives an instructive display of how to
handle the classic 'good' knight against 'bad' bishop. 31 Nd4 Kd7 32
h4 White gradually 'fixes' Black's pawns on light squares. 32 . . . Rga
33 Rc3 ! White's rook takes up residence on the third rank in order to
harass Black's pawns at the appropriate moment-33 . . . Rha 34 g3
Rga 35 f4 Black is well and truly tied up, his pieces are passively
placed with no enemy targets to attack. 35 . . . Rea: This will allow
the black king to go for a wander, but where is he going to go? 36 Kg2
Kc7 37 g4! White starts his breakthrough on the king side while
Black's forces are occupied. 37 . . . Kb6 3a Kg3 Bd7 39 g5 Rha ! This
prevents White's intended h5 and it seems as if his position has
reached an impasse, but he can still strengthen his grip of the dark
squares. 40 b4 ! Rh7 41 R� 5 (restraining 41 . . . a5 and giving him
absolute control of the dark squares) 41 . . . Re7 ? (A fatal error
allowing a quick breakthrough. 41 . . . Rha was better when White
could try 42 a4! threatening a5 + and Rc7 when Black would be
completely stymied.) 42 h5! This temporary pawn sacrifice gives
White a powerful passed g-pawn. 42 . . . gxh5 43 Kh4 White's abso
lute control of the dark squares means that all the adjacent light
squares are vulnerable. 43 . . . Bea 44 Rea. Finally White's rook
penetrates Black's position--a victory for patience! 44 . . . a5 45 a3
axb4 46 axb4 resigns. Black hardly has a move, e.g. 46 . . . Bg6 47 Rga
Bea 4a g6 intending Kxh5 and Kg5-f6+ - or 46 . . . Ka6 47 Rxea!
Rxea 4a Kxh5 when White's g-pawn proves too strong for the black
rook.
Our next example illustrates the superiority of a bishop over a
M O R E C O M P L EX E N D I N G S 1 33
knight in an endgame where there are pawns on both sides of the

board:
Diagram 1 09
Bellin-Povah
Borehamwood
1 978
Black in fact enjoys the aforementioned advantage twice over, for he
has two bishops against two knights! White, therefore, feels that it
would be wise to exchange off one of the bishops, but in so doing he
gives Black a passed pawn-an even greater evil! Play continued 41
Nxc4 ?! dxc4 42 Ke4 Ke6 43 d5+ ?! This pawn advance is dubious on
two counts-first it increases the scope of Black's bishop and second
the d-pawn is weak. 43 . . . ·Kd6 (Black's advantage is clear since he
threatens 44 . . . f5 + 45 Kf3 Bd4 followed by 46 . . . Kxd5 - +) 44 g4
f5 + ! Black temporarily sacrifices a pawn to force a breakthrough on
the queen side. 45 gxf5 gxf5 + 46 Kxf5 Bd4 47 Nb5 + Kxd5 48 Nc7 +
White dare not exchange pieces because Black's c-pawn is much too
quick for the white f-pawn 48 . . . Kc6 (not 48 . . . Kc5 ? allowing 49 '
Ne6 + Kd5 50 Nc7 + with no progress) 49 Ke4 Bgl; with his king
active White would like an exchange of minor pieces; but Black is
not having any of that! 50 Ne6 Bxh2 51 Kd4 (The power of the bishop
is illustrated by a variation such as 51 f5 c3 52 f6-not 52 Kd3 ? Be5
when Black's h-pawn and mobile king should see him win-52 . . .
Bd6 53 f7 h5 58 Kd3 h4 59 Ng5 Bb4 when the bishop is performing
several useful functions until the king can be activated) 51 . . . Kb5
52 Nc7 + (if 52 f5 Kb4 53 a3 + Kxa3 54 Kxc4 a4 55 f6 Kb2 56 f7 Bd6
and the bishop can be sacrificed for the f-pawn leaving White's
knight the impossible task of restraining two ambitious rook pawns;
alternatively 53 f6 c3 54 f7 Bd6 55 a3 + Kb3 56 Nc5 + Kxa3 57 Nd3
Kb3 - +) 52 . . . Ka4 (52 . . . Kb4 allows 53 Nd5 +) 53 Kxc4 Bxf4
54 Ne6 Be3 ! notice how the bishop keeps the knight at bay 55 Nd8
Ka3 56 Nc6 a4:
1 34 M O R E C O M P L EX E N D I N G S
Diagram 1 1 0
Bellin-Povah
There are very few pawns left so the win is tricky, for Black must
watch out so as not to be caught with bishop and 'wrong' rook pawn
(i.e. bishop and h-pawn would be drawn if the White king reaches
ht). White tries to keep his a-pawn with 57 Nb4 ?! (he would have
had better chances with 57 Kb5 h5 58 Ne5 Bf4 59 Nf3 Bg3 60 Ng5 h4
6t Nh3 Bet ! 62 Nf4 Bd2 63 Nh3 Be3. Again the bishop restrains the
knight, and in fact this time the knight is trapped, so White's move is
forced 64 Ka5-other moves would allow 64 . . . Kxa2 without loss of
Black's a-pawn-64 . . . Kb2 ! ! Black cannot afford to capture the
a-pawn if he wants to win the race to the trapped knight. 65
Kxa4-not 65 Kb4? a3 winning easily-65 . . . Kc3 66 Kb5 Kd4 67
Kc6 ; obviously not 67 a4 Ke4 68 a5 Kf3 69 a6 Kg4 ! 70 a7 Bxa7 - +
when we see the strength of the bishop who is capable of monitoring
both sides of the board at once; 67 . . . Ke4 68 a4 Kf3 69 Kd5 Kg4 70
Ke4 Kxh3-not 70 . . . Ba7 ? 7t Nf4 with equality-7t Kxe3 Kg3 72 a5
h3 73 a6 h2 74 a7 ht = Q winning by one tempo!) 57 . . . Bd2 58
resigns. The conclusion might have been 58 Nc2 + Kxa2 59 Kb5 Kb3
60 Nd4 + Kc3 6t Ne2 + Kd3 when one of Black's pawns will reincar
nate.
We now conclude this chapter with a brief examination of several
important and common situations.
1. QUEEN VERSUS ROOK (AND PAWN)
A, queen will nearly always win against a rook unless the defending
side can catch his opponent with a stalemate trap. Consider Diagram
ttt:
Diagram 1 1 1
In order to win , White must force his opponent's king to the edge of
the board-1 QeS + Kd7 2 KdS Rc7 other moves are naturally pos
sible but they come to a similar conclusion: 3 Qe6 + Kd8 4 Qg8 + (not
4 Kd6 ?? allowing Black one of his resources 4 . . . Rc6 + ! 5 Kxc6
stalemate) 4 . . . Ke7 5 Qg7 + Kd8 6 Qf8+ Kd7 7 Qf4 ! (The simplest
winning move, although both 7 Qh8 and 7 Qb8 win) 7 . . . Kea 8 Kd6
Kb8 9 QeS ! Rb7 (no better are 9 . . . Rc8 10 Kd7 + ! Kb7 11 Qbs + + - or 9
. . . Rc2 10 QbS + winning the rook next move; all other moves have a
similar fate) 10 Kc6 + Ka8 1 1 Qal + Kb8 (not 1 1 . . . Ra7 12 Qh8 mate)
12 QaS ! We now have another position analysed by Philidor, who
proved the win back in 1 7 7 7 ! 12 . . . Rbl (the reader might like to find
the winning sequences after 12 . . . Rh7, 12 . . .. Rb3 and 12 . . . Rf7;
naturally 12 . . . Rb2 loses instantly to 13 QeS + forking king and
rook) 13 Qd8 + Ka7 14 Qd4 + Ka8 15 Qh8 + Ka7 16 Qh7 + picks up
the rook. A queen: should always be able to force the win against a rook,
unless an exceptional case such as that given below were to arise:
Diagram 1 1 2
This position i s amusingly drawn with Black t o move-1 . . . Rh7 + 2

Kg2 Rg 7 + 3 Kf3 Rf7 + 4 Kg4 (Ke4 Re7 is no good) 4 . . . Rg 7 + 5 Kf5
Rf7 + 6 Kg6 Rg 7+ and now 7 Kf6 Rg6 + draws as already mentioned,
and so does 7 Kh6 Rh7 + ! 8 Kxh7 stalemate. If it were White to move,
then the position is naturally won.
The case of queen against rook and pawn is not so clear cut, and we
will leave the study of this intriguing ending to the reader. Suffice it
to say that in general the ending is a win, but it is possible to draw if
the pawn is unmoved and the rook can keep the enemy king at bay.
Philidor demonstrated in 1803 that Diagram 1 1 3 is a draw:
Diagram 1 1 3
Philidor
1 803
Black can play . . . Rc6 and back to e6 so zugzwang is ruled out and
progress for White is impossible, e.g. 1 QbB + Ke7 2 QgB Rc6 3 Qg7 +
Kd8 4 QfB + Kc7 5 Qa8 Re6 , etc.
2 ROOK AND MINOR PIECE VERSUS ROOK
This particular ending of rook and minor piece against rook is

usually drawn, although there are certain winning positions-more
so with the bishop.
Rook and Bishop versus Rook

This case is quite common in tournament play, and because the
defence is difficult it often results in a win; therefore we shall
concentrate on the correct defensive procedure:
Diagram 1 1 4
Flohr-Reshevsky
Semmering-Baden
1 937
Play continued-1 . . . Rd2 2 Ke5 Kc8 ! 3 Bc5 Rd7 4 Be7 Kb7 5 Ke6
Kc6 6 Rht Rd2 7 Rel + Kb5 8 Bd6 Re2 + and it was clear that White
was getting nowhere; in fact a further ten moves saw them reach a
position analogous to the diagram and a draw was agreed-9 Kd7
Re4 10 Rc5 + Ka4 1 1 Kc6 Kb3 12 Kd5 Re8 1 3 Rb5 + Kc2 14 Bc5 Kd3 1 5
Rb3 + Ke2 16 Bd4 Rd8 + 1 7 Ke4 Re8 + 18 Be5 Ket 1 9 Rb2 Re7 H. By
maintaining the pin on the bishop White cannot set up mating
threats. The reader should guard against over-simplification, for this
ending deserves careful study from specialist endgame books.
Rook and Knight versus Rook

The defence here is even easier because the defending king cannot
be driven to the edge of the board as is shown in the following
example:
Diagram 1 1 5
Toth-Pritchett
EEC Team Toumament
1 978
1 38 M O R E COM PLEX E N D INGS
The players entered this ending on move 7 6 and accurate defence by

Pritchett had taken them to this position. Play continued 90 . . . Kg4
91 Rea Kf3 92 Rc3 + Kg4 93 Rd3 Rc5 94 Ra3 Rb5 95 Ke4 Rb1 96 Ra5
Kg3 9 7 Ra4 Rba 9a Nf6 Rb3 99 Nd5 Rba 100 Ra3 + Kg4 101 Ne3 + Kg5
102 Ra5 + Kg6 and Toth soon gave up in despair.
If the defending king is already on the edge of the board then a win
is sometimes possible, but the complexity of this ending is such that
we could not do it justice here. Again the diligent reader should
consult basic endgame texts.
3 ROOK VERSUS MINOR PIECE
This ending is again usually drawn, but as always there are excep
tions when the defending king is on the edge of the board, so beware!
Rook versus Bishop

Diagram 116 shows two contrasting cases:
Diagram 1 1 6
Draw Win
First, on the left-hand side White cannot win, although his pieces are
well placed , because 1 . . . Bba 2 Rda is stalemate and 2 Rg5 Bd6 3
Rga + Bba always threatens a stalemate.
In the second case on the right-hand side the win is trivial 1 . . .
Bd6 2 Rea + Bfa 3 Raa and Black is in zugzwang 3 . . . Kha 4 Rxfa
mate. From these two cases we can draw this conclusion:
If the defending king is forced to the edge of the boo.rd he should
aim for the corner not controlled by his bishop, so that he can block a
check and set up a stalemate position. This principle, in turn,
implies that the superior side should try to prevent the defending
king from reaching the 'safe' corner, and Kling and Horwitz give an
impressive example of how this may be achieved:
Diagram 1 1 7
Kling and Horwitz
1 851
White wins no matter where Black places his bishop, so long as the
black king does not cross the f-file to seek sanctuary on aB-1 . . . Bgt
(other bishop moves would allow a double threat of mate and loss of
the bishop, e.g. 1 . . . Bc3 2 Rc7 wins) 2 Rft ! This forces the bishop
into the open 2 . . . Bh2 3 Rf2 Bg3 (3 . . . Bgt 4 Rg2 is the same) 4 Rg2 !
Bd6 (if instead 4 . . . Bf4 or 4 . . . Bh4 then 5 Kf5 + or 5 Kh5 + is decisive)
5 Rd2 Bc7 (5 . . . Be7 6 Ra2 ! wins since 7 Ra8 + cannot be prevented
and unfortunately 6 . . . KfB 7 Ra8 + would force 7 . . . Bd8 to provide
a flight square for the black king) 6 Rd7 Bb6 7 Rb7 wins.
Rook versus Knight

More often than not drawn, but there are a few exceptions. Diagram
118 shows two drawing examples:
. -�.
• • •
. -�· .
. � .
• • • •
,, " J • • •
. . . �
• 'lia"
� -, • �
m -
� Diagram 1 1 8
1 40 MORE COMPLEX ENDINGS
In the bottom half of the diagram the draw i s simple but play must be
precise--1 Rb2 + Kal 2 Rb8 (if 2 Rh2 Kbl 3 Rd2 Kal 4 Kb4 Kbl 5 Kc3
Na2 + ! draws, but not 5 . . . Kal ? 6 Rdl Kbl 7 Rhl zugzwang) 2 . . .
Ne2 ! (the only move and not all that obvious! 2 . . . Na2 ? 3 Kb3 Kbl 4
Rb7! Ncl + 5 Kc3 + Kat 6 Kc2 Na2 7 Rbl is mate while 2 . . . Nd3 ? 3
Kb3 Ncl + 4 Kc2 reaches the same end) 3 Kb3 Kbl ! This time the
black king is threatening to escape via c l . 4 Re8 Ncl + 5 Kc3 Na2 +
and Black gives perpetual check or gets his king off the back rank.
The case in the top half of the diagram is also drawn, but a bit more
complicated-1 Rb7 Nh6 2 Rh7 Ng8! The knight must stay near his
noble monarch. 3 Rf7 + Ke8 4 Rg 7 Kf8! (forced , since 4 . . . Nh6 5 Rg6
wins the knight) 5 Rh7 Ke8 6 Rf7 Nh6 7 Rfl Ng8 8 Rat Kf8 9 Ra7 and
White has gone full circle without making progress.
That concludes this chapter on more complex endings , but it
cannot be over-emphasized that the ambitious student must study
these endings in greater depth if he is to tip the balance on those
frequent half-points which are on the win/draw borderline.
General Advice
In this chapter we hope to cover a number of points which are not

normally mentioned in the average instructional text. Since this
book aims at helping the young ambitious player to improve his
game, we shall consider some important ideas which are too often
neglected.
HOW DO CHESS PLAYERS ANALYSE?
This question must have intrigued almost everyone who knows

anything about chess. The novice often asks the strong player, 'how
many moves do you "see" ahead?' Obviously there cannot be a
standard reply because the answer will always depend on the nature
of the position under consideration. Yet one would have thought
that amid the hundreds of thousands of books written on chess (to
which this is a new addition) there would exist a comprehensive
study on the methods of analysing chess positions. Unfortunately,
however, very few authors have attempted to tackle this highly
complex topic, but one who should be mentioned is the Soviet
Grandmaster Alexander Kotov. In his excellent book Think Like a
Grandmaster he explains 1that naturally there are various ways of
approaching an analysis of a given position. One important feature
that does emerge from his exposition is that a logical approach to any
position will generally give the best results. We shall, therefore,
devote some space to what Kotov calls 'the tree of analysis' and the
selection of candidate moves. For this purpose let us return to a
position considered in Chapter 2 (see page 38) :
1 42 G E N E R AL ADV I C E
Diagram 1 1 9
Let us assume that a Master is confronted with this position for the
first time and that he is to provide a fairly accurate and comprehen
sive analysis of this position, with -Black to move.
His first observations would, needless to say, be general comments
such as 'Black is way behind in development' (about 5 tempi) and
'White has good domination of the centre and is poised to attack' : the
material balance being level means that White has a clear advantage.
This assessment procedure would take only a matter of seconds.
Next the Master would try to pinpoint weaknesses in Black's
position in order to formulate some attacking plans for White. Black,
for instance, has a weak and vulnerable pawn on b5, and weak
squares on d5 and c6. Furthermore, Black's king in the centre is an
obvious target. At this stage the Master would want to start analysing
some concrete variations to back up his observations , and he would
probably begin by mentally listing Black's different options in the
diagrammed position. These moves are referred to as 'candidate
moves', for it is one of these that Black will eventually choose to play.
His thoughts could run as follows, 'Well, Black is threatened with
the loss of his b-pawn by 1 1 Nxb5 so unless he is going to sacrifice a
pawn he must take account of the threat. He could defend b5 with 10
. . . Rb8, 10 . . . c6 or 10 . . . Qd7 , but none of those does much for his
development. Black could try 1 0 . . . c5 counter-attacking White's
queen, but that is anti-positional because it creates a weak pawn on
d6 and a weak square on d5; what is more it doesn't solve the
problem of the b-pawn. He could try 10 . . . Bd7; in fact any bishop
move would do since the rook on a8 would then be guarded. So both
10 . . Bb7 and 10 . . . Be6 come into consideration. Now what should
.
I play in this situation? Well it strikes me that he is rather vulnerable

G E N E R AL A D V I C E 1 43
on the e-file with White intending a timely e5, so perhaps 10 . . . Be6

is most sensible, providing his king with some shelter on the file.
Yes, that's what I think he will play, so let us start analysing the
consequences of that move. '
The reader should realize that this internal dialogue helped the
Master identify seven possible candidate moves , but he quickly
chose one as the prime contender and was then going to delve into
the ramifications of this move.
By continuing this type of dialogue with oneself it is possible to
explore the intricacies of a position. Obviously some positions are
easier to analyse and assess than others. The position given in our
example is quite complicated and leads to the tree-diagram with
many branches, shown on page 144.
If one takes into account that most of the analysis just considered
arose from only one branch of our tree, we can see that the position
was very complicated. Simpler examples with long forcing varia
tions are easier to analyse, and Kotov refers to these as 'bare
trunks'-for instance, Diagram 120:
Diagram 1 20
R. E. Spurgeon-Povah
Paignton
1 974
This position arose at move 38 in one of the author's games. The

time-control had just been reached and adjournment was not far
away. White continued 39 b4 and Black played 3 9 . . . Net ! winning
a pawn because of the twin threats of 40 . . . Nxg2 + and 40 . . . Nc2 + .
It i s worth mentioning a useful tip here: i f you are going t o have to
seal a move and you have a forced or relatively obvious move, then it
often pays to seal that move. In this case I quickly played my 39th
move supposing that it would make him think, when much to my
surprise and embarrassment he played 40 f3 , and I was asked to seal
1 44 G EN E RAL ADVICI;
Tree of analysis for position in Diagram 119
1 0 Qxd4 . . .
1 0 . I . Bb7 10 . . ; Bd7 1 0 . . . c6 10 . . . Qd7 1 0 . . . RbB
plausible
10 . . . cs
anti�positional quite Don't aid development
11 b4 11 Nds
, I-
(positional move (attacks in
secures weakness at a6 11 Bf4! (continue centre but
but too slow) development) easily repelled
�.-m.....
by 11 . . . c6)
12
13
14
lS
r----
es dxes
Qxes Bd6
Qxf4 0-0
Qe3 Bxf4
I
12 Nds (threatening
1 3 Nxf6 to smash
12 Radl Be7
and Black
castles
safely
K-side pawns)
is fine for Black
1 2 . . . Nxds? 1 2 . . . Bxds?!
13 exds! 1 3 exds + . . .
wins bishop 12 . . . Be7
+-
T
1 3 . . . Be7 1 3 . . . Kd7
king is
permanent
target
1 3 . . . Kxe7
0-0
1 4 es � 13 . . . Qxe7
1 4 es . . . 14 Qe3 ? Nxds 1 4 Nh4!
+ ts Nfs Rea
16 Re3 BfB
17 Rael :!:
14 . . . Nhs ? 14 . . . Nds
ts . . . exd6 cXd6 15 . . . exd6 cXd6
1 6 . . . Bxd6 1 4 . . . dxes 16 . . . Qxg7
I
+- +-
1 s Rxes 0-0 1s Nxes 0-0 1 s Bxes 0-0

16 Bxf6 Qxf6
lS QxeS! . . . 1 7 Qxf6 gxf6
:!:
0-0
0-0
15 . . . Ra7 ? 15 . . .. cs 15 . . .
1 6 Nd4! 1 6 Nh4 16 Nd4! :!:
+- 17 Nf5 QdB
18 Rad! Qb6
19 Nxh6+
+-
GEN ERAL A DV I C E 1 45
my next move. The problem was that I had the difficult choice
between two equally attractive moves 40 . . . Nxg2 + or 40 . . . Nc2 + .
Had I done the sensible thing I would have sealed 3 9 . . . Net and
then I could have analysed the two alternatives at move 40 during
the adjournment. However, it was not to be and I settled down to
analyse the position , to choose my sealed move, as follows:
40 f3
40 . . . Nc2 + 40 . . . Nxg2 +
I 41 Kf2 exf3
42 Kxf3 f4!
40 Kd2? 40 Kf4
Nxa3 I
41 fxe4 43 Bxf4 43 Bf2 43 Kxg2
fxe4 40 . . . exfJ ?! 40 . . . e3 ! ? Nxh4+ Ne3 fxg3
4 2 Ke3 4 1 gxf3 nxa3 14 kxf5 e2 + 44 Kxf4 44 Kxg3
Nc2 + 42 Kxf5 4� Kg6 el = Q Nc2 Kxds
43 Kxe4 unclear 43 Bxel Nxel unclear 45 Kf4
Nxb4 unclear Ke6
-+ [44 g4!) 46 Kg5
b5!
47 Kg6 47 Kxh5
Ke5 Kf5!
48 Kxg7 zugzwang
Kf5 48 a4
49 Kh6 bxa4
Kg4 49 b5
-+ a3
50 bxa6
•2
51 a7
al = Q
52 a 8 = Q
52 . . . Qxa8?
I
52 . . . Qd l +
STALEMATE MATES
Because of the relative simplicity of the position I was able to

analyse one variation 13 moves in depth. However, I did not have
faith in my analysis and thought that one analytical error in the 40
. . . Nxg2 + variation could be very costly so I sealed the wrong
move--namely 40 . . . Nc2 + . Fortunately the story has a happy
ending , for my opponent played the inferior 40 Kd2 ? and lost quite
quickly; had he played 40 Kf4 and found 44 g4! which I had com
pletely underestimated when sealing, he would have had very good
chances of drawing at least! The moral to this story is to have faith in
your own analysis, because if you don't-nobody else will!
1 46 G E N E RAL ADV I C E
If the keen student wishes t o improve his analytical powers then

he should make a point of studying Master games with good verbal
annotations. Verbal assessments of positions will often give an
observer greater insight into a position than any amount of detailed
analysis. The reader is advised to consult some of the books recom
mended at the end of this chapter.
In order that the student might realize what is expected of him,
here are a few exercises (answers to be found on page 1 68) :
Diagram 1 2 1 Diagram 1 22
Povah-Baljon Boleslavsky-Kotov
Wijk aan Zee I l l Zurich Candidates
1 975 1 953
White played 28 Nxe5 . Analyse Assess this position-Black

the consequences of this move played 13 . . . Na5 . Was this an
and draw the relevant tree error and should he have con
diagram.· sidered 13 . . . Nb4 or 13 . . .
Nd5 ?
G E N E RAL ADVICE 1 47
Diagram 1 23 Diagram 1 24
Fi�cher-Petrosian Teichmann-Bernstein
Candidates (7) St Petersburg
1 971 1 909
Should White play 1 3 Bb5 or What positional advantages

can he hope to profit from his does White enjoy, and how can
advantages in some other fash he exploit them to the full?
ion?
ANNOTATING CHESS GAMES
This is an exercise which is sadly neglected by too many players.

Botvinnik was an avid enthusiast for the annotation of games . He
used to write detailed notes on most of his games and claimed that
this was one of the most effective methods of improving one's play.
The author, while not claiming to do detailed annotations of his
games, does keep a thorough record of all the games he has played,
most of which have light notes. Such a practice has several advan
tages (a) good cross-referencing enables one to improve opening
play; (b) one becomes familiar with certain middle game positions
and how to handle them; (c) it becomes easier to prepare for oppo
nents whom one meets regularly.
As an exercise we give the following game score without com
ment, and the reader is advised to write notes on the game and then
compare them with those given in the annotated version. You
should time yourself on this task, spending no less than half an hour
and ideally about one hour.
N. E. Povah-J. Sanz (Spain) , Charlton, 1979

1 e4 e5 21 Bh4 f4
2 Nf3 Nc6 22 Nd5 f3
3 Bb5 Nge7 23 g3 Rf8
4 Nc3 Ng6 24 b3 Nc6
5 d4 exd4 25 Qd2 Qxe4
6 Nxd4 Bc5 26 Rhe1 Qf5
7 Be3 Bxd4 27 Qc3 Nce5
8 Bxd4 �o 28 Rd2 Rd7
9 Be3 d6 29 Kb2 b6
10 h4 f5 30 Ne7 Qt'7
11 Qd5 + Kh8 31 Nc6 Qg7
12 Bg5 Qe8 32 Nd4 Rg8
13 h5 Nge5 33 Re4 c5
14 h6 gxh6 34 Nf5 Qg6
15 Bxh6 Rt'7 35 Nxd6 RfB
16 ��o Ng4 36 Rxe5 Nxe5
17 Bc4 Nce5 Black lost
18 Qd4 Be6 on time
19 Bxe6 Qxe6
20 Bg5 Rg7
When you have completed your annotations turn over and compare
your notes with those of the author.
N. E. Povah-J. Sanz, Charlton, 1979
1 e4 e5
2 Nf3 Nc6
3 Bb5 Nge7
An unusual third move in the Ruy Lopez. Its main advantage is its
surprise value.
4 Nc3
4 0-0 is also possible, but 4 d4 exd4 5 Nxd4 Nxd4 6 Qxd4 Nc6
justifies Black's strategy.
4 . . . Ng6
4 . . . g6 is a common idea in this variation, but here it is decidedly
dangerous, e.g. 5 d4 Bg7 (5 . . . exd4 6 Nds Nxds 7 exds Qe7 + 8 Kft
Ne5 9 Qxd4 f6 10 Bf4 Bg7 1 1 Rel ± P. Johner Tartakover, Karlovy

Vary 1 9 1 1 . ) 6 B g 5 ! h6 7 Be3 exd4 8 Nxd4 0-0 9 Qd2 intending
0-0-0 with the advantage.
5 d4 exd4
6 Nxd4 Bes
White's play up to here has been very natural, but now comes the
problem of how to continue and what plan to adopt. I wanted to
castle on the queen side so that I could launch my king-side pawns in
an assault on his king (I expected him to castle short) . So play
continued-
� Be3 Bxd4
8 Bxd4 0-0
If 8 . . . Nxd4 9 Qxd4 0-0 then White stands better due to his freer
development.
9 Be3
White retains the two bishops.
9 .d6 . .
10 h4!?
A provocative move: White foresees that Black intends . . . f5 , so he
launches his king-side advance immediately.
Theory mentions the game Barczay-Sydor, Lublin 1 969: 10 0-0
Kha 1 1 Qd2 f5 12 f4 fxe4 1 3 Nxe4 Qe7 14 Ng3 �-
10 . . . f5
Not 10 . . . Nxh4 ? 1 1 Qh5 Nxg2 + 12 Kd2 ! h6 1 3 Bxh6 with a crushing

attack.
11 Qd5 +
This queen check enables White to castle quickly.

11 .Kha. .
12 Bg5
Now the bishop will not be disturbed by . . . f4.
12 .Qe8. .
13 hs Nge5
1 3 . . . Nge7 i s also possible, but 14 Qd2 followed b y 0-0-0 leaves

White well placed.
14 h6! gxh6
Not the tempting 14 . . . g6? keeping the h-file closed , but allowing
15 f4 ! Ng4 16 Bxc6 bxc6 17 Qd4 + + -
15 Bxh6
15 Rxh6 is answered by the annoying 1 5 . . . Ng4 .
15 . . . Rf7
-
Not 1 5 . . . Rg8 ? because of the sequence 16 f4 Nf7 1 7 Bxc6 bxc6 1 8
Qd4 + again.
16 0-�?!
Castling for the sake of it! It is true that early castling is wise, but one
should only castle when there is no better move, and here 16 f3
preventing . . . Ng4 was to be preferred.
16 . . . Ng4!
An annoying move threatening both . . . Nxf2 and . . . Nxh6.

17 Bc4?!
This doesn't help, better was 17 f3 ! Nxh6 (not 17 . . . Nf2 ? 18 Bxc6

bxc6 19 Qd4 + winning the knight) 18 Rxh6 when Black's exposed
king guarantees White a lasting initiative.
17 ... Nce5
18 Qd4 Be6
Black now starts to gain the initiative.

19 Bxe6 Qxe6
20 Bg5 Rg7
21 Bh4 f4!
This secures the e5 square for Black's knight.

22 Nd5
White hopes to play 23 f3 , repelling one of the frisky horses , but

he doesn't want to allow 22 . . . Ne3 . But now Black gets there
first-
2 2 . . . f3 !
2 3 g3
Not 22 gxf3 ? Nxf3 followed by 23 . . . Qh6 + and 24 . . . Nxh4 - +

23 . . . RfB
24 b3?!
Missing the point of Black's last move, which was to guard the f6
square. White should protect his f2 pawn with 24 Rd2 when chances
are about level.
24 . . . Nc6
25 Qd2
White must lose a pawn and the f-pawn is more important.

25 . . . Qxe4
26 Rhet
Fortunately White now gains active play for his pieces to compen
sate for his pawn deficit.
26 . . . Qf5
27 Qc3
White's main chances are based on his control of the dark

squares-particularly the at-h8 diagonal. Black cannot grab the
f-pawn, e.g. 2 7 . . . Nxf2 ? 28 Bf6 ! Rxf6 29 Nxf6 threatening 30 Res +
27 . . . Nce5
28 Rd2 Rd7
Black is running short of time and the position is difficult for him.
Perhaps he should try to force the issue with 28 . . . c6 although this
leaves the d6 pawn vulnerable.
29 Kb2 b6
White's last move was played simply to make Black think and to rush
him into an error-the clock plays its part!
30 Ne7 Qf7
31 Nc6 Qg7
32 Nd4!
Not 32 Nxa7? c6 ! and the knight is lost. Now White's knight has a
lovely choice of squares.
32 . . . Rg8
33 Re4!
Another White piece comes into the attack, and combinational
possibilities begin to arise.
33 . . . c5?
This loses as it weakens the d6-pawn, but it is difficult to find
constructive moves for Black anyway.
--
34 Nf5 Qg6
Better was 34 . . . Qf8 3 5 Nxd6 ! Rxd6 36 Rxg4 Rd4 ! (36 . . . Rxd2 3 7
Qxe 5 + Rg7 38 Rxg7 Qxg7 39 Bf6 + - or 36 . . . Rxg4 3 7 Qxe 5 + Qg7 3 8
Qxg7 + Kxg7 3 9 Rxd6 + - ) 3 7 Rdxd4 ! cxd4 3 8 Rxg8+ Kxg8 3 9 Qxd4
±.
3 5 Nxd6?
Missing the most accurate win by 3 5 Rxg4 ! Qxf5 ( 3 5 . . . Qxg4 36
Bf6 + Rgg7 3 7 Bxg 7 + wins material) 36 Rf4 ! followed by Bf6 +
winning the exchange.
35 . . . Rf8?
Black's best hope was 3 5 . . . Rxd6! 36 Rxe5! Rd4 (not 36 . . . Nxe5 ? 3 7
Qxe 5 + + - or 36 . . . Rxd2 3 7 Rg5 + Qf6 3 8 Rxg8 + Kxg8 3 9 Bxf6 + - )
3 7 Rxd4 cxd4 ( 3 7 . . . Nxe5 38 Rd5 wins the knight) 38 Qxd4 Qg7 (38
. . . Nxe 5 ? 39 Qxe5 + Rg7 40 Bf6 and White reaches a won king and
pawn ending) 39 Re4 Qxd4 40 Rxd4 h6 (40 . . . Nxf2 ? ? 41 Bf6 + Rg7
42 Rd8 mate) 41 Rd2 Re8 ! 42 Kc3 Re2 43 Rxe2 fxe2 44 Kd2 Nxf2 4 5
Kxe2 with an ending which should be won for White, but Black i s not
without hope.
36 Rxe5 Nxe5
and Black lost on time; however he was clearly worse as the follow
ing line shows: 3 7 Qxe 5 + Qg7 38 Qxg7 + Rxg7 (not 38 . . . Kxg7 ? 39
Nf5 + + - ) 39 Ne4 ± White's two minor pieces are far superior to the
rook, and Black's f-pawn will soon fall, e.g. 39 . . . Rgf7 40 Bg5
followed by Bf4 , Ng5 and Nxf3 + - .
It is hoped that the keen student will adopt the habit of annotating
most of his games as thoroughly as possible, for the dividends
should be high.
LEARNING TO HANDLE A CHESS CLOCK
It is very important that an ambitious player learns how to handle the

chess clock in the most professional fashion. Alekhine emphasized
this fact by saying that time trouble is as little justification for a
blunder 'as the claim by a law-breaker that he was drunk at the time
of the offence'. He continued, 'the inability of an experienced Master
to cope with his clock must be regarded as just as serious a fault as an
oversight'.
So how should a player cope with time trouble? Well, first he
should naturally try to avoid it by not squandering so much of his
time--a crime of which many players are guilty. This goal can be
achieved in several steps: (a) adopt a more professional approach, do
not spend so much time wandering around the room when your
opponent is thinking, but use that time to analyse the position and to
make general assessments and formulate plans; (b) improve your
analytical skills so that you don't repeatedly analyse the same varia
tions , thereby wasting valuable time; (c) good opening preparations
will also save time as well as increasing the pressure on the oppo
nent.
If despite these guidelines you still run into time trouble, then you
must train yourself to be able to think quickly but calmly. It is very
important that you look calm when in time trouble for this has an
unsettling effect upon many opponents who expect you to be flus
tered. Botvinnik advises the use of training games with very little
clock time, in which the aim is to make good use of yuur time rather
than worry about the quality of the moves played. Repeated practice
enables the ' clock-sinner' to improve both his attitude and analytical
skills under time pressure.
One practice which should be avoided is moving swiftly to keep
the opponent in time trouble. This simply makes the game a lottery
and the first player to blunder usually loses. Why should you take
such risks if you have more time ? However, one time-trouble prac
tice tliat often does work is to analyse long and complicated
sequences of moves in advance and then to rattle them off quickly.
The advantage of this technique is that your opponent cannot
possibly analyse the position to the same degree of depth as
you, especially as he does not know which first move you are
considering.
1 54 G E N E R A L ADV I C E
Finally there are occasions when time trouble can provide a player
with his salvation. Usually this arises when you are clearly worse but
your opponent is running short of time, so you should seek an
opportunity to complicate the issue , and these attempts sometimes
border on desperation! Consider this example:
Diagram 1 25
P. G. Wann-Povah
London
1 979
Black's pawns on b2 and a3 gave him a winning initiative, but White

was not dead yet and he played 23 Qxc6 ! With both players short of
time this move came as a nasty shock, and it should have been
enough for at least a draw. Play continued-2 3 . . . Bxc6 ?! 24 Bxc6 +
Qd7 25 Rxd7 Ra6 (not 2 5 . . . a2 ?? 26 Ra7 + + - ) 26 Rc7 + ? ? (Black
had accepted the proffered queen on move 2 3 , having expected 26
Bb5 ! and intending the blunder 26 . . . a2 ? because 27 Ra7 + ? Kd8 28
Rd1 + Rd6 ! 29 Rxd6 + Kc8 30 Ba6 + Kb8 3 1 Rb7 + Ka8 wins for Black,
but 27 Rdd1 + ! Ke7 28 Bxa6 axbl = Q 29 Rxb1 Ra8 30 Bd3 Ra1 31 c3·
wins for White! However Black has chances of a draw with 26 . . .
Ra5 ! 27 Rb7 + Rxb5 28 Rxb5 a2 29 R5xb2 axbl = Q + 30 Rxb1 Kd7 with
a technically difficult win for White.) 26 . . . Rxc6 2 7 Rxc6 a2 28
resigns.
In the next example Kotov has a hopeless position against
Reshevsky, so he tries a sensible gamble in his opponent's chronic
time trouble:
Diagram 1 26
Reshevsky-Kotov
Zurich Candidates
1 953
Kotov played 34 . . . Qe2 and Reshevsky glanced anxiously at his
clock and fortunately found 35 Qxf8 + ! (had he played 35 Rft then 3 5
. . . Rdt 36 Nd2 Qxd2 3 7 Bc4 Rxft + 3 8 Bxft Qg5 would allow Black
to stave off White's immediate threats) 35 . . Rxfa 36 Rxe2 and
.
Reshevsky made the time control and won. Although Kotov's

gamble failed, it was undoubtedly his best try and quite justifiable
under the circumstances.
HOW TO IMPROVE
How does the young ambitious player gain the necessary experience
to enable him to aspire to his desired level of success ? Basically there
are two methods of improving one's game--playing and studying.
But it is not adequate to just play hundreds of games-the playing
conditions need to be stringent and the opposition should be of
reasonable quality, so the following factors should be taken into
account and sensible ratios must be determined:
1. Quantity and quality of opposition
2. Tournament Chess
3. Study time
1. Quantity and Quality of Opposition

It is very important that the student strikes up a good balance be
tween time spent playing and studying. The author found that he
1 56 G E N E RA L A D V I C E
was quite capable o f playing about seventy competitive games

(games played with a chess-clock for school, club and county teams)
a year from the age of fourteen through to the early twenties, when he
increased his activity to an average of about 120 games a year. The
serious student should consider this figure of seventy games as a
fairly good target and he should certainly aim to play at least fifty
games a year; there are very few top class players who have managed
to make it to the top with a lower average.
By playing at least thirty match games a season , a player will also
get a published BCF (British Chess Federation) rating which will be
_
revised annually. The following chart gives a rough indication of
various playing levels:
Grade Standard
Up LU H J U Beginner
1 00-1 2 5 Low: Club standard
1 2 5-1 50 Average: Club standard
1 50-1 75 High: Club standard
1 7 5-200 County strength
200-2 2 5 National Expert and Master levels
225+ International Master strength
One of the big problems for the fast-improving youngster is to find

good-quality opposition. It is well known that one generally benefits
most of all from playing people slightly stronger than oneself.
In many instances the junior player is prevented from meeting
such opposition, either because he is played on a very low board in
team events or because he tends to play opponents of his own age
(i.e. school matches and junior events) . These problems can easily be
overcome if the youngster is ambitious enough and chooses his
tournaments carefully-for instance, Tony Miles, Britain's top
player and our first Grandmaster, played in and won the British
Under-2 1 Championship when he was only fifteen! There is nothing
to stop many youngsters, who have experienced a measured degree
of success at a given age group, from playing in older sections.
Similarly, a player should strive to push himself to higher grading
limits. There are many tournaments nowadays which incorporate
graded sections, such as a major tournament limited to players under
about 160, or a minor tournament limited to players under about 1 30.
A player graded at 125, who has played in several minor sections
G E N E RAL A D V I C E 1 57
with reasonable results, should consider moving up into major sec

tions as soon as he feels ready to do so. The principle is to stretch
yourself to the limit, and gradually you will alter your horizons!
2. Tournament Chess
Tournament chess offers many players the most effective method of
gaining competitive experience and, with the British tournament
scene being the envy of Europe, there are plenty of opportunities for
the keen enthusiast to improve. One problem that does arise is that of
finding details about tournaments. Until one gets on the tournament
circuit mailing list, the best solution to the problem is to enquire at
your local chess club or to consult the 'Forthcoming Events' sections
in chess magazines�hess , British Chess Magazine or the BCF's
monthly publication Newsflash .
As mentioned earlier, it is important to choose one's tournaments
carefully so as to get the maximum benefit from the experience.
There are literally dozens of week-end 'Swiss' tournaments through
out Britain now and these gruelling events, which usually last six
rounds (one round Friday night , three rounds Saturday and two
Sunday) , have been largely responsible for the rapid improvement in
British chess throughout the 1970s.
A skeleton list of the BCF Chess Calendar, including most of the
large 'Open Swiss tournaments', is given below:
January Hastings Chess Congress (numerous events)
February City of London Congress
March Blackpool Congress
April Aaronson Masters Tournament
May Surrey Congress; Surrey Junior Championships
June Charlton Congress
July Evening Standard Congress; Manchester Congress
August British Championships plus supporting events;
Lloyds Bank Masters Tournament
September Aaronson Congress; Benedictine International
(Manchester)
October LARA (Lambeth Arts & Recreational Association)
Congress
November
December Islington Congress; London Junior Championships
(All events are in London, unless stated otherwise)
The keen enthusiast would be well advised to try and play in a
1 58 G E N E R AL A D V I C E
selection of these events and, i f h e does, h e may find the following

tips useful:
(a) ' Swiss' tournaments invariably have fast time controls usually
followed by 'blitz finishes' hence it pays to play fairly quickly con
serving time for the latter stages.
(b) Attacking play should be favoured since it is more difficult to
find the most accurate defences with fast time-controls. It is even
worth sacrificing a pawn for nebulous attacking chances-a danger
ous practice at the slower international time-control of 40 moves in
two and a half hours. Avoid short draws!
(c) The large number of entrants in the 'Swiss' events normally
ensures that at least 5 points out of 6 are needed for a winning score;
so do not give up if you lose an early game, for 5 points usually
wins a prize!
3. Study Time
Throughout this book the student has been encouraged to spend as
much time studying as he possibly can. It should be quite possible
for most enthusiasts to put in at least three to four hours of study a
week; some may even try to emulate Fischer who literally worked
like a Trojan. Bisguier said of Fischer, ' what was important was that
he was playing a tremendous amount of chess and learning and
improving all the time. He devoured all the chess books he could get
his hands on and was slowly but surely assimilating this knowledge,
utilizing what he could make use of immediately and storing the rest
for future reference, like a squirrel eating enough nuts to assuage his
hunger and hoarding the remainder for later needs. '
No matter how m !lch time a player spends studying, the important
thing is that his time is well spent. As already mentioned, most
players concentrate on developing an effective opening repertoire,
but the best method of improving play must be the study of complete
games. By playing through games with good notes the student is able
to see how the Master's ideas develop and how they are carried
through to their conclusion. In this way the student not only learns
his openings, but he relates them to certain middle game themes, and
on some occasions he will learn how to handle particular endings.
We conclude this chapter with three games played by junior
players who certainly have made the grade.
The first, played when Fischer was only thirteen, has been heralded
by many as 'the game of the century'!
D. Byrne-R. J . Fischer, Rosenwald Tournament, 1956
1 Nf3 Nf6
2 c4 g6
3 Nc3 Bg7
4 d4 �
5 Bf4 d5
The Grunfeld defence-Fischer's great favourite.
6 Qb3 dxc4
7 Qxc4 c6
8 e4 Nbd7
Bisguier-Benko, US Championship 1963/4 , went 8 . . . b5 9 Qd3 Qa5

10 Be2 b4 11 Ndl c5 12 0-0 Ba6 with good prospects for Black.
9 Rdl
Fischer refutes 9 e5 as follows-9 . . . Nd5 ! 10 Nxd5 cxd5 1 1 Qxd5

Nxe5! - + .
9 . . . Nb6
10 Qc5
Provocative! 10 Qb3 was more prudent.
10 . . . Bg4
11 Bg5?
This error, moving a piece for the second time before completing his
development, allows Fischer to start some fireworks. 11 Be2 fol
lowed by 0--0 was natural and best.
1 1 . . . Na4!
12 Qa3
If 12 Nxa4 then Black regains his material with interest, e.g. 12 . . .

Nxe4 13 Qcl Qa5 + 14 Nc3 Bxf3 +: or 1 3 Bxe7 Nxc5 14 Bxd8 Nxa4 - +
or 13 Qxe7 Qxe7 14 Bxe7 Rfe8 1 5 Be2 Rxe7 1 6 0--0 b5 winning a piece
or 13 Qb4 Nxg 5 ! 14 Nxg5 Bxd1 1 5 Kxd1 Bxd4 16 Qd2 Bxf2 + .
12 . . . Nxc3
13 bxc3 Nxe4
Sacrificing the exchange but opening the e-file.

1 60 G E N E RAL A D V I C E
14 Bxe7 Qb6!
15 Bc4
Not 1 5 Bxf8 Bxfa 16 Qb3 Nxc3 ! 1 7 Qxb6 axb6 1 8 Rat Rea + with a
winning attack-Fischer.
15 . . . Nxc3
16 Bc5
16 Qxc3 Rfe8 would allow Black to regain his piece with an extra
pawn.
16 Rfe8 +
1 7 Kh
Diagram 1 27
D. Byrne-A. J. Fischer
Rosenwald Tournament
1 956
Here Byrne must have expected 1 7 . . . Nb5 when 18 Bxf7 + would
give him the advantage, but Fischer comes up with a real gem !
17 . . . Be6 ! !
' . . . a stunning masterpiece o f combination play . . . ' Hans Kmoch.
18 Bxb6
The alternatives are no better:
(a ) 18 Bxe6 Qb5 + 19 Kg1 Ne2 + 20 Kf1 Ng3 + 21 Kg1 Qf1 + 22 Rxf1
Ne2 mate.
(b) 18 Qxc3 Qxc5! +
(c) 18 Re1 Nb5 +
18 . . . Bxc4 +
19 K g 1 Ne2 +
20 Kf1 Nxd4 +
21 Kg1
G E N ERAL ADVICE 1 61
ff 2 1 Rd3 axb6 22 Qc3 Nxf3 ! - +

21 . . . Ne2 +
22 Kft Ne3 +
23 Kg1 axb6
24 Qb4
If 24 Qd6 Nxd t 2 5 Qxdt Rxa2 intending . . . Rat .
2 4 . . . Ra4!
25 Qxb6 Nxdt
Black's material advantage is decisive and the rest is a massacre 26
h3 Rxa2 27 Kh2 Nxf2 28 Rel Rxe1 29 Qd8 + Bf8 30 Nxet Bd5 31 Nf3
Ne4 32 Qb8 b5 33 h4 h5 34 Ne5 Kg7 35 Kgt Be5 + 36 Kft Ng3+ 37
Ket Bb4 + 38 Kd1 Bb3 + 39 Ket Ne2 + 40 Kbt Ne3 + 41 Ket Re2 mate.
G. Kasparov-L. Polugayevsky, 46th USSR Championship, 1978

This sensational victory by the fifteen-year-old Gary Kasparov in the
latest Soviet championship, over one of the world's top Grandmas
ters, heralds the arrival of another star of the future. Kasparov is
currently considered to be the best player for his age in the world and
it is not unreasonable to talk of him as a future World Champion.
1 e4 e5
2 Nf3 e6
Polugayevsky is a leading exponent of the Sicilian Defence, but his

name is more commonly associated with the Najdorf variation 1 e4
c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 a6, etc.
3 d4 exd4
4 Nxd4 a6
5 Ne3
This rather natural move has been overshadowed in the last few
years by either 5 Bd3 or 5 c4.
5 . . . Qe7
6 Be2 b5
Black wishes to mobilize his queen side rapidly, but he is behind in
development and Kasparov is quick to take advantage of this fact.
7 Bf3 Bb7
8 0-0 Ne6
1 62 G E N E RA L A D V I C E
Boleslavsky-Getman, USSR 1965, went 8 . . . d 6 9 a4 b 4 10 Na2 Nf6

1 1 Rel e5 1 2 Nf5 g6 1 3 Bg5 Nbd7 14 Nh6 Qb6 1 5 c3 ± .
9 Nxc6 dxc6
9 . . . Bxc6 would allow White to play 10 Rel with threats of Nd5 .
10 e5!?
This was a new move which Kasparov had probably prepared at
home; the pawn sacrifice is quite logical since White must open up
the position to utilize his lead in development.
10 • • • Qxe5
Polugayevsky can hardly decline the offer, otherwise the pawn on e5
would gradually cripple his game by controlling two useful squares
f6 and d6.
1 1 Rel Qc7
12 Bh5
A very strange-looking move with a threat of Rxe6 . Presumably
Kasparov got the idea from some earlier games where the idea was
tried in another setting, e.g. Stein-Gipslis, USSR 1 965.
12 . . . Be7
Polugayevsky probably assumed that Kasparov could not capture on
e6 because of the reply 1 3 . . . g6, but
13 Rxe6! g6
The alternative is 13 . . . Nf6 when Kasparov could have played 14
Ne4 with great complications thus:
(a ) 14 . . . Nxe4 1 5 Rxe4 when White's lead in development gives
him the advantage.
(b) 14 . . . 0--0 ? is obviously dangerous, e.g. 1 5 Nxf6 + Bxf6 16 Rxf6
gxf6 1 7 Bh6 ± since White regains the exchange with the initiative.
(c) 14 . . . Nxh5 15 Qxh5 and now:
(i) 15 . . . 0--0--0 16 Bf4 Qxf4 17 Rxe7 ±
(ii) 1 5 . . . g6 ? 16 Nf6 + Kd8 1 7 Qdl + Kc8 1 8 Bf4 + -
(iii) 1 5 . . . 0--0 ! 1 6 Qe5 Qxe5 1 7 Rxe5 with equality.
This is the variation Polugayevsky should have chosen, but he prob
ably missed Kasparov's next move when choosing to play as he did.
14 Re1 !
The obvious try 14 Qd4 is inadequate, e.g. 14 . . . fxe6 1 5 Bf4 ( 1 5

Qxh8 ? 0-0-0! threatening both 1 6 . . . gxh5 and 1 6 . . . Bf6) 1 5 . . .
e5 ! 16 Bxe5 Qd8 ! 1 7 Qe4 gxh5 18 Bxh8 Qd7 + .
14 . . . Rd8?!
Polugayevsky declines the sacrifice presumably because of the self
assured manner of the young Kasparov. We cannot be sure of how
much analysis Kasparov had done at home, but the ability to play
one's moves quickly and confidently is of great importance in such
situations. A victory for home analysis ?
Diagram 1 28
G. Kasparov-L. Poluga
yevsky 46th USSR
Championship 1 978
lf Polugayevsky had accepted the proffered bishop 1 4 . . . gxh5 then

play might have continued-1 5 Qd4 ! f6 1 6 Bf4 when White's lead in
development and Black's unco-ordinated position offer Kasparov
good attacking chances, e.g. 16 . . . Rd8 17 Qe3 Qc8 18 Qc5 ! main
taining pressure against e7, blockading Black's c-pawn and eyeing
up the h5 pawn. The reader is advised to examine this position for
himself to see if he can find adequate attacking chances for White
against attempted defences such as 18 . . . Kf8 , 1 8 . . . Kf7 or 18 . . .
Rd7 .
15 Qf3 c 5
The bishop i s even more poisonous now 1 5 . . . gxh5 1 6 Bf4 Qd 7 1 7
Qxh5 with a very powerful attack due to threats such as Be5 , Rad 1 ,
Ne4, etc.
16 Bf4 Qb6
The exchange of q'Ueens does not help Black-16 . . . Bxf3 1 7 Bxc7

1 64 G E N E R A L ADV I C E
Bxhs 1 a Bxda Kxda 1 9 f3 g 5 20 Rad 1 + Kea ( 2 0 . . . Kea 2 1 Nd5 Bd6 2 2

Rea + when Black will never b e able t o untie himself) 2 1 N d 5 h 6 22
Rd3 Rh7 (22 . Kfa 23 Nxe 7 + - ) 2 3 Rxe7 + ! Nxe7 24 Nf6 + Kfa 25
. .
Nxh5! when Black's pieces are in no position to defend his queen

side pawns.
17 Qg3 gxh5
18 Bc7 Qg6
19 Bxd8 Qxg3
20 hxg3 Kxd8
21 Rad1+ Kc7
22 Nd5+ Bxd5
23 Rxd5
Diagram 1 29
Kasparov-Polugayevsky
An ending has been reached where Black has a bishop and knight for
rook and pawn. This material balance would normally favour Black
(partk:ularly so in a middle game) . but in this case Black's pieces are
poorly co-ordinated, as bishop moves would allow the annoying pin
Rea and the knight must defend the bishop , so development is
painful!
23 . . . h6
24 Rxh5 Rh7
25 Rhe5 Kd7
26 R5e3 Rg7
27 Rd3 + Kc7
28 Ra3 Rg6
29 Rf3 Bf6
Bellin in his notes to the game cites 2 9 . . . Bf6 ? as a decisive blunder,
G E N ERAL ADV I C E 1 65
saying that 29 . . . Rg7 would make it difficult for White to better his
position.
30 c3 Kd7
31 Rd3 + Kc7
32 Rea Ne7
This loses the f-pawn, but it is difficult to find constructive moves for
Black, e.g. 32 . . . Be7 allows 33 Ra8 when Black hardly has a decent
move.
33 Reda Nc6
34 Rad7 + Kb6
35 Rxf'7
Now that the f-pawn has fallen Kasparov is well on the way to
victory-3 5 . Be7 36 Re3 Bd6 3 7 f4 c4 3a Kh2 Bc5 39 Re2 b4 40 Re4
. .
bxc3 41 bxc3 Bf2 42 Rxc4 Bxg3 + 43 Kh3 Bet 44 a4 Na5 45 Rb4 + Kc5?
46 Rf5 + resigns.
Black loses his knight, but even after 45 . . . Kc6 he would be
struggling to restrain White's f-pawn. Although this game is not as
spectacular as Fischer's classic victory over Byrne, it is still a
remarkable achievement for the young Kasparov to play with such
maturity and to overwhelm such a formidable opponent as
Polugayevsky.
Finally, a game played by one of Britain's top juniors Julian Hodg
son. This game caused quite a sensation when it was played, for it
was a victory which helped Julian (who was only fifteen at the time)
to score an amazing 2! points from three consecutive games against
Grandmasters. The game is annotated by Julian especially for this
book.
J. M. Hodgson-J. D. M. Nunn, Aaronson Open, 1 978
1 e4 c5
2 f4
'After the game I found that my opponent had never played against
this line before. '
[Britain's fourth Grandmaster John Nunn, i s a relatively new
advocate of the Sicilian, hence it is not too surprising that he should
be caught out by this unusual variation. (-Author.)]
2 Nc6. . .
3 Nf3 g6
'3 . . . e 6 is also a possibility, but it i s not s o sharp.'
4 Nc3 B g 7
'If 4 . . . e6 then 5 d4 gives White the advantage. '
5 Bc4 d6
'Here my opponent chooses a move which I think is not as good as 5
. . . e6, for if Black manages to play . . . d5 then he has at least
equalized. '
6 0-0 Nf6
7 d3 0-0
8 f5?!
' White should play 8 Qel and Qh4 before f5 . '
8 . . . gxf5
' Black must accept the pawn sacrifice, otherwise his queen-side
pieces will be blocked in. '
9 Qel fice4
10 dxe4 Bg4
11 Qh4 Bxf3?
'Here-as was later pointed out by Jonathan Kinlay-Black should
play 1 1 . . Bh5 and . . . Bg6 with a very solid defensive position.'
.
12 Rxf3 Ne5
13 Rh3!
Diagram 1 30
J. M . Hodgson-J . D. M .
Nunn Aaronson Open
1 978
G E N ERAL A D V I C E 1 67
13 . . . Ng6
'After 13 . . . Nxc4 14 Nd5 White has a winning attack. ' [If Black tries
to escape with 14 . . . Re8 then 1 5 Bh6 ! appears decisive.
(-Author.)]
14 Qg3 Qd7
15 Nd5!
'To prevent 15 . . . Qg4. '
1 5 . . . Nxd5
'If 1 5 . . . Nxe4 then 16 Qf3 Nf6 1 7 Nxf6 + Bxf6 18 Qh5 with a crushing
attack. '
16 Bxd5 e6
17 Bb3 d5
18 Qf3 !
' White threatens both 19 exd5 and 19 Qh5 . '

18 . . . c4
19 Ba4! Qxa4?
'Black's only chance was 19 . . . Qe7 20 Qh5 Bd4 + 21 Khl when he

has either . . . f6 or . . . f5 . If 21 . . . f6 22 c3 Be5 23 exd5 exd5 24 Bc2
gives White a strong attack. However 21 . . . f5 is not so clear;
possibly 22 Bg5 Bf6 23 exf5 exf5 24 Bh6 Bg7 25 c3 with a difficult
position for both sides . '
20 Qh5 Rfd8
21 Qxh7 + Kf8
22 Bh6 Bxh6
23 Rxh6!
' White will soon have queen and two rooks against Black's king with
the black major pieces out of play. '
23 . . . Ke8
'Already White was threatening 24 Rxg6 followed by 25 Rft + and
26 Qf7 mate. '
24 Rfl Rd7
25 Qg8 + !
1 68 G EN E R A L ADV I C E
'Not 2 5 Rxg6? fxg6 2 6 Qha + Ke7 2 7 Qxaa Qxc2 when it is Black who
has the advantage. '
25 . . . NfB
26 Rxe6 + ! Kd8
[Not 26 . . . Re7 ? 27 Qxf7 + winning. (-Author.)]

2 7 QxfB + Kc7
28 Qc5 + Kd8
29 Rh6 resigns.
'Black cannot prevent Mate . '
Solution t o Exercises o n page 146

Diagram 1 2 1
28 Nxes
f'-0f
28 axb3 28 . . . a3 ! 28 . . . Bxes
_,, �
7? 29'�dg)(al + 29 Rd7 ! !
• 30 i<d2Rd8
31 QhB+ ! BxhB
32 RxhB+ kg7
29 bxa3! bxa3
30 Nd7 + !
29 Nd7+ Ke7
30 Bxc7? axb2+ 2 9 . . . Qbe 29 . . . Bx ;;
z "l , , . . �',b2t
":_J,
31 Kdt bt =Q+ 30 Qh7 30 Qhe mate
32 Ke2 Ba6+ +-
33 Kf3 Rxc7
-+
I
30 • . . Qxd7 30 . . . Kee 30 . . . Ra6? 30 . . . Ke7 30 . . . Qe6
31 Rxd7 . . . 31 Bxc7 a2 31 Bf6 + - 3 1 Bf6+ . . . 31 Bf6 Kee
I
32 Qh8 + ! BxhB 32 Rd3!
�
7 +-
3:3:�: L
Ct. QhBJ
I
31 . . . Kee 31 . . . Kd6 31 . . . Ke6
32 Rxb7 32 Rd3+ Kc6 32 Qs4+ Kd6
+-
=
33 Rxh6+ Kg7 33 Rxd7 Kxd7 33 Rd3+
Ct. QhB + )
I
� Bes mate ·
34 +- +-
34 Bes? at=Q+
35 Bxat Rxat + 1--0 (in 4 3 moves)
36 Kb2 RxhB
37 Kxal Rd&
-+ 3 4 Bd6 + !
I I
1
34 . . . Ke6 34 . . . Kxd7
35 Nxcs+ RxcS 35 Bes + Ke&
36 Rxa8 BxaB 36 Rxc8
37 Kb2 o!:
I I
36 . • . al = Q + ? 36 . . . Bxc6 36 . . . Rxc6
37 Bxat Rxat+ 37 Bal Bb7 37 Bat Bxe4
38 Kb2 36 f3 + - "'
+- t. c4 , Rdt,
Kc2, Bc3
Rat, Kb2, etc.
G E N ERAL ADVICE 1 69
Diagram 1 2 2
I n this typical Queen's Gambit Accepted position White has his
normal pressure down the e-file and particularly against the e6
square. Bronstein points out in his classic book on the tournament
(see Recommended Reading, page 1 70) that White has , in effect, a
lead of three "tempi-both rooks are well placed and his white
squared bishop is on a good diagonal. Kotov played 13 . . . Na5
which was branded an error by most commentators because of Boles
lavsky's forceful play-14 d5! Nxb3 (obviously capturing on d5
loses a piece) 15 dxe6 Qb6 ( 1 5 . . . Bxf3 1 6 exf7 + Kha 17 Rxda Bxe2 ta
Rxaa Rxaa 19 Rxe2 allows White to regain his piece with two extra
pawns) 16 axb3 fxe6 1 7 Nd4 ! + - Bd6 1 8 Qxe6 + Kha 19 Nf3 Rada 20
Bf4! Bxf3 (20 . . . Rfea 2 1 Rxd6 Rxe6 2 2 Rxb6 Rxb6 23 Be7 leaves
White better placed) 2 1 Rxd6 Rxd6 22 Qxd6 Qxd6 23 Bxd6 Rea 24
Rxea + Nxea 25 Bes with an easily won ending 1-0 (42).
After the game Stahlberg and Euwe recommended 1 3 . . . Nb4 , but
some years later Rauzer demonstrated that 14 ds Nbxds 15 Nxds
Bxd5 16 Bxd5 Nxd5 1 7 Bxe7 Qxe7 ta Rxd5 wins for White. This
prompted Najdorf to suggest 13 . . . Nd5 14 Nxds Bxg5 1 5 Nc3 Nb4 as
being all right for Black. Bronstein claims_ that the consequences of
1 3 . . . Nd5 are not so clear after 14 Nxds Bxg5 1 5 Nb6 ! Qxb6 (if 1 5 . . .
Rb8 1 6 Nxg5 Qxg5? 1 7 Nd7 + - or 1 5 . . . Ra7 1 6 d5! gives a very
strong attack) 1 6 Nxg5 with unpleasant threats against Black's unde
fended king side, e.g. 16 . . . Nxd4 (not 16 . . . h6 ? 17 Nxf7 ! Rxf7 t a
Qxe6 with a strong attack) 1 7 Qd3 Nf5 t a Nxe6 fxe6 1 9 Bxe6 + Kha 20
Bxf5±so it seems that Black's position was in a bad state anyway, but
the lesser evil appears to be 13 . . . Nd5.
Diagram 1 2 3
Fischer spurned 1 3 Bb5 ! ? axb5 14 Qxaa 0-0 presumably fearing that
Black's two bishops and passed d-pawn would give him ample
compensation for the exchange. Instead he gives a masterful illustra
tion of how to capitalize on the enemy pawn weaknesses at a6 and
dS-1 3 Rel ! Qxa4 t4 Nxa4 Be6 1 5 Be3 0-0 ( 1 5 . . . Nd7 fighting over
the c5 square is answered by 16 f4! g6 1 7 Bd4 with a grip on the dark
squares) 16 Bc5 ! Fischer offers the exchange of dark-squared bishops
so as to make it easier to attack the weak pawns on white squares for,
as Bronstein says , ' . . . a dark square weakness meant that the pieces
on the light squares were also weak. The key to the attack on the dark
squares was to occupy them with my forces , by which I attacked my
opponent's men on the light squares. ' 1 6 . . . Rfea 1 7 Bxe7 Rxe7 1 a

b4 ! (fixing the pawn at a6 , because 1 a . . . a 5 i s met by 1 9 b 5 with a
powerful passed pawn) 1 a . . . Kfa 1 9 Nc5 Bea 20 f3 Rea7 21 Re5 Bd7
22 Nxd7 + ! Typical Fischer! He exchanges his ' good' knight for the
'bad' bishop in order to gain entry squares on the c-file. 22 . . . Rxd7
23 Rel Rd6 (to stop 24 Rc6) 24 Rc7 Nd7 25 Re2 Black hardly has a
decent move, so we give the rest of the game without comment 2 5 . . .
g6 26 Kf2 h5 2 7 f4 h4 2 a Kf3 f5 2 9 Ke3 d4 + 30 Kd2 Nb6 3 1 Ree7 Nd5
32 Rf7 + Kea 33 Rb7 Nxb4 34 Bc4 resigns.
Diagram 124
White has a dominant knight on d5 and temporary control of the
b-file; while Black's dark squares in the vicinity of his king are very
weak, so Teichmann attempts to exchange dark-squared bishops-1
Bh4 ! Bd4 + (1 . . . Bc6 2 Nf6 + Bxf6 3 Bxf6 would leave Black defence
less on the al-ha diagonal) 2 Kh1 Kg7: this king move defends the f6
square, but shortens the black bishop 's line of retreat. 3 Bf2 Bxf2 (3
. . . Be5 ? 4 f4 doesn't help) 4 Rxf2 Qa5 (trying to prevent 5 Qc3 + ) 5
Qe2 f6 6 Qb2 ! Now White controls both the al-ha diagonal and the
b-file. 6 . . . Rfa 7 g4 h6 a h4 g5 9 f4 ! White is hell-bent on attacking
the f6-square. 9 . . . gxh4 10 Nxf6 ! and White soon won, since 10 . . .
Rxf6 1 1 g5 is hopeless for Black.
RECOMMENDED READING
The author recommends the following books as being particularly

useful for study, although the study of most chess books will prove
instructive:
Kotov, A. , Th ink Like a Gra n d m aster (Batsford, 1 9 7 1 )
Kotov, A. , Play Like a G ran d master (Batsford, 1 97a)
Chernev, I. , The most in structive ga m es of Chess ever played ( Faber ,
1 965)
Bronstein, D. , The Chess S t ruggle in Practice (McKay , 1 97a).
Chess U n l i m ited
Chess like music, like love, has the power to make men happy
-Tarrasch.
In this, the last chapter of the book we shall take a brief look at the
lighter side of chess.
There are many facets to chess which ensure that the enthusiasts
should never tire of this great game. For instance, chess can be
played at differing speeds, with Blitz Chess at one extreme and
Correspondence Chess at the other.
BLITZ CHESS
Blitz Chess is a popular pastime with nearly all chess players includ
ing the great Masters. In fact most top-class players pride themselves
on the standard of their blitz play.
There are two different forms of blitz chess. One is the very popu
lar 'five-minute' game, in which each player's clock is set at five
minutes to the hour and the game must be completed before one of
the flags falls , otherwise that player will have been deemed to have
lost on time. Naturally it is possible to play ten-minute, fifteen
minute or whatever time control the players see fit. The fact that an
unofficial World Blitz Championship was held in 1 970 immediately
after the USSR v Rest of the World match shows that the top Grand
masters take blitz chess seriously. Twelve of the world 's top Grand
masters competed in an all-play-all five-minute tournament and the
victor was the mercurial Bobby Fischer with a phenomenal 19 out of
2 2 . Runner-up was the ex-World Champion Mikhail Tal with 14!
points , (4! points behind) while further down the field were such
illustrious names as Korchnoi, Petrosian, Bronstein, Hort, Smyslov
and Reshevsky, to name but a few.
1 72 C H ESS U N L I M ITED
The other form of blitz chess i s 'buzzer' o r 'lightning' chess , in

which a buzzer sounds every ten seconds (or whatever rate is prefer
red) and the player whose turn it is to move must move on the buzzer.
This form of blitz chess is not as popular as five-minute chess and it
is mainly played in England where there is , in fact, an annual British
Lightning Championship.
The author strongly advises the ambitious student to make full use
of 'five-minute' chess as a means of sharpening his quick play and as
a general training aid for the study of various positions. Further
more, blitz chess is great fun!
CORRESPONDENCE CHESS
Correspondence or postal chess is another very effective way of

improving your game. Moves are posted back and forth between the
players, who obviously have plenty of time to analyse the intricacies
of an opening, middle game or even endgame position. Postal chess
is very good for developing a methodical approach to your analysis
of positions. One can write out a detailed analysis of certain posi
tions at great length and this not only improves your understanding
of the position, but also helps you to undertake your analysis in a
logical and thorough fashion. Postal chess is also a good method of
learning a new opening because one has plenty of time to learn the
basic ideas behind that opening. Because correspondence chess
takes quite a time (games usually take about a year and international
games may take several years) , most top correspondence players
have ten to twenty games going at once; hence there are plenty of
opportunities to learn and experiment.
The keen student is again advised to try postal chess to see if it
suits him.
CHESS PROBLEMS AND STUDIES
Throughout a game of chess the two players are confronted with a

series of problems, some specific and some general. General prob
lems fall into the category of basic principles such as which pieces to
exchange off and where to post one's pieces, etc. Specific problems
on the other hand are of the nature: does Bxf3 result in the win of
C H E S S U N L I M IT E D 1 73
material, or do I have a mating attack here? The solutions tend to be

more concrete and we can often see a definite answer to the question.
Consider the problem in Diagram 1 3 1 : can White draw?
• • •
• • • •
•t• •
-� ·"
.. . � • B
M R
r�
• • •
. v.• • •
• t • .\ti>. Diagram 1 31
. . � H. K. Mattison
The reader might like to try and solve this one, arrd the solution will
be given at the end of this section.
The world of chess problems and studies is quite vast, with solu
tions requiring mate, help-mate or even stalemate. The field of chess
problem composition and solving is so complex that we will not
attempt to do it justice here. The author recommends that the study
of such problems should be conducted for enjoyment; their practical
value is only evident for positions which have a strong resemblance
to a serious ' game-like' position.
Solution to position 1 3 1 1 a4 + Kb6 (naturally not 1 . . . Kxa4 2
Rxc5 Kb3 3 Bg5; 1 . . . Kc4 2 Bf2 intending 3 Rxc5; 1 . . . Kb4 2 Bel +
Kc4 3 Bd2 when White draws at least, in all cases) 2 Bf2 cl Q 3 Rxc5
=
Qxc5 (Any queen move is met by a rook discovered check winning

the queen) 4 Kh1 ! ! (Black is in zugzwang and must capture the
bishop on f2) 4 . . . Qxfz stalemate!
CHESS LITERATURE
There are literally thousands of chess publications and it is well

known that more books have been published on chess than all the
other games in the world put together. With such a plethora of chess
books to choose from , the student finds himself in the difficult
position of having to be selective. Chess is an international game and
1 74 CH ESS U N L I M ITED
knowledge of algebraic notation will enable the reader t o read games

in various languages, although he may not be able to read the notes.
The English-speaking chess world has the advantage since most
books are published in English.
To help the student decide which books to consult the author
gives a short bibliography of recommended reading. The reader is
advised to buy or consult the following books, which are classified
according to the three stages of chess:
Opening
A. Matanovic (ed.), Encyclopaedia of Chess Openings (Batsford/
FIDE) vols A-E. (These five volumes are very expensive and are
best used as reference works.)
In addition, the Batsford series of chess books about the openings
are particularly useful.
Middle Game
A. Nimzovich My System (Bell, 1 92 9) .
Bobby Fischer My 60 Memorable Garnes (Faber, 1 969) .
D r M. Euwe, H . Kramer The Middle Game Book 1-Static Features
(Bell, 1964), Book 2-Dynarnic and Subjective Features (Bell, 1965).
There are many other books worth reading on the middle game.
These, plus those mentioned on page 1 70, were of great interest to
the author and it is hoped that the reader may also derive some
benefit from them. It should be emphasized that any collection of
games from one of the all-time greats, such as Alekhine, Capab
lanca, Lasker, Tarrasch, Fischer, Spassky, Tal, Karpov, Korchnoi,
etc. will be a useful training aid.
Endgame
P. Keres Practical Chess Endings (Batsford, 1974) .
Grigory Levenfish and Vassily Smyslov Rook Endings (Batsford ,
1971).
Reuben Fine Basic Chess Endings (Bell, 1941).
CHESS MAGAZINES
Apart from the thousands of books on chess, there are many

magazines in many different languages. Some of the more interest
ing ones are:
CH ESS U N L I M ITED 1 75
1 . Chess Informant vols 1-26. A FIDE publication. This magazine

appears every six months and contains over 600 complete games
with notes. It is an international publication with no language--very
good value for the serious player.
2. British Chess Magazine. Available from 9 Market Street, St
Leonards on Sea, East Sussex TN3 8 ODQ.
3. Chess. Available from Chess, Sutton Coldfield, Warwickshire.
Both these magazines are monthly publications and are good value
for most chess players in Britain.
4. Rabbits. Review. Available from The Editor, 7 Billockby Close ,
Chessington, Surrey.
This is a relatively new magazine, which caters for the beginner; it is
very informative and good value.
VARIATIONS ON A mEME
Although chess ranks as the most complex game in the world (a

conservative estimate has shown that 25 x 10115 different chess
games are possible! ) , there are still those who like to experiment
with the many fascinating offshoots of chess. We shall briefly
examine a few of these hybrids in case the reader wants some light
entertainment.
1. ' Must-take' Chess

In this crazy game the object is simply to lose your pieces as quickly
as possible! Captures are imperative, although if there is more than
one, then naturally you may choose. The first player to lose all his
pieces, including his king , is the winner.
2. Progressive Chess
This is a very entertaining game in which White plays one move,
Black two consecutive moves; White then has three consecutive
moves, and so on. The object is to checkmate as normal, but any
check in a sequence of moves concludes that sequence, e.g. if you
have five moves and check on your third move, then you forfeit the
other two moves.
3. Cylindrical Chess
In this game the pieces move in exactly the same fashion, but the
1 76 C H E S S U N L I M IT E D
board i s considered t o b e a hollow cylinder: that i s the a and h files

are adjacent, but the first and eighth ranks are not in contact, other
wise the rooks would be able to capture each other. A miniature
could go as follows-1 f4 ? ! g6 2 h4 Bg7 3h5?? c5 + ! (check along the
d8-a5-h4-e1 diagonal! ) 4 Rh4 Qxh4 + 5 g3 Qxg3 mate.
4. Kriegspiel
Kriegspiel is one of the more popular chess variants. Three chess sets
are required-each of the two players has a set to himself, and a
referee controls a master board. The two players play normal chess
except neither has knowledge of his opponent's moves. The referee
maintains the actual position on his master board which is out of
view of both players. The players move in tum when the referee tells
them that their opponent has moved. If a capture is made, then the
piece is removed from the board of the player who lost the piece, and
the other player is informed that he has made a capture--but is not
told what piece he has captured. If a check is given, both players are
informed and the referee specifies whether the check is along a rank,
file or diagonal, or by a knight. A player can make various inferences
from the moves that he is allowed to play as well as those that the
referee will not allow (the illegal moves) . To save time, a popular
question is, 'Are there any pawn captures?' If there are, the referee
replies, 'Try' and the player must try at least one pawn capture.
Otherwise the referee will reply 'No'. Pawn captures by en passant
are announced by the referee one� they have been played.
It is customary for the opponents to sit back to back with the master
set between them, and the stage is set for the battle of luck and logic
to begin.
Although there are numerous other hybrids of chess, such as
hexagonal chess, pocket-knight chess, exchange chess, to name but a
few, we shall leave it to the reader to discover the wonders of this
incredible game for himself.
The author hopes that this brief chapter has shown the keen
devotee, that although chess can be taken very seriously (and should
be if you are ambitious! ) it also has a lighter side, which can bring
hours of enjoyment. This should come as no surprise; after all chess
is a game--and what a game ! !
Also in Faber Paperbacks
LEO N A R D BA R D E N ' S C H ES S P UZZLE B O O K

An Evening Standard C hess Book
Leonard Barden
J U N I O R C H ES S P UZZLES
Raymond Borr a n d Stanley Morrison
L O G I CA L C H ES S
M ove b y M ove
Irving Chernev
TH E M O ST I N ST R U CTIVE GAM ES OF C H E S S EVE R P LAY E D

6 2 M asterpieces o f M odern C hess Strategy
Irving Chernev
T H E C H ES S C O M PA N I O N
Edited b y Irving Chernev
WI N N I N G C H ES S
H ow t o Perfect Yo u r Attacking Play
Irving Chernev a n d Fred Reinfeld
F I R ST B O O K O F C H ES S
I. A . Horo witz a n d Fred Reinfeld
I M P R OV I N G YO U R C H ESS
The N i ne Bad M oves a n d H ow to Avoid Them
Fred Reinfeld
Please write to
Faber & Faber Limited, 3 Queen Square, London WC T N 3AU
for a complete list of Faber Paperbacks

Povah, Nigel - Chess Training-Faber and Faber (1981)

Uploaded by

Document Informationclick to expand document informationEntrenamiento en ajedrez

Document Informationclick to expand document information

Copyright:

Available Formats

Povah, Nigel - Chess Training-Faber and Faber (1981)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Povah, Nigel - Chess Training-Faber and Faber (1981)

Uploaded by

Copyright:

Available Formats

<I; CHESS TRAINING

t/f Nigel Povah

FABER AND FABER

©Nigel Povah 1981

This book is sold subject to the condition that it

shall not, by way of trade or otherwise, be lent,

British Library Cataloguing in Publication Data

ISBN 0-571-11 604-3

I should like to express my deepest gratitude to Doris Hall for the

Preliminary General Advice 16

2. Openings in Theory and Practice 35

5. Tactical Complexities of the Middle Game 89

Many hardened fans of English descriptive notation may regret the

The references are determined by labelling the eight files (col­

The pieces are designated by capital letters a s follows:

The following moves, written first in extended algebraic and second

As can be seen extended algebraic notation is much more precise

White Black White Black

Note the marked decrease in the number of dashes, etc.

It is important that the reader makes sure that he or she is familiar

Preliminary General Advice

Although the penultimate chapter of the book also gives general

The diagrammed position was reached after White's 1 1th move in

1. King and Queen versus King

2. King and Two Rooks versus King

Consider the following position-White: K e t Rat Rht , Black:

4 . King and Two Bishops versus King

6 . King and Two Knights versus King

KING AND PAWN ENDINGS

Pawn on the Sixth Rank

would continue-1 Ke2 Ke7 (gaining the opposition since White

SOME GENERAL THEMES

1. King and Rook's Pawn versus King

2. The 'Outside' Passed Pawn

For instance, in Diagram 1 4 White has an outside passed pawn on

White has an extra pawn, but at the moment no passed pawns;

3. The 'Protected' Passed Pawn

bxc4 + (if 3 . . . b4 then 4 c5 leaves Black the impossible task of

4. Effective versus Ineffective Pawn Majorities

tinue-18 f4 ! exf4 ( 18 . . . Ke7 19 fxe5 fxe5 when White has a poten­

Chess is a game which can conveniently be divided into three stages:

I t should be noted that these three concepts ar e frequently related,

Notice the economy of White's development; in four moves he has

weakening the queen side, and to no good effect. Remember, 'PAWN

and this.would be anti-positioned, for it would create a weak square

In the diagrammed position White enjoys a substantial lead in

The next example illustrates the dangers o f neglecting one's

A logical capture doubling White's pawns before Nge2 (connecting

Immediately exploiting the white square weakens with the intention

White continues to neglect his development and, instead of moving

Black threatened 10 . . . exf3 1 1 Qxf5 fxg2 winning. Now the threat is

Notice the activity of Black's pieces compared with White's, which

Black intends to fianchetto his light-squared bishop, but he is neg­

White has regained his pawn and controls the centre.

3. An open or half-open file for a rook (usually the e- or f-file) is an

Early castling is invariably advisable in the opening in order to

queen-side expansion with . . . Rb8 and . . . b5, etc.

The references are determined by labelling the eight files (col

tinue-18 f4 ! exf4 ( 18 . . . Ke7 19 fxe5 fxe5 when White has a poten

Black intends to fianchetto his light-squared bishop, but he is neg