First published in 2016 by Gloucester Publishers Limited, Northburgh House,

10 Northburgh Street, London EC1V 0AT

Copyright © 2016 Cyrus Lakdawala

The right of Cyrus Lakdawala to be identified as the author of this work has been
asserted in accordance with the Copyrights, Designs and Patents Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, electronic,
electrostatic, magnetic tape, photocopying, recording or otherwise,
without prior permission of the publisher.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.

ISBN Kindle: 978-1-78194-324-3

ISBN epub: 978-1-78194-325-0

Distributed in North America by National Book Network,

15200 NBN Way, Blue Ridge Summit, PA 17214. Ph: 717.794.3800.

Distributed in Europe by Central Books Ltd.,

99 Wallis Road, London E9 5LN. Ph 44(0)845 458 9911.

All other sales enquiries should be directed to Everyman Chess,

Northburgh House, 10 Northburgh Street, London EC1V 0AT
email: info@everymanchess.com; website: www.everymanchess.com

Everyman is the registered trade mark of Random House Inc. and is used in this work under
licence from Random House Inc.

Everyman Chess Series

Chief advisor: Byron Jacobs
Commissioning editor: John Emms
Assistant editor: Richard Palliser

Typeset and edited by First Rank Publishing, Brighton.

Cover design by Horatio Monteverde.
Printed by TJ International Limited, Padstow, Cornwall.
About the Author
Cyrus Lakdawala is an International Master, a former National Open and American Open
Champion, and a six-time State Champion. He has been teaching chess for over 30 years,
and coaches some of the top junior players in the U.S.

Also by the Author:

Play the London System
A Ferocious Opening Repertoire
The Slav: Move by Move
1 ... d6: Move by Move
The Caro-Kann: Move by Move
The Four Knights: Move by Move
Capablanca: Move by Move
The Modern Defence: Move by Move
Kramnik: Move by Move
The Colle: Move by Move
The Scandinavian: Move by Move
Botvinnik: Move by Move
The Nimzo-Larsen Attack: Move by Move
Korchnoi: Move by Move
The Alekhine Defence: Move by Move
The Trompowsky Attack: Move by Move
Carlsen: Move by Move
The Classical French: Move by Move
Larsen: Move by Move
1 ... b6: Move by Move
Bird’s Opening: Move by Move
The Petroff: Move by Move
Fischer: Move by Move
Anti-Sicilians: Move by Move
Contents
About the Author
Bibliography
Introduction

1 The Early Years

2 World Champion and 1960-1970
3 The Later Years

Index of Complete Games

Bibliography
Attack with Mikhail Tal, Mikhail Tal & Iakov Damsky (Cadogan 1994)
Botvinnik: Move by Move, Cyrus Lakdawala (Everyman Chess 2013)
Mikhail Tal’s Best Games Vol. 1, Tibor Karolyi (Quality Chess 2014)
My Great Predecessors Vol. 2, Garry Kasparov (Everyman Chess 2003)
Selected Chess Games of Mikhail Tal, J.Hajtun (Dover 1961)
Study Chess with Tal, Mikhail Tal & Alexander Koblencs (Batsford 1978)
Tal-Botvinnik 1960, Mikhail Tal (Russell Enterprises 2001)
The Life and Games of Mikhail Tal, Mikhail Tal (Everyman Chess 1997)
The Magic Tactics of Mikhail Tal, Karsten Müller & Raymund Stolze (New in Chess 2012)

Use was also made of certain annotated games from both ChessBase’s Mega Database and
ChessPublishing.com, involving annotations by Alexander Baburin, Lubomir Ftacnik, Rainer
Knaak, Tony Kosten, Ian Rogers, and Alexander Volzhin.
Introduction
Things are not what they appear to be; nor are they otherwise. – Surangama Sutra

The nature of miracles is they contradict our understanding of what we consider ‘truth’.
Perhaps the miracle itself is a truth which our minds are too limited to comprehend. Mikhail
Nekhemevich Tal was just such a miracle worker of the chess board. His games make us
believe that anything, however unlikely, is possible, if we are able to back it up with
imagination, daring and computer-like calculation/intuition. He revelled in confusing
opponents, who were otherwise regarded as competent.

One of the most – if not the most – beloved players of all time was born on November
9th, 1936. Tal was a born prodigy, and not just in chess. He was reading by age three, and
could perform three figure multiplications in his head by age five. His memory was close to
photographic, as well. By age 15 he had already skipped three grade levels and begun
University. In 1955, by age 19, he made it to the USSR Championship and his high-risk
fantasy play caught the eye of several GMs, as well as the public and the press who
immediately made him their darling. In 1957 Tal did the impossible by winning the USSR
Championship at age 20. There followed dominating tournament performances, including the
Interzonal. He normally won by completely squashing the lower half of the field.
Tal’s reputation itself became a weapon, so much so that his opponents began to play
like children spooked by an adult’s ghost story. In this way, Tal essentially began each game
as if he had already won. When it came to the young Tal, it appeared as if he were hell-bent
on self-destruction, yet didn’t succeed, since he kept winning! To many of his opponents, it
felt as if he balanced out his incredible luck with a lack of common sense. From the late
1950’s on, with each year that passed, Tal’s rivals felt less and less equal against his rising
skills. Each game against Tal was an ordeal to be survived, rather than a chess game to be
played.

The World Championship Match with Botvinnik

What has chess come to? – Mikhail Botvinnik after losing his 1960 world championship
match against Tal.

To Botvinnik’s and Smyslov’s deep annoyance, the Tal-loving chess public considered his
play a supreme act of transcendence, of which no other mortal was capable. For those who
looked upon Tal’s impertinent style with scepticism, it is human nature to distrust that which
we don’t understand. In 1960, Tal was undisputedly the best tournament player in the world,
but matches were a different thing altogether. How would Tal fare against the ageing iron
logician?
 In a duel between a strategist and a tactician, the former tends to have a reach and speed
advantage in the ring, while the latter relies on a single colossal knockout punch. In a feud
between an honest person (Botvinnik) and a criminal (Tal, of course!), the criminal holds an
advantage, since he is unconstrained, outside the law.
In this match, to everyone’s surprise Tal dared to challenge Botvinnik strategically –
although the trademark tactics were clearly woven within all the games. Tal obviously
altered his style and played Botvinnik with an uncharacteristic degree of respect. Now you
may argue that Tal playing strategically is a little like Mozart signing the devil’s contract to
be the greatest composer of his day. The trouble is Mozart could have had his wish without
owing the devil a thing. I’m convinced that Tal was the dominant player in the world in
1960, and could have beaten Botvinnik – who was then past his prime – playing in his
normal, con-artist tactical style (which he indeed did in some of the games of the match).
Even playing in this careful style, Tal still managed to turn a position from stable to
chaotic, almost without any sense of transition. That which we would consider platitudinous
advice today, like: “In the opening phase, develop quickly and strive to control the centre
before attacking,” was akin to divine revelation in Morphy’s era. The most impenetrable
barrier in the universe is the person who examines the data with a mind already made up,
even though circumstantial evidence suggests the exact opposite.
By the Botvinnik/Smyslov era, the top players had somehow settled into a false belief that
chess had to be played only one way: correctly, from a strategic base. Tal, in a sense, the
new Morphy, radically altered the Botvinnik/Smyslov ‘correct play’ paradigm, by
successfully violating this deeply held belief, and in doing so, rekindled our sense of
wonder of the game. He single-handedly turned the Botvinnik/Smyslov age of correctness
and reason on its head, by playing and winning in a style which people couldn’t even
imagine.
The chess world suddenly realized that our game is not exclusively some technical
exercise, but instead, a place of imagination and bluff. Organization is just as important as
content. A library may have a book within its walls which holds the deepest secrets of the
universe, yet if we are unable to locate it, it’s as if the book never existed in the first place.
In the same way, Botvinnik knew perfectly well that many of Tal’s moves were semi-sound
or outright unsound. The problem he was unable to solve was: where was the refutation
embedded within the maze of calculation?
Tal, like no other player, preyed on our primal fear of the darkness. He complicated until
Botvinnik’s vaunted logic felt useless, as he groped about to orient himself in the dream-like
sense of dislocation which threatened to envelop and flood his brain, assailing it with a
hundred competing and contradictory impulses. Game after game, Botvinnik experienced
rising discomfort, as the familiar and comfortable slipped away irretrievably, only to be
replaced with the terrible unknown, which in turn induced confusion, fatigue and time
pressure errors. Botvinnik soon realized that a game against Tal was in essence an attack
upon his psyche.

The Disastrous Rematch

Profit and loss, defamation and fame, praise and blame, suffering and joy; all of these
are impermanent; and thus why should any of them cause satisfaction or dissatisfaction?
– The Dhammapada

We are reminded of the cliché: Easy come, easy go. Tal, who endured a total of 12 surgeries
in his life, lost his return match, almost without a fight, due to his declining health of a
kidney. In Tal’s weakened condition, Botvinnik was able to enforce maddening order upon
virtually every game, where Tal’s powers of logic were almost childlike when compared
with Botvinnik’s stark perfection in that realm. In his weakened state, Tal was incapable of
pushing the positions to the brink of madness – and beyond – as he pulled off in the first
match.
Tal’s reign as world champion was the shortest ever. To make matters worse for Tal,
Botvinnik played bolder in the rematch, refusing to back down from calculation battles. And
with Tal’s poor health, this critical power was removed from the equation. Pretty much
every one of Tal’s slumps may be attributed to his poor general health. He didn’t exactly
help matters by his heavy smoking, and he drank like a man who owned a spare liver.
In the 1965 Candidates’ matches, Tal eliminated both Portisch and Larsen, but then was
defeated by the surging Spassky, who went on to defeat Petrosian for the world
championship.

Resurgence in Later Years

I’m always making a comeback, but nobody ever tells me where I’ve been. – Billie
Holiday

Just as it looked like Tal’s failing health would force him into a dignified early retirement,
around 1972 Tal went on a tear, shattering Petrosian’s non-losing streak (68 games without a
loss), by going an unbelievable 83 games without loss. Add to that the risks Tal took in his
games, and it reaches the level of unbelievable. Then in the early 1980’s he did it again, this
time going 80 games without a loss. In 1988, Tal won the second World Blitz Championship
at Saint John, New Brunswick, ahead of the reigning world champion Kasparov and also ex-
champion Karpov. I can’t explain it. Can you?
Tal’s dazzling powers finally began to fade by the early 1990’s, in lock step with his
declining health. Normally, age brings with it increased caution, but with Tal, not so much!
As he grew older, his aggression continued unabated, with only the pretext changing. Time
itself lacked the power to alter his core style, which was an unrepentant optimism. The
natural optimist like Tal always envisions what he can gain, rather than worry about the
potentiality for loss. It’s as if he only looks forward and is incapable of seeing what is
behind him.

Tal’s Style
Please allow me to introduce myself, I’m a man of wealth and taste.

Our own unique styles are rare melodies, never to be exactly replicated, when we pass from
the world. There are many different forms of currency in chess. To Tal, initiative, attack and
the creation of combinational potential were the three most important. His motto in
rationalizing his shady sacrifices (many of which were maddeningly vague abstractions,
more than concrete ideas): “Better to be poor and among the living, than to be rich and
dead.” He just naturally picked moves which left almost zero possibility of an amicable
settlement with the opponent.
Tal famously said that he attempted to clear a path in the “dense forest ... where there is
only room for one.” He also tried to create positions where “two times two is five”. Chess
journalist John Henderson wrote: “Just like Shabalov’s legendary mentor, the great Mikhail
Tal, there’s a ‘fear factor’ element when you played him, and he won many spectacular
games with unsound sacrifices because his opponents saw ‘phantom attacks’ that proved
afterwards to be unsound.”
Tal was either unwilling or unable to recognize shared realities, which he revoked and
replaced with his own: i.e. “It’s not a bad thing to be down material.” Tal disagreed and
many of his opponents must have doubted his hold on reality. Sometimes Tal’s sacrificial
indulgences would go horribly awry and he would lose like an amateur. Yet he believed in
his own magic. So much so, that in the very next game he would be back to his old tricks,
completely unchastened and unrepentant, so that his opponents found it difficult to call him
to account. His erratic thoughts and speculations over the board were utterly unguessable by
the rest of us, whose minds functioned on the basis of the rational.
Tal realized early on, that his mark always falsely believed that he or she is the actual
con-artist, who plans to come away with something for nothing. He would lure them into the
darkness with sacrifice (I am hard pressed to find Tal games in this book where he didn’t
sacrifice), and once the relentless campaign of mischief began, it was all but impossible for
the opponent to restore order. Imagination tends to fall flat if combined with
incomprehension of a position’s essence. If Capablanca’s play epitomized intuitive elegant
simplicity, then Tal was the evil twin who strove for anarchy.
Yet, like Capa, Tal achieved his aims just as intuitively, where his tactical intuition
would allow him to successfully navigate utterly incomprehensible positions. But don’t get
me wrong. There were few players who were capable of out-mathing Tal’s famed
intuition/calculation abilities. He was essentially the modern day computer, versus most of
his opponents, who represented the 1970’s high school nerd with a slide rule. Yet Tal still
relied on his intuition, which doesn’t require calculation in every instance. Instead, images
were formed in his subconscious, with the solution instantaneously delivered to his
conscious mind. I sense that is how Tal kept his bearings in the most bewildering of
positions.
When I play over Capablanca’s or Fischer’s games, it is easy to fall sway to the
delusion: ‘I too can play like this!’. When I play over Tal’s dream-like games, the thought to
play like him is impossible, since his self-ordering chaos feels utterly unreachable, for me,
the ordinary. Tal was the only player who was capable of producing miracles and magic on
the chess board. Complex calculations drifted through his mind with unconscious ease of
motion, while we ordinary mortals couldn’t untangle some his positions if given the luxury
of a week of study.
Tal’s great contribution to our game is that he showed us that chess is not just an
accountant’s ledger, with scored tallies of debt and income. He used sacrifices as routinely
as we use money to purchase what we want. In Tal’s case, what he ‘bought’ were three
items:

1. Energy for his initiative.

2. Attacking chances.
3. Enhanced tactical opportunity.

The Patron God of Sacrifice: Tal’s Gambles

“It ain’t right, Atticus,” said Jem. “No, son, it ain’t right.” – To Kill a Mockingbird

There are no atheists among gamblers, who all pray to their lucky star. Tal never believed in
the philosophy that a small portion of a certainty is sometimes better than going for 100%,
with an all-or-nothing gamble. Tal, in direct contradiction of Abe Lincoln, sincerely
believed that he really could fool all of the people, all of the time. Tal was the greatest
entertainer within the theatre of the mind. Why? In game after game he was willing to risk it
all, just to give the audience a thrill. I guess bravery, foolhardiness and innocence have a lot
in common.
Tal was a born gambler, on par with Katherine Parr, the last (and luckiest) of Henry
VIII’s wives, who managed to pull off the impossible, by retaining her head during her
tenure by cleverly outliving Henry. Most GMs believe: when a variation is inherently
incorrect, precedence and numbers don’t matter. Even though a billion people believe a lie,
the force of their numbers fails to magically turn the lie into the truth. Now most great
players are seekers of truth on the chess board. Shockingly, Tal didn’t care about abstract
notions of eternal truth. He just wanted to win, and he was willing to play deliberately
shady moves and sacrifices to achieve his aims: to confuse and fluster the opponent.
Have you ever argued with someone, realized that their reasoning is flawed, but were
unable to refute their illogic? Well, now you know just how Smyslov, Botvinnik and many,
many others felt when they lost to Tal. After intense study of Tal’s games, my own games felt
passionless and depressingly devoid of ups and downs. A sacrifice or a combination’s
stature is revealed not by its length, but by the depth of originality. In this book there are
probably more sacrifices and more combinations than in all my other player-books
combined.
Time after time, Tal rushed into Death’s arms and survived the dance. Just look at some
of the positions he reached:
 The attack always whispered seductive invitations to Tal, who could never resist. In this
position Tal was White against Tolush from a Poisoned Pawn Najdorf (to enter this line as
Black against Tal is the same as walking into a seedy looking bar in the rough part of town,
despite the fact that there is no bouncer). With his g5-bishop hanging, Tal played Re1!!, a
move backed up by today’s comps, yet incomprehensibly difficult to find over the board. He
found the hidden refutation’s pattern, so randomly and so deeply embedded within the
geometry, that it is next to impossible for a human (other than Tal!) to decode it.

Restlessness is the great tempter, which whispers into our ear: ‘You can have more, on
condition you are willing to commit yourself to risky action.’ In this case Tal, leading the
Candidates’ tournament with Smyslov as Black chasing him, earlier rejected a favourable
unloseable ending, and entered this mess. On his last move he played Qh4, allowing ... bxc4.
Tal managed to sufficiently confuse a former world champion and swindle his way to
victory. When we are busted, we tend to discard reason to embrace madness. Tal is the only
player I can think of who embraced madness when his position was perfectly good! His
move decisions in this game are the conscienceless ‘Id’ acting upon every crazy impulse,
without fear of consequences.

To tell Tal to refrain from a semi-sound sacrifice, was the equivalent of reasoning with a
two-year-old as you pass by the mall’s toy store. He was merely a helpless passenger to his
creative desires, which always took control of his decision-making process – even with the
world championship on the line. The threat of debtors’ prison never pressed on Tal’s
conscience, and this is clearly a case of will co-opting reason. He led Botvinnik by a point
in his 1960 match.
In this game Tal sensed that Botvinnik had imposed his will upon the position, and that
Tal would slowly be ground down strategically, unless he could alter the landscape
radically. So on his last move he unleashed ... Nf4!?!? – a move which nearly caused a riot
among the spectators, and which reminds us of a celebratory ape swinging through the
jungle’s trees. After making such moves, he would wait for a receptive part of his
unconscious mind to attune to the new circumstances. For Tal to have the audacity to commit
to such an obviously unsound sacrifice – and against Botvinnik, with the world title on the
line – is akin to the newly ordained priest declaring to his congregation: “I think the
scriptures are wrong about quite a few things. Let me explain why.”
Tal always believed his fairy godmother would rescue him, just before disaster struck.
In this case the move took its toll on the flustered Botvinnik’s clock, while Tal’s forces
exploded in exultant rage, allowing chaos to lead them where it would. He intuitively
understood that an act of rash folly can also be a weapon, since its very audacity disorients
a rational opponent. And who was more rational than Botvinnik?
 Tal, playing Larsen as White in their Candidates’ match, put it all on the line with Nd5,
an objectively dubious sacrifice, which of course, worked!
After going through some of the games in this book, you may get an odd feeling that
nearly every one of Tal’s victories was a freak statistical anomaly. Here is a game from the
very young Tal:

Game 1
V.Ragozin-M.Tal
Riga (simul) 1948
Slav Defence

1 d4 d5 2 c4 e6 3 Nf3 c6 4 Nc3 dxc4

The Noteboom Variation. I’m almost certain the young Tal was winging it here, rather
than playing a line he actually studied.
5 a4 Bb4 6 g3!?
 This move turns the game into a kind of shady Catalan Gambit, living a lifestyle past
what White can afford.

Question: Why a shady version?

Answer: White tossed in a too-early a4, a move normally anathema to such positions.
This gives Black access to b4 and renders White’s compensation problematic at best. Today,
most people play 6 e3 b5 7 Bd2 a5 8 axb5 Bxc3 9 Bxc3 cxb5 10 b3 (10 d5 is also played
here) 10 ... Bb7 11 bxc4 b4 12 Bb2 Nf6, reaching a Noteboom tabiya position.
6 ... c5!?
Undoubling his c-pawns, at the cost of enhancing White’s light-squared bishop down the
h1-a8 diagonal. 6 ... Nf6 is Black’s most common move here.
7 Bg2 Nc6 8 Be3
Protecting d4, while pressuring c5.
8 ... Nf6 9 0-0!?
Allowing a seemingly trivial matter to fester unresolved today, invites tragedy
tomorrow. I would go for 9 dxc5 Nd5 10 Bd2 Bxc5. Even though down a pawn, White looks
better off than in the game’s continuation.
9 ... Ng4!
Black already stands better since he picks up the bishop-pair in an open position, while
weakening White on the dark squares.
10 Ne4
10 Ne5 Nxe3 11 Bxc6+ bxc6 12 fxe3 0-0 looks promising for Black, who may make use
of his bishops in the open position.
10 ... 0-0
I would have removed White’s options by playing 10 ... Nxe3.
 11 Rc1!?
Frowned upon by Komodo, but I doubt White gets full compensation for a pawn after 11
Bg5 f6 12 Bd2 Bxd2 13 Nfxd2 cxd4 14 Nxc4.
11 ... cxd4 12 Bxd4 f5!

Even as a kid, Tal effortlessly seizes the initiative from an experienced GM.

Question: How can a GM play an opening so lamely?

Answer: This game was played in 1948, a database-less, intuitive era, where even some
of the strongest players winged it in the opening phase. To look down upon another
generation’s mistakes (with the benefit of the knowledge of our own), strikes me as similar
to one of those Sci-fi movies or novels, where the aliens invade earth, and then soundly
trounce us humans – not all that surprising, given that the aliens’ technological advances pre-
date ours by about 10 million years, or so. The reason we understand ideas and concepts
today is due to the mistakes of the masters of the past.
13 Neg5
This looks awful, but if 13 Nc3?? e5 14 Be3 Nxe3 15 Qxd8 Rxd8 16 fxe3 Bc5 White is
hopelessly lost in the ending.
13 ... h6 14 Nh3 e5!
White’s h3-knight is completely cut off from the centre.
15 Bc3 Bd6?
Kid Tal unwisely dodges the exchange of queens. Surplus ambition, coupled with a
gambling addict’s recklessness, were the two attributable causes of most of Tal’s future
problems on the chess board. In this instance he hopes to whip up a kingside attack. The
trouble is at this stage, Tal’s attack – if it can be called that – remains in the larval stage,
where as yet, it is neither one thing or another. He is winning after 15 ... Be6! with an extra
pawn and a dominating position.
16 Qd5+! Kh7 17 Qxc4 Qe7 18 e4?!
An overreaction which weakens both f3 and f2. White would have been better off
playing 18 Rfd1.
18 ... f4
This is exactly the kind of position Tal’s future opponents would fear. Of course,
Ragozin was oblivious to such feelings, since in 1948, all he saw across from the board was
a kid.
19 Rfd1 fxg3 20 hxg3 Ne3?

When a mad scientist builds his creature, he always does so with big, humanitarian
plans. Then the creature’s power and free will grow beyond its creator’s control. Tal was
the mad scientist, who loosed his creatures upon the world – and sometimes upon himself,
as in this instance.
I reveal to you a dark secret about positional players: we all loathe positions of
unaccustomed calculation, which make our atrophied brains strain with exertion. Tal
understood this, even as a child, and did all he could to force his opponents to calculate. He
was one of the few great players who craved change, just for change’s sake. The reason Tal
is so beloved (his style is the standout, at a time of overlapping signatures) is that he is our
spectral reflection of how we all want to play: without rules or inhibitions. He was the
drunken guy beating up the sober one in a bar brawl. His last move is spectacular yet
incorrect (which was a good way to describe Tal’s play for his entire life!). Black has a
close-to winning position after the simple 20 ... a5!.
21 Nhg5+?
Ragozin should simply accept with 21 fxe3! Bxh3 22 Bxh3 Rxf3 23 Bf5+! g6 24 Kg2
Rxe3 25 Rxd6! gxf5 (25 ... Qxd6?? 26 Qf7+ Kh8 27 Bxg6 Qe7 28 Bxe5+! overloads black
queen and knight to force mate) 26 Qd5 when Black is the one in deep trouble.
 21 ... hxg5 22 fxe3 Bc5
Black has a clear advantage if he plays 22 ... g4! 23 Nh4 g6 with ... Be6 to follow.
23 Bd2 Bb6
23 ... b6! was more accurate.
24 b4 Bg4
Tal should have taken time out for the defensive move 24 ... a6.
25 Rf1 Rad8
When we set our trap, we are the frozen-still heron on the lake, patiently waiting for the
unlucky fish who swims across its path.
26 Rc2?
26 Qe2!, covering both d2 and e3, was necessary.
26 ... Bxf3! 27 Bxf3

Exercise (combination alert): Black has access to

a winning continuation. What should he play?

Answer: 27 ... Nd4!

We get an early glimpse into the mind of the player who made us rethink our belief in an
ordered chess universe, run by ironclad rules. He was an unrivalled master of warping a
sedentary position to a chaotic one, to meet his wicked ends. Tal could read a position’s
hidden tactical clues, the way a child reads a fairy tale. Even stronger though was:
Answer no.2: 27 ... Rxd2! 28 Rxd2 Bxe3+ 29 Kg2 (29 Rdf2?? is met with 29 ... Rxf3)
29 ... Bxd2 30 Rh1+ Kg6 31 Bh5+ Kf6 32 Rf1+ Bf4. The king is unexpectedly safe and
Black emerges up several pawns when White captures the f4-bishop.
28 exd4
28 Kg2 puts up greater resistance, although after 28 ... Nxf3! (most certainly not 28 ...
Nxc2?? 29 Rh1+ Kg6 30 Bh5+ Kh7 31 Bf7 mate) 29 Rxf3 Rxf3 30 Kxf3 g4+! 31 Kg2 (31
Kxg4?? Rf8 gives Black a winning attack) 31 ... Qd6 32 Qe2 Qg6 33 Rc4 Qh6 34 Be1 Qh3+
35 Kg1 Rd1! White won’t hold the game.
28 ... Rxd4 29 Kg2!?
Well, I wouldn’t describe this move as a denim-tough defence. His move is made with
the philosophy: the most shocking way to meet a surprise is with another surprise. When
faced with multiple urgencies, all we can do is prioritize them and turn our attention to the
most solvable one. Ragozin has little choice but to live off food stamps and welfare checks.
He may have resigned if he realized he was playing a future world champion. 29 Qa2
Rxd2+ 30 Kh1 Rxc2 31 Qxc2 g4 is also hopeless for White.
29 ... Rxc4 30 Rxc4 g4!
Tal, setting a trap, obviously doesn’t believe in the criminal’s principle: after the
successful perpetration of a crime, lay low.
31 Be2
31 Bxg4?? Rxf1 32 Kxf1 Qf7+ picks up the rook.
31 ... Rxf1 32 Kxf1 Qf6+ 33 Ke1 Qf2+ 34 Kd1 Qg1+ 35 Kc2 Qxg3
The passed g-pawns are decisive.
36 a5 Bd4 37 Rc7 Qg2 38 Kd3 Qh3+ 39 Kc4 g3!
The best liars cunningly add a grain of truth to their lies to make them completely
believable. It looks like Tal just wants to push his passed pawn, yet the move also contains a
hidden agenda.
40 Rxb7??

Exercise (combination alert): A tempting yet flawed move is a seeming truth,

rooted in a lie. White just blundered in an already lost position. How did kid Tal
end the game in a single move, made in a tone which forbids all further argument?
 40 ... Qc8+ 0-1
Answer: Double attack. Even stronger, however, was to ignore the free rook:
Answer no.2: Mating net. 40 ... Qe6+! 41 Kd3 Qb3+ 42 Bc3 Qxc3 mate.

Why Study Tal’s Games?

Tal is the only player I can remember who did not calculate lengthy variations: he simply
saw through them! – Garry Kasparov

One of my misguided students who only studies opening theory (I keep his name anonymous
since his statement is so profoundly wrongheaded!) once mused aloud that the “mindless”
study of great past players is a waste of time, since current knowledge evolved past the
players’ old knowledge. Now I know for a fact from 35 years of teaching experience, that
students who fail to study past great players struggle with gaps in their understanding of the
game.
The players we study shape our own chess natures. They teach us not what to think, but
the infinitely more important skill of how to think. The oddest thing is after playing through
just a few of Tal’s games, his sacrificial and strategic outrages quickly become – dare I say
it? – normal. When I was a kid, I would try and ape Tal’s style, which was so utterly at odds
with my own, with disastrous results. I was unable to play at optimal strength since I was
pretending to be something I wasn’t.
The trick is to play over a great player’s games to inspire – not necessarily imitate. If you
play through the games with this mind set, you will extract maximum benefit from this book
and other books like it. Tal’s most memorable games play like a novel, where one sub-plot
leads to another and another, until they fuse into the main story. To him, chess was a
boundless mystery which took you from one adventure to another. It was also a place where
imagination rules, where 3+1 doesn’t equal 4, and where water runs uphill. So remarkable
were some of Tal’s ideas, that they feel supported more by enchantment, rather than fact. If
we can extract this sense from his games, your time will have been well spent.

Acknowledgements
Many thanks to editor GM John Emms, and to IM Richard Palliser for the final edit, and also
thanks to trusty CEO of proofreading, Nancy.

May Tal’s imagination spark our own to reach for the impossible.

Cyrus Lakdawala,
San Diego, California,
January 2016
Chapter One
The Early Years
Tal’s early games are notable for their absence of the following traits:
1. Subtlety.
2. Caution.
3. A blatant disregard for long-term weaknesses, or material deficit.
4. An unapologetic obsession with delivery of checkmate against the opponent’s king.

The perfect storm of the above factors makes this chapter the most entertaining of the book.
In the young Tal, we chess fans get the impossible: a world-class coffee house player!

The following diagram gives us a glimpse into the young Tal’s thinking process. In the
first game of the chapter, Tal’s opponent has just played 14 Rfd1.

Tal, like no other player, had a propensity for conjuring drama, from even the most arid
positions. To us, such a position feels like dead terrain. Yet Tal sensed the presence of
sentient life, hidden deep within its recesses. In response, Tal played the inverted:
14 ... Rf8!?
A move bathed in restless energy. Why on earth would he deliberately un-develop? The
answer is that Tal hated all swaps, since it reduced his ability to complicate. To him, such
quiet positions were akin to captivity, and he would take on appalling risk to alter the
position’s nature. The inherent cruelty of chess is that a player with the deeper understanding
of the game can still lose to a kid, who knows next-to-nothing strategically, yet calculates
like a demon and sees tactics infallibly. And Tal was that annoying kid, who continued to
beat his more knowledgeable elders, who could only watch helplessly, as their strategic
superiority dissolved in a haze of complications.
When we guess, is it better to choose yes, or no? A positional player cautiously veers to
the latter. Tal, of course, put his money on the former, always willing to sacrifice if it ‘felt’
right to him. Sometimes we just can’t imagine that his sacrifices are sound. Then after we
play through the game, we realize we need to imagine harder, since Tal pulled it off and
won!
Here is what I’m talking about. In the diagram below, Bobotsov has just played his
knight to d5.

Instead of retreating his queen, Tal played ... Nxd5!, and when White took the hanging
queen on a5, Tal took the bishop with ... Nxe3, with only two pieces for the queen. Now you
may argue that both Caligula and Commodus would criticize Tal’s decision for its violent
excess. But can we legitimately criticize a player when his pattern – sound or unsound –
worked over and over again, even against the strongest opposition of the day?

Game 2
I.Bleks-M.Tal
Latvian Championship, Riga 1952
Pirc Defence

1 e4 d6 2 d4 Nf6 3 Nc3 g6 4 g3 Bg7 5 Bg2 Nbd7 6 Nge2 e5 7 0-0 0-0 8 dxe5

This move hands Black easy equality. White’s best shot at an edge lies in 8 h3.
8 ... dxe5 9 h3
One move late. White wants to post his bishop on e3, without fear of future ... Ng4
ideas.
 9 ... c6 10 Be3
White can also prevent Black’s potential for queenside expansion with 10 a4 a5 11 Be3
Qe7 12 Qd2 Re8 13 Rad1 Qb4 14 b3 Bf8 15 Bh6 Bxh6 16 Qxh6 Qf8 with dry equality,
B.Esen-M.Erdogdu, Ankara 2007.
10 ... Qe7

11 Nc1
Nothing is happening on the kingside so White hopes to build a bind on the other side
with Nb3 and a4.

Question: Isn’t 11 f4, intending a kingside assault, White’s most natural plan?

Answer: It’s been tried in two games, and it looks premature: 11 ... exf4! (this natural
move is so far unplayed by Black) 12 gxf4 Nxe4! (Black eventually regains the lost piece,
with advantage, in all lines) 13 Nxe4 f5 14 Qd6 Re8 15 Qxe7 Rxe7 16 c3 fxe4 17 Ng3 Nb6
when White experiences difficulty in regaining his lost pawn, and even if he does, stands
worse due to an inferior structure.
11 ... Rd8 12 Qe2 Nc5 13 Nb3 Ne6
Tal keeps an eye out for occupation of d4.
14 Rfd1 Rf8!?
‘Swap’ is a dirty word in the natural attacker’s lexicon.
 Question: Why on earth did Tal back away his rook like this?

Answer: Deviation from the norm tends to either delight or offend. The reason for
Black’s bizarre move-choice was that the teenage Tal (and even the adult Tal!) hated swaps,
since each one reduced complications. In this case he makes a large concession, without
much in return. Surprisingly though, even with the concession, I couldn’t find a way to even
an edge for White. Still, Black should probably play the patient 14 ... b6 15 Rxd8+ Qxd8 16
a4 Qc7 17 a5 Rb8 18 axb6 axb6 with equality.
15 Qc4
I would play 15 a4 which should be met with 15 ... a5.
15 ... Ne8!?
Tal is intent on a kingside attack with a future ... f5. The sensible path lies in 15 ... Bd7
16 a4 b6 17 Rd2 Rfc8, but, of course, Tal never played sensibly.
16 Rd2
Once again, I would toss in 16 a4.
16 ... Kh8
In preparation for ... f5.
17 Re1
Discouraging ... f5 by placing his rook on the same file as Black’s queen.
17 ... a5!?
To Tal, his position was the chess equivalent of the drug Ambien. So to remain awake,
he twists the position around to suit his complicating mood. His move is in clear violation
of the principle: Don’t play on your opponent’s strong wing. Tal invites White’s next move.
18 Na4!
Black is strategically lost if White’s knight reaches b6.
 18 ... b5!

This move is forced. For Tal, peaceful negotiation is a remote concept. He begins a
tactical sequence which should end in White’s favour, but gives Black all sorts of tactical
and practical chances – which is all he ever asked for in any position.
19 Qxc6 bxa4 20 Qxa8 Bb7 21 Qxa5 axb3 22 axb3
This wasn’t such a great deal for Black, who only got two knights for White’s rook and
three connected queenside passers. Black must create quick kingside counterplay, or else he
is just busted if White gets time to push his queenside passed pawns.
22 ... f5!?
This move involves another sacrifice. The anonymous ChessBase annotator gave this
move a dubious mark. I see no other choice for Black, since quiet play is fatal, and White
would simply continue to build on the queenside.
23 exf5 Bxg2
 24 Kxg2
The safe route allows Black dynamic equality. White can also risk the line 24 fxe6!?
Bxh3 25 Bc5 Qb7 (threatening mate on g2) 26 f4 Rf5 27 e7 h6 28 Qb4 Qc6 (28 ... Qf3 is
met with 29 Qc3) 29 fxe5 Rg5 30 Qc3. This is probably a kind of position Tal was after
when he began the forcing sequence with 18 ... b5. The e7-pawn ties down Black’s knight.
I’m not so confident Black generates enough play to justify the material deficit, although Tal
certainly has practical chances, based on the weakened light squares around White’s king. I
have a feeling kid-Tal would have swindled his opponent if they had reached this position.
24 ... gxf5
Now ... f3 and ... f3+ is in the air.
25 Bb6
Alternatively:
a) 25 Qa7 Qf6 26 Bc5 Rg8 27 Bb4 (intending Bc3) 27 ... e4 28 Bc3 Qh6 29 Bxg7+
N8xg7 30 Rd7 f4 31 Rxe4 fxg3 32 Qe3 Ng5 33 h4 Qc6 34 hxg5 Qxd7 35 Qxg3 when I think
Black has more chances to win, than to lose.
b) 25 Qb6 f4 26 Bxf4 Nxf4+ 27 gxf4 Rxf4 28 Qd8 Qe6 29 Qd5 Qg6+ 30 Kh1 and Black
should be able to generate enough play against White’s king, before White is able to push his
queenside pawns down the board. Komodo assesses at dead even.
c) 25 f4? allows 25 ... exf4 26 gxf4 Nxf4+ when Black has a winning attack.
25 ... f4 26 Qd5
Bleks insists on hanging on to all his material. Maybe he should give something back
with 26 Rxe5 Bxe5 27 Qxe5+ N8g7 28 Bd4 with approximately even chances.
26 ... Qf6 27 Bd8?!
Vague unease morphs into outright apprehension. White worries about a knight reaching
g5. His last move eyes a coming ... Ng5, where he intends to chop it with the bishop. Still,
the bishop looks awkward on d8.
 27 ... Qg6
Threatening ... fxg3 and ... Nf4+ cheapos.
28 Qe4 Qh6
Of course, a queen swap is fatal for Black, since it depletes his only trump: White’s
weakened king’s position.
29 Rdd1?
This move passes safety’s last outpost, and his position slowly inches toward the level
of incurable. White plays as if he isn’t of one mind on the correct plan. This looks like a
case of the wrong rook. When under constant pressure, we tend to become inarticulate if
asked about our defensive plans.
29 Red1! is correct, since it enables future Rd6 ideas, as well as Rd8 plans to swap
rooks. It’s funny how even a small course-correction like this, may be enough to save a game
which would otherwise be lost. Following 29 ... Nf6 30 Bxf6 Qxf6 31 h4! (preventing ...
Ng5) Komodo says the position remains even. I would take White’s side if given a choice,
since he has a clear plan to push his queenside pawns. The burden of proof lies upon Black
to generate kingside counterplay.
29 ... Nf6 30 Bxf6
30 Qxe5?? loses on the spot to 30 ... f3+.
30 ... Qxf6 31 Rf1?! Ng5!

Advantage Black, who seizes control over the key light squares e4 and f3, while
watching h3.
32 Qd5 Qf5
Also worth a thought is the line 32 ... f3+ 33 Kh2 Nxh3 34 Qd7 Nxf2!! (not 34 ... Ng5?
35 Rd6!, forcing queens off the board, after which Black is in trouble) 35 Rxf2 e4 36 Rdf1
Qg6 (threat: ... Be5!) 37 Qd5 Bxb2 38 c4 Rg8 39 Rxf3 (39 Rg1?? Qh6+ forces mate next
move; variations appear before us with the incoherence of a half-remembered dream) 39 ...
exf3 40 Rxf3 Qc2+ 41 Kh3 Qe2 42 Qd3 Qe1 and Black has all the winning chances.
33 g4?!
White lacks time for this structural reconstruction effort. His best defensive chance lies
in 33 Qd7 Qe4+ 34 f3 Qxc2+ 35 Rd2 Qxb3 36 Qd5 Qe3.
33 ... f3+ 34 Kh1 Qf4!?
idea ... e4 and ... Be5. An illuminating moment. Young Tal is so completely obsessed
with the delivery of checkmate, that he refuses to pause to pick up a key pawn with 34 ...
Qxc2!, after which White is busted.
35 Qd2 Nxh3!?
Allowing a queen swap puts Black’s win in jeopardy. Correct was 35 ... Qf6! 36 Qd6
Qf7 37 Qd7 Qg6, and if 38 Qd6 Black can backtrack and pick off c2 with 38 ... Qxc2,
retaining a winning position.
36 Qxf4 Nxf4 37 Kh2
Maybe White can try 37 Ra1! preventing 37 ... e4 due to 38 Ra4. After 38 ... Re8 39
Kh2! Bxb2 40 b4 Be5 41 b5 Ne2+ 42 Kh3 Bd4 White still has some chances to save the
game.
37 ... e4!
Freeing his long suffering bishop. Now White’s kingside begins a process of
calcification.
38 Rd7

Exercise (planning): An omen-conscious person doesn’t believe in

coincidence. Too many black pieces hover near White’s king. Despite
the ending status of the game, it just feels like Black has something. White’s
b2-pawn is hanging. Should Black take it? If not, does he have a better plan?
 Answer: Mating net. Black threatens to lift his rook to f6 and then h6, with a mating net.
38 ... Be5! 39 Kg3?
Gulp! Well, I guess nobody lives forever. White, who failed to appreciate the true
danger to his king, had to go into desperation mode with 39 Re7 Ng6+ 40 Rxe5 Nxe5 41
Kg3 e3! 42 fxe3 Kg7 when he may have chances to hold the game.
39 ... Ng2+
Even stronger is 39 ... Nd3+! 40 Kh3 (40 Kh4 Nxf2! is also decisive) 40 ... Nxf2+! 41
Rxf2 e3 42 Rf1 e2 which wins.
40 Kh3
The king belies his nervousness when his tea cup rattles as he places it down in the
saucer.
40 ... Rf6!
Threatening mate.
41 g5
White’s last move allows a forced mate, but 41 Rd8+ Kg7 42 Rd7+ Kg6 43 Re7 Nf4+
44 Kh4 Bd6 45 Ra7 e3! 46 fxe3 Ng2+ 47 Kh3 Nxe3 48 Rf2 Nxc2! 49 Rxc2 f2 is decisive
too.

41 ... Rf4!
There is no good defence to swinging the rook to h4.
42 Kh2
Now the puppet dances to the hidden manipulation of Tal’s strings.
42 ... Rh4+ 43 Kg1 Nf4
Threat: ... Ne2 mate.
44 Rd2 Nh3+ 45 Kh1
 Exercise (combination alert): Black to play and force mate:

Answer: 45 ... Nxg5+

Answer no.2: Even quicker is 45 ... Bd4! 46 Kh2 (or 46 Rxd4 Nf4+ 47 Kg1 Ne2 mate)
46 ... Nxf2+ 47 Kg3 Rg4+ 48 Kh2 (White’s king comes to that awful place in life, where he
sees himself exactly as others see him) 48 ... Be5 mate.
46 Kg1 Bh2+! 0-1
At this point Tal must have indulged in a knowing smile, of one who saw this moment
coming a long time ago. 47 Kh1 Bg3+! 48 Kg1 Nh3+ 49 Kh1 Nxf2+ 50 Kg1 Rh1 is mate.
Note that White’s three queenside passers – which in the end, provided little solace when
compared to his wretched king’s position – didn’t have time to move up even a single
square!

Game 3
M.Tal-G.Lisitsin
USSR Championship, Leningrad 1956
Sicilian Defence

I included this untypical Tal game into the book for sentimental reasons. This is the first Tal
game I ever played over (I think the year was around 1969), and I marvelled at Tal’s
superhuman king, who seemed to single-handedly win the game.

1 e4 c5
Compare the position Tal got, to this one which arises from the Pirc, Austrian Attack: 1
... d6 2 d4 Nf6 3 Nc3 g6 4 f4 Bg7 5 Nf3 c5 6 Bb5+ Bd7 7 e5 Ng4 8 e6 Bxb5 9 exf7+ Kd7
10 Nxb5 Qa5+ 11 Nc3 cxd4 12 Nxd4 h5 13 h3 Nc6 14 Nde2 Nh6. I think White holds an
edge here and he stands far better than in the position Tal reached in the opening of his game.
2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 g6 6 f4

This is now considered a rather harmless line of the Dragon. If Tal were alive today, he
would play the Yugoslav Attack and deliver a quick mate on 99.99 percent of his
opposition!
6 ... Nc6

Question: Why not 6 ... Bg7 and castle first?

Answer: When I was a kid, annotators always accompanied this move with a question
mark. However, today’s comps contradict the assessment, and it’s fully playable for Black:
7 e5 Nh5! (the only move, but an adequate one) 8 Bb5+ Bd7 9 e6 fxe6 10 Nxe6 Bxc3+ 11
bxc3 Qc8 (there is a position in the Pirc, Austrian Attack which is similar to this one, which
actually looks to me like a better version for Black than its Pirc counterpart) 12 Bxd7+
Kxd7! 13 Ng5 Qxc3+ 14 Bd2 Qc4 15 Rb1 b6. White has compensation for the pawn, but no
more. If given a choice, I would take Black, M.Al Modiahki-A.Motylev, Dubai 2005.
7 Nxc6
White’s idea is to follow with the disruptive e5.
7 ... bxc6 8 e5 Nd7!
Avoiding 8 ... dxe5 9 Qxd8+ Kxd8 10 fxe5 Ng4 11 Bf4 Bg7 12 0-0-0+ Bd7 (12 ...
Ke8?? walks into 13 Nb5!) 13 Re1 with a difficult ending for Black, J.Bolbochan-
H.Rossetto, Buenos Aires 1948.
9 exd6 exd6 10 Be3 Be7 11 Qf3
11 Qd2 is White’s main continuation.
11 ... d5
 I already prefer Black, who controls the centre and also looks to have the more
promising attack, if White castles long, since the b-file is open.
12 0-0-0 Bf6
This move allows White to swap away the dark-squared bishops, which I think benefits
Tal’s side. I would play 12 ... 0-0 13 h4? (this move weakens the light squares; 13 g4 is a
better try) 13 ... Nf6 14 Be2 h5 15 Qg3 Bf5 when Black has a grip on the kingside light
squares, similar to what happened in the game continuation, M.Lindinger-V.Golubev,
Hamburg 1999.
13 Bd4!
Otherwise Black’s bishop is just too influential on its diagonal.
13 ... 0-0 14 h4?
This stab fails to fit the position’s template. In some positions we must ruthlessly
amputate stylistic preference, and just play the best move, the way a gambler decides his
horse on the racing form – strictly based on the odds, and not by some ‘lucky hunch’ based
on the horse’s name, which happens to be the same as his mother’s. A steep price tag of
weakened light squares is attached to this move. Sometimes we affirm our individuality with
our errors, and our impulsive acts are often followed by slowly withering enthusiasm, until
it reaches the level of full regret. This rash move is typical of the young Tal, who weakens
his kingside light squares, playing for an imaginary mate. Correct was 14 Qf2.
14 ... Rb8
Also promising is 14 ... Bxd4! 15 Rxd4 Nf6! 16 Bd3 Bg4.
15 Qf2

Question: Shouldn’t White get on with his attack with 15 h5?

Answer: This ‘attack’, much like the creatures in Maurice Sendak’s children’s book
Where The Wild Things Are, is imaginary and without true threat to Black’s king. Your move
fails to take into account Black’s threat: 15 ... Bxd4 16 Rxd4 Qb6, after which the double
attack on d4 and b2 is decisive.
15 ... Rb4! 16 Bxf6 Nxf6 17 a3 Qb6!?
I think this move gives away some of Black’s advantage. I would retain queens with 17
... Ng4! 18 Qd2 Rb8 19 h5 Bf5 when Black’s king looks safer than White’s.
18 Qxb6 Rxb6 19 Na4

Tal begins to fight for control over c5.

Question: Does White’s control over c5 give him an edge?

Answer: Chernev and Hajtun seem to think that White stands better here, but I think this
assessment may be influenced by the final result of the game. To me, Black looks slightly
better, due to his central control and kingside light-square hole on g4.
19 ... Rb7 20 Bd3 Nh5 21 Rhf1 Re7
Seizing the open file.
22 f5!
We tend to respond to confrontation with either withdrawal or fury. Tal hopes to scour
away some of the pawn weaknesses in his position. Sensing trouble, he correctly offers a
pawn to damage Black’s structure and plug the f5 hole with a pawn. The natural 22 Nc5?!
lands him in difficulties after 22 ... Bg4 23 Rde1 (23 Rd2? walks into the tactic 23 ... Nxf4!
which exploits White’s weakened back rank) 23 ... Rxe1+ 24 Rxe1 Nxf4 25 g3 (25 Re7?
Nxg2 26 Rxa7 Nxh4 and Black’s h-pawn looks fast in the queening race) 25 ... Nh5! (not 25
... Ne6?? 26 Nd7! winning material) 26 Kd2 f5 (now ... Nxg3 is a real threat) 27 Re3 f4.
22 ... gxf5
This is okay, but Black would have done better by declining the sacrifice with 22 ...
Ng3! 23 f6 (White’s sacrifice feels unsound after 23 fxg6?! Nxf1 24 gxh7+ Kh8 25 Rxf1 f5,
after which the h7-pawn falls) 23 ... Re5 24 Rfe1 Rfe8 25 Rxe5 Rxe5 26 Nc5 Bg4 when f6
and h4 are potential targets in the future, and White’s austere position – which isn’t
furnished with creature comforts – borders on overextended.
23 Rfe1 Rfe8 24 Rxe7 Rxe7 25 Kd2 Ng3 26 Kc3 f4 27 Kd4
While dark-square weakness is endemic throughout Black’s position, his two extra
pawns also matter. White’s minor war successes (his admittedly impressive looking king)
are not able to compete with his failures on the home front. Tal is still in trouble, despite his
picturesque king.
27 ... Bf5
Stronger was to toss in 27 ... Bg4! 28 Rb1 and only then 28 ... Bf5 with a superior
version of the game’s continuation.
28 Rd2 Re6
The rook makes it clear that c6 isn’t open for rent. It can also swing over to the kingside
and attack h4 along the third rank.
29 Nc5 Rh6?!
Black retains an edge after the correct 29 ... Re1!.

Exercise (planning): Black’s last move, going after White’s h-pawn,

allows White counterplay. Come up with a plan for White.

Answer: King infiltration. White’s active king virtually acts like an extra piece in the
ending. His king threatens both f4 and also infiltration to d6, if Black’s rook vacates the
third rank.
30 Ke5! Bxd3 31 cxd3 Rxh4
Covering f4, but allowing king entry to d6.
 32 Kd6!
White doesn’t stand worse here, despite Black’s two extra pawns, since his super-active
king makes up the difference.
32 ... Rh6+
Alternatively:
a) 32 ... Nf5+ 33 Kxc6 Rh2! (threat: ... f3) 34 Nb3 Ne3 35 Nd4 Rxg2 36 Rxg2+ Nxg2 37
Kxd5 h5 38 Ke4 h4 39 b4 h3 40 Nf3 Kf8 41 b5 Ke7 42 a4 f6 43 a5 Kd6 44 d4 Ne3 45 Kxf4
Nc4 46 a6 Kd5 47 Kg3 h2 48 Kxh2 Nb6! (48 ... Ke4?? 49 Nd2+! Nxd2 50 b6 wins) 49 Kg3
Ke4 50 Kg4 Nd5 51 Kg3 when neither side can make progress and the game is drawn.
b) 32 ... Rh2 33 Rf2 Nf5+ 34 Kc7! Ne3 35 Rxf4 Rxg2 36 b4, and if 36 ... Rg6 37 Kb7
h5 38 Kxa7 Rg4 39 Rf2 when it’s anybody’s game.
33 Kc7
Threatening b4 and Kb7, picking up Black’s a-pawn.
33 ... Nf5 34 Kb7 Nd4?!
Black should hold the game after 34 ... Rh2! (this idea proves to be a countering source
of irritation for White) 35 Rf2 Ne3 36 b4 Rxg2 37 Rxf4 Rg4 38 Rf2 h5 39 Kxa7 h4 40 Rh2
f5 41 Kb6 f4 42 Nd7 (threatening a cheapo on f6) 42 ... Kg7 43 Kxc6 f3 44 Ne5 Rf4 45
Nxf3 Rxf3 46 Rxh4 Nc2 47 b5 Rxd3 48 b6 d4 49 b7 Rc3+ 50 Kb6 Rb3+ 51 Kc7 Nxa3 52
b8Q Rxb8 53 Kxb8 and the rook versus knight ending is a trivial draw for Black.

Exercise (critical decision): The advantage begins to

swing to White, but only if we find the correct move. Our
candidates are 35 Kxa7 and 35 Rf2. Which one should White play?

Answer: 35 Rf2!
In anticipation of Black’s threatened ... Rh2! followed by ... f3. The hasty 35 Kxa7??
allows Black his intent with 35 ... Rh2! 36 a4 f3 37 a5 fxg2 and game over.
35 ... a5?
This blunder robs Black of a chunk of efficiency. Black may still save the game after 35
... Rh2! 36 Rxf4 Ne6 37 Rg4+ Kf8 38 Nxe6+ fxe6 39 Kxa7 Ke7 40 a4 h5 41 Rg8 Kf7 42
Rg3 Rh4! 43 a5 Rb4 44 a6 Rxb2 45 Rh3 c5 46 Rxh5 Kf6 47 Ka8 c4 48 dxc4 dxc4 49 Rc5
Rxg2 50 Rxc4 Rb2 with a drawn rook and pawn ending.
36 Rxf4
More accurate is 36 a4!.
36 ... Ne6 37 Rg4+ Kf8 38 Kxc6!
For many moves White’s king cast longing eyes upon the c6-pawn. Black has no useful
discovery.
38 ... Nxc5+ 39 Kxc5 Re6
Likewise, if 39 ... Ke7 40 b4 axb4 41 axb4 Ke6 42 b5 Rh1 43 Rb4 Rc1+ 44 Kd4 Rc7
45 b6 Rb7 46 Kc5 Kd7 47 Kxd5 and White wins.
40 Kxd5 Rb6 41 b4
The rook and pawn ending is a trivial win for White.
41 ... axb4 42 axb4 Ke7 43 Kc5 Rf6 44 Rd4!

Cutting off Black’s king.

44 ... Rf5+ 45 Kb6 Rf6+ 46 Kc7 Rf5 47 Re4+!
Driving Black’s king even further away.
47 ... Kf6 48 Kc6 Rf2 49 g4 h5 50 gxh5 Kg5
Black’s lone passed pawn isn’t enough to save him.
51 b5 f5 52 Rb4 f4 53 b6 f3
Black’s f-pawn moves slower than a day of fasting.
54 b7 1-0
The end might be 54 ... Rc2+ 55 Kb5 f2 56 b8Q f1Q (if you aid and abet a murder, are
you not equally guilty as the one who pulls the actual trigger? Black promotes to a new
queen, only at the cost of his king, and as we all understand, there is no second place trophy
in a chess game) 57 Qg3+ (we note that there is an awful lot of fawning and curtsying from
Black’s pieces, when they are in the presence of White’s tyrannical queen) 57 ... Kf6 58
Qg6+ Ke7 59 Re4+ Kd8 60 Qd6+ Kc8 61 Re8+ Kb7 62 Qb6 mate. “I view you as an
unnecessary complication in my life,” says the queen.

Game 4
M.Tal-A.Tolush
USSR Championship, Leningrad 1956
Sicilian Defence

1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 a6 6 Bg5 e6 7 f4 Qb6 8 Qd2 Qxb2 9 Rb1
Spassky famously crushed Fischer with 9 Nb3 in their world championship match: 9 ...
Qa3 10 Bxf6 gxf6 11 Be2 h5 12 0-0 Nc6 13 Kh1 Bd7 14 Nb1 Qb4 15 Qe3 d5? (a violation
of the principle: Avoid opening the position when behind in development; Black is
probably still okay after 15 ... Ne7) 16 exd5 Ne7 17 c4 Nf5 18 Qd3 h4? 19 Bg4 Nd6 20
N1d2 f5 21 a3 Qb6 22 c5 Qb5 23 Qc3 fxg4 24 a4 and Black’s queen is trapped, B.Spassky-
R.Fischer, Reykjavik 1972.
9 ... Qa3 10 e5!?

Principle: Open the game and create confrontation when leading in development.
Today, this line exists on the fringes of disreputability (although some brave Shirovian GMs
still adhere to it), and has mostly been replaced with 10 f5.
10 ... dxe5 11 fxe5
White gets an open f-file and displaces the f6-knight, at the cost of a chronically weak e-
pawn.
 11 ... Nfd7 12 Ne4!?
White’s most principled move. Tal offers his a-pawn and dodges ... Bb4 and ... Qa5
ideas, while eyeing the d6- and f6-squares. Instead, after 12 Bc4 Qa5 13 0-0 Nxe5 14 Rbe1
Nxc4! (or 14 ... Nbc6 15 Nxc6 Nxc6 16 Qf4 Bc5+ 17 Kh1 0-0 18 Ne4 Be7 and objectively,
Black should survive and convert, although it won’t be easy with so many pieces hovering
over his king, A.Kosten-J.Tisdall, London 1982) 15 Qf4 Nd6 16 Ne4 Qc7 17 c4 Nb5 18
cxb5 Qxf4 19 Bxf4 f6 I prefer Black, despite the fact that his development stands at zero.
Queens have been removed from the board and Black’s two pawns should be decisive,
since his position remains almost weakness free, H.Gaida-S.Jalanskis, correspondence
2002.
12 ... Qxa2
Black can also toss in 12 ... h6 13 Bh4 Qxa2 14 Rd1 Qd5 15 Qe3 Qxe5 16 Be2 Bc5 17
Bg3 Bxd4 18 Rxd4 Qa5+ 19 Rd2. The comps like Black, while White scores 72.6% from
583 games.

Question: Why then the discrepancy between

computer evaluation and human performance?

Answer: Conclusion: White’s practical chances may outweigh Black’s objective

advantage, mainly since we humans have the bad habit of blundering when on the defensive
– even when our position is objectively a good one.
13 Rb3
This allows Black to disrupt White’s castling. Superior is 13 Rd1 Qd5 14 Qe3 Qxe5 15
Be2 Bc5 16 0-0. White may have full attacking compensation for his three missing pawns,
I.Cheparinov-M.Roeder, Kallithea 2008.
13 ... Qa1+ 14 Kf2
 This position occurred seven times in my database. Black’s score? Zero!

Question: Is the position inherently bad for Black?

Answer: Not at all. Komodo assesses at dead even. The problem, of course, is that
White generates a fierce initiative for the sacrificed pawns, and for a human, navigating the
maze for Black is extraordinarily difficult to achieve, on a practical level. The only player I
can think of who consistently won from Black’s side of the Poisoned Pawn line was Fischer,
and even he got clobbered in the line when he faced Spassky.
14 ... Qa4?
A genie’s deepest wish is to escape the confines of her bottle. But then when she gets
out, she must serve Larry Hagman (Major Nelson) for the remainder of his life.
Komodo suggests the unplayed 14 ... h6! 15 Rc3 Nc6 16 Nxc6 bxc6 17 Bb5!! Qxh1 18
Bxc6 hxg5 19 Bxa8. Now Black escapes with 19 ... Bc5+! 20 Rxc5 Nxc5 21 Nd6+! Ke7 22
Qxg5+ Kd7 23 Qd2 f6 24 Qd4 fxe5 25 Qxc5 Rf8+ 26 Bf3 Rxf3+! 27 gxf3 Qxh2+ 28 Kf1
Qh1+ 29 Ke2 Qg2+ 30 Ke1 Qh1+ and White should take perpetual check, since 31 Kd2?
Qxf3 32 Nxc8?? loses to 32 ... Qd5+ 33 Qxd5+ exd5 with too many pawns for the piece.
 Exercise (critical decision): White must sacrifice before Black catches up
in development. Our candidate moves are 15 Nxe6 and 15 Bb5. One of
them wins, while White stands worse in the other. Which one should Tal play?

15 Bb5?!
Attacks are creatures who grow surly when unfed. Nature designed Tal’s brain for
imagination, not specificity. Korchnoi once had the colossal gall to call Tal a “weak
calculator”. What he meant by that was that Tal calculated intuitively, rather than in a pure
numbers-cruncher-fashion, like Lasker, Korchnoi and Kasparov. “The wrong piece and on
the wrong square!” writes Kasparov.
Answer: Tal incorrectly rejected 15 Nxe6!!. Clearance. Sometimes a simple geometric
association may spring a hidden combination’s lock within our mind: 15 ... fxe6 16 Nd6+
Bxd6 17 Qxd6 (threatening mate on e7) 17 ... Rf8+ 18 Kg3 Nf6 (the only move; if 18 ...
Rf7?? 19 Qxe6+ Kf8 20 Bc4 and Black must resign) 19 exf6 gxf6 20 Be2! (threatening a
nasty check on h5) 20 ... Qd7 21 Bh5+ Rf7 22 Qc5 Qc6 23 Qxc6+ Nxc6 24 Bxf6 with a
winning position for White.
15 ... axb5 16 Nxb5 f6!
Living in a harsh, unforgiving environment tends to burn away all traces of
sentimentality, since our prime directive is survival itself. Black must take action, since the
growing threat of White’s attack can no longer be tolerated. Tolush finds the only defence,
giving air to his king. 16 ... Qxe4?? walks into 17 Nc7 mate.
17 exf6!
Tal offers a second piece avoiding the natural but weak continuation 17 Ned6+? Bxd6
18 Nxd6+ Ke7 19 exf6+ gxf6. Black remains up a piece, while his king remains safe
enough, for now at least.
 17 ... gxf6?!
A move based on the philosophy: just because we don’t fully comprehend a danger,
doesn’t mean we should deny its existence. Black incorrectly declined. Sometimes I get
jealous of my chess comps and begin to believe that my books arise from their non-existent
minds, rather than my own faulty, yet existent one. In this position though, there is no way
any human’s mind is capable of working his way to the correct path – at least without the
help of chess computers.
Black should eliminate a dangerous attacker with 17 ... Qxe4! (sometimes it’s just easier
to say “yes” to an opponent’s demands, rather than expend the energy required to resist it)
18 fxg7 Bc5+ 19 Re3 Qf5+ 20 Ke2 Qg4+ 21 Ke1 Ra1+ 22 Kf2 Qf5+ 23 Ke2 Rg8! (23 ...
Qg4+ is perpetual check) 24 Rxa1 Qg4+ 25 Ke1 Rxg7 26 Rg3 Qe4+ 27 Kd1 Na6 28 Nd6+
Bxd6 29 Qxd6 Qe5!. Black forces the queens off the board and retains all the winning
chances in the coming ending.
17 ... Nxf6 is also superior to Tolush’s choice: 18 Nxf6+ gxf6 19 Bxf6 Bc5+ 20 Ke1
Rg8 21 Nc7+ Kf7 22 Rf3 Qe4+ 23 Kd1 Bb4 24 Qxb4 Qxb4 25 Bc3+ Kg6 26 Bxb4 Ra1+ 27
Ke2 Rxh1 28 Rg3+ Kf7 29 Rf3+ and the game ends in perpetual check.
18 Re1!!

A luminous example of Tal’s intuitive approach to irrational positions. His move is

played with the philosophy: a magnifying glass can’t generate energy without sunlight. In
order to reach a correct conclusion, we must exclude the insignificant (his hanging bishop
and knight!), just as much as we must include the necessary (his once dormant rook entering
the battle via e1). Tal is determined to sequester every possible resource to his war effort,
even at the risk of depleting his attacking forces through sacrifice. This is his third piece
offer, which Kasparov called “a brilliant flash of genius.” Every white piece participates in
Tal’s attack.
Everyone else on earth would play 18 Nxf6+?! Nxf6 19 Bxf6 Bc5+ 20 Ke1 Qe4+ 21
Kd1 e5! (threat: ... Ra1+) 22 Nc7+ Kf7 23 Nxa8 Bg4+ 24 Kc1 Kxf6 25 Qh6+ Qg6 26 Rf1+
Ke6 27 Qd2 with an exceedingly difficult position to assess. Black is up material, but his
king still remains in danger. The comps say dead even, which is code for unclear.
18 ... Ra6?
Tolush fights for ownership over d6, which lapses unresolved. Correct was the
counterintuitive 18 ... Nc6! (18 ... fxg5?? 19 Nc7+ now wins, in view of 19 ... Kf7 20 Rf3+)
19 Bxf6 Qxe4 20 Nc7+ Kf7 21 Bxh8!? (or 21 Rxe4 Nxf6 22 Rf3 Ra5 23 Rxf6+ Kxf6 24
Qc3+ e5 and Black’s king still doesn’t look very safe to me) 21 ... Qh4+ 22 Kg1 Ra3!
(preventing White from slipping his rook into the attack, via the third rank) 23 Nxe6 Rxb3
24 Nxf8 Nxf8 25 cxb3 Ne6 when Black stands worse, but harbours reasonable chances to
survive.
19 Bxf6 Nxf6 20 Nxf6+ Kf7 21 Rf3
White’s attack is out of control. Even stronger was 21 Ne4!, keeping Black’s queen out
of defensive reach: 21 ... Rg8 22 Qf4+ Ke8 23 Nc7+ Kd8 24 Nxa6 Nxa6 25 Qf7! leaves
Black completely busted.
21 ... Qh4+ 22 Kf1!?

Exercise (critical decision): Black can play 22 ... Qc4+,

picking up the b5-knight. Should he do it?

Answer: It’s a trap! Tolush correctly avoids the c4 check.

22 ... e5!
Black’s defensive idea, although slow in presenting itself, finally begins to assume form.
It takes a brave person to willingly allow a discovered check. Tolush devotes the full force
of his concentration upon the safety of his lightly guarded king. With his last move he allows
his a6-rook to participate in the defence, while freeing his light-squared bishop.
 Tolush correctly avoids Tal’s trap with the greedy 22 ... Qc4+?? 23 Kg1 (of course,
White is completely busted if he saves the knight at the cost of swapping queens, with 23
Qe2? Qxe2+ 24 Rxe2 Ke7; Black’s king is safe and he will convert his extra piece in the
ending) 23 ... Qxb5 24 Nd5+! Ke8 (likewise, if 24 ... Kg8 25 Qg5+ Bg7 26 Ne7 mate or 24
... Kg6 25 Rf6+ Kh5 26 Re5+ Kg4 27 Qf4 mate) 25 Nc7+ Ke7 26 Nxb5. Black loses his
queen and his king is unable to escape mate.
23 Qd5+?!
In positions of such a complexity level, it’s virtually impossible to find the best
pathways. 23 Nh5+! Kg6 24 Rg3+ Bg4 25 h3 is decisive.
23 ... Be6!
The psychologically most difficult thing about defending, is that we must temporarily
place aside our desire for a just retribution. Tolush keeps finding only moves. 23 ... Kg6?
allows 24 Ne8! Bg4 25 Qe4+ Kh5 (the king only finds temporary shelter in his abandoned
hovel) 26 Qxe5+ Qg5 27 Qxh8 Qxb5+ 28 Kg1 Bxf3 29 Re5+ winning.
24 Nd7+?!
A once small crack begins to widen. With each move, Tal’s advantage decreases.
Correct was 24 Nh5+! (the long awaited apocalyptic final battle begins) 24 ... Kg6 25 Qxe5
(threat: Rf6+) 25 ... Nd7 26 Rg3+ Bg4 27 Qxh8 and then:
a) 27 ... Kh6! is Black’s most tenacious defence: 28 Nc7 Qxh5 29 Nxa6 bxa6 30 h3
Qf7+ 31 Kg1 Bf5 32 Qa1!! is not such an easy move to find over the board. Threat: Qc1+,
followed by Qg5 mate. Amazingly, Black’s minor pieces are no match for White’s rooks,
since the x-factor is Black’s unsafe king. 32 ... Bg6 33 Rf3 Qd5 34 Rd1! Rxd7 is coming. 34
... Qc6 35 Rxd7! Bc5+ 36 Rd4 Qd5 37 c3 leaves White up with too much material.
b) 27 ... Kxh5? 28 Qxh7+ Bh6 29 Qf7+ Rg6 30 Re6 Nf6 31 Re5+ Bg5 32 Nd4! Kh6 (32
... Qxh2? 33 Rxg5+! Kxg5 34 Nf3+ wins) 33 Qf8+ Kh5 34 Nf3! Bxf3 35 Rexg5+!, and if 35
... Rxg5 36 Qf7+ Kh6 37 Qxf6+ Kh5 38 Rxg5+ Qxg5 39 Qxg5+ Kxg5 40 gxf3 with an easily
winning king and pawn ending.
24 ... Kg6?
The position’s truth merely flickers at the rim of both players’ consciousness. Black just
might save the game with 24 ... Ke7! 25 Qc5+ Ke8 26 Qc8+ Ke7 27 Qxb8 Bg7! 28 Qxb7
Qc4+ 29 Kg1 Qc8! 30 Qc7 Qxc7 31 Nxc7 Rc6 32 Nxe6 Kxe6 33 Rg3 Kxd7 34 Rxg7+ Kd6,
with chances to hold the draw.
 25 Nxe5+ Kg7
The king gingerly places one foot forward, the way an inexperienced tight rope walker
tests his balance. 25 ... Kh5 is of no help: 26 Qd1!! (threatening a nasty discovered check
with his f3-rook; Kasparov is of the opinion that Tal would have found this comp-shot over
the board) 26 ... Kh6 27 Qd2+ Qg5 and now White wins with the pretty deflection
combination 28 Rh3+! Bxh3 29 Nf7+, forking king and queen.
26 Rg3+! Qxg3
Not much choice in the matter. Black must place aside his old ways of material comfort,
with only bleak deprivation in his future:
a) 26 ... Kh6 27 Nf7+! Bxf7 28 Qd2+ Kh5 29 Re5+ Qg5 30 Qxg5 mate.
b) 26 ... Kf6 27 Qd8+ Be7 28 Qxh8+ Kf5 29 Rf3+ Qf4 30 Qxh7+ Kg5 31 h4+ Qxh4 32
Qg6 is also mate. “You are my honoured guest. Welcome to my home,” says the queen.
27 Qxb7+ Nd7 28 hxg3
Black’s problem is that the d7-knight is pinned, while he is already behind in material,
with an insecure king. Conclusion: Black is completely busted.
28 ... Rb6 29 Qc7 Bc5
Or 29 ... Rxb5 30 Nxd7.
30 Nxd7!
Tal, unafraid of ghosts, correctly ignores the coming ... Bc4+.
30 ... Bc4+ 31 Re2 1-0
31 ... Bxe2+ 32 Kxe2 Re8+ 33 Kd3 leaves Black down too deep materially.

Game 5
M.Tal-V.Simagin
USSR Championship, Leningrad 1956
Caro-Kann Defence
1 e4 c6
The position we reach normally arises after the move order 1 ... d6 2 d4 Nf6 3 Nc3 c6.
2 d4 d6!? 3 Nc3
3 c4 either leads to a King’s Indian or Old Indian set-up.
3 ... Nf6 4 f4 Qb6?!

A move made with the philosophy: only a rigid mind perceives unorthodoxy as a crime,
to be punished. I’m all for creativity, but this is perhaps going too far. We get the distinct
feeling that this bizarre home remedy tincture isn’t likely to heal Black’s soon-to-be-ailing
position.

Question: Why did Black bring his queen out so early in the game?

Answer: For now, Black submerges his intent. The trouble is not that Black brought the
queen out. Instead, she was brought to the wrong square. Simagin probably hoped his queen
move tied Tal’s dark-squared bishop down to defence of b2. It ended up being a target for
White’s pieces.
Better was 4 ... Qa5 (Black threatens ... Nxe4 and plays for an ... e5 break) 5 e5 Ne4 6
Bd2 (White hands over the bishop-pair; the main lines run 6 Qf3 and 6 Bd3 Nxc3 7 Qd2) 6
... Nxd2 7 Qxd2 d5 8 Nf3 Bg4 9 Be2 e6 10 h3 Bxf3 11 Bxf3 c5 and Black achieved a decent
French-like position, without a bad, light-squared bishop, D.Grazian-C.Lakdawala, San
Diego (rapid) 2010.
5 Nf3 Bg4 6 Be2 Nbd7 7 e5!
Tal plays e5 before his opponent can do so.
7 ... Nd5
After 7 ... dxe5?! White has a pleasant choice of opening the f-file with 8 fxe5 or playing
8 dxe5, after which Black is virtually forced to sacrifice a pawn with 8 ... Nd5 9 Nxd5 cxd5
10 Qxd5 e6 11 Qb3 when he lacks compensation.
8 0-0 Nxc3!?
This move only helps White. I would avoid opening the position and play 8 ... e6.
9 bxc3
Of course, White is happy to capture with the pawn, since this strengthens his centre,
while opening the b-file in preparation for a tempo-gaining Rb1.
9 ... e6 10 Ng5!
When an opponent dared to provoke Tal, his mood grew excitable – which for him was
synonymous with murder. Tal’s last move was played with an undertone of implied warning.
He increases his development lead to dangerous levels, as he contemplates knight sacrifices
on either e6 or f7.
10 ... Bxe2 11 Qxe2 h6
Black’s best move in an already bad situation:
a) 11 ... g6? fails to halt White’s f5 ambitions: 12 f5! gxf5 (12 ... exf5?? loses
instantaneously to 13 exd6+) 13 Qh5 Nxe5 14 Nxe6 Ng6 15 Nxf8 Nxf8 16 Re1+ Kd8 17
Qxf7 and Black soon resigned in A.Pereira-F.Ribeiro, Lisbon 1995.
b) 11 ... Be7? 12 f5! is a winning attack for White.
12 Nxf7!

Black’s king goes for a ride. I bet Tal made this move instantly. The decision to sacrifice
and play for mate was really made on his 10th move, and it’s too late for a change of heart.
12 ... Kxf7 13 f5
Central lines open with alarming rapidity.
13 ... dxe5
Instead, 13 ... Nxe5 14 Be3 regains the piece, since Black can’t afford 14 ... Nd7?? 15
fxe6+ Kxe6?? 16 d5+ winning the house, or if 13 ... Kg8 14 exd6! Nf6 15 Bf4 exf5 16 Qe6+
Kh7 17 Qxf5+ Kg8 18 Qg6 Qd8 19 Be5 Nd5 20 Rxf8+! Qxf8 21 Rf1 Nf6 22 Rxf6 Qe8 23
Qg3 Rh7 24 c4 with a dominating position for White.
14 fxe6+
Even stronger is the line 14 Qh5+! Ke7 (14 ... Kg8 15 fxe6 instantly regains the piece,
since any knight move is fatal: for example: 15 ... Nf6?? 16 Qf7+ Kh7 17 Rxf6 and there is
no defence to the coming Rxh6 mate) 15 fxe6 Kxe6 16 Be3 Qa5 (16 ... Nf6 17 d5+ cxd5 18
Qf5+ wins the queen) 17 dxe5 Nxe5 18 Qf5+ Kd6 19 Bf4 Re8 20 Rae1 and there is no
remedy to Rxe5, followed by Re1.
14 ... Kxe6 15 Rb1!?
This visually startling shot (incorrectly given an exclamation mark by Irving Chernev,
and awarded an overly generous double exclam by Tal and Damsky) may only be White’s
second best continuation. Black is unlikely to survive the simple 15 Be3! (threat: d5+ and
Bxb6) 15 ... Qa5 16 Qg4+ Ke7 (16 ... Kd6? 17 dxe5+ Kc7 18 Rf7 Rd8 19 Rd1 is hopeless
for Black) 17 dxe5 Nxe5 18 Qf5 and it becomes obvious that Black’s king won’t last long.

15 ... Qxb1!
Black’s only move.

Question: I see that Black’s last move loses his queen. Can he decline with 15 ... Qa6?

Answer: In the case of the decline, White whips up his attack with 16 Qg4+ Kd6 17
dxe5+ Kc7 18 Bf4 when Black can resign.
16 Qc4+!
Forcing Black’s king to a dark square, after which Tal wins his opponent’s queen with
the a3 discovery.
16 ... Kd6?
The king decides to pack up stakes and move out West, to begin a new sister-free life.
But then he soon remembers she can still visit. We may reach an eminently logical
conclusion, based on reasoned argument. Yet if our calculation is off by even a hair, we will
still be wrong. Believe it or not, this natural move, where Black’s king heads for refuge on
c7, is the losing move.
Much stronger is the counter-intuitive 16 ... Ke7!!. No human in the world would play
this move, which looks like suicide. The move may actually save Black, in a position which
feels utterly vacant of defensive opportunity: 17 Ba3+ Kd8 18 Rxb1 Bxa3 19 Rxb7 Rc8 20
Rxa7 Bd6 21 Qf7 Rc7 22 Ra8+ Rc8 23 Ra6 Rf8 24 Qxg7 Rf6 and Black still retains a
prayer for survival. How painful when our opponent becomes the unexpected beneficiary of
favourable geometry – without doing a damn thing to deserve it!
17 Ba3+ Kc7 18 Rxb1 Bxa3

Exercise (combination alert): Do you see Tal’s little tactic, which he

must have foreseen when he began his combination with 15 Rb1! ?

Answer: Double attack. Not only does White pick off b7, but he also drives Black’s
king into the middle of the board.
19 Qb3! Be7 20 Qxb7+ Kd6
Black’s king is clearly housed in a dilapidated fortress. There isn’t much choice, since
dropping back to d8 hangs two rooks.
21 dxe5+ Nxe5 22 Rd1+ Ke6
 Question: Doesn’t Black have plenty of material for his queen?

Answer: Black gets two minor pieces and a rook for the queen – more than enough.
However, material parity is not the issue. Black’s king is in dire trouble in the middle, and
this factor is destined to cost him material later on.
23 Qb3+! Kf5
23 ... Kf6 24 Rf1+ Kg6 is close to the game’s continuation. I can’t tell you what comes
next, otherwise I give away the answer to a future combination alert!
24 Rf1+ Ke4!?
Simagin clearly has a death wish, where his king oversteps the bounds of neutral
observer, to active participant in the attack against him. Common wisdom dictates that it’s
not such a good thing to have your king strolling about the middle of the board, when playing
Mikhail Tal! When our plans go horribly wrong, we become Dr. Frankenstein, who
continues to place hope in his diseased creation. 24 ... Kg6 once again reaches the position
we talked about in the earlier note.
25 Re1+
This move, which disrupts White’s timetable for war, violates the principle: When
attacking, don’t chase the enemy king. Instead, cut off exit squares. Even stronger is 25
Qe6! (threat Re1+) 25 ... Bc5+ (25 ... Bf6 26 g3! threatens Rf4+, driving the king even
deeper into White’s territory, and if 26 ... Bg5 27 Rf5 Bf6 28 Kf2! Rad8 29 Rf4 mate) 26
Kh1 Rhe8 27 Re1+ Be3 (or 27 ... Kf4 28 g3+ Kg5 29 Rxe5 mate) 28 Qh3! Nc4 29 Qh5!
Rad8 30 Qc5! and wins.
25 ... Kf5 26 g4+!
Another attacker is introduced. Both king and knight are overloaded and unable to
capture g4.
 26 ... Kf6 27 Rf1+ Kg6

Exercise (combination alert): An otherwise honest person may surrender to the

complaints of an empty stomach to steal food. White to play and win material:

Answer: Double attack/pin.

28 Qe6+! Kh7
The trouble with 28 ... Bf6 is that Black loses the piece anyway, after 29 Qf5+ Kf7 and
now White exploits the pin with 30 Qxe5.
29 Qxe5 Rhe8 30 Rf7 Bf8
Naturally not 30 ... Bf6?? 31 Qxf6.
31 Qf5+ Kg8
The king’s new job is to be his sister’s yes-man. His duties: smile, nod and
enthusiastically say “Yes!” to her demands, no matter how irrational.
32 Kf2!
Tal prevents ... Re2.
32 ... Bc5+ 33 Kg3! Re3+ 34 Kh4!
The two kings reversed roles, with Tal’s on the fourth rank, and Black’s back at home.
34 ... Rae8
34 ... g5+?? is a helpmate after 35 Kh5 Rh3+ 36 Kg6 and Black can resign.
 Exercise (critical decision): This is getting pretty scary
for White’s king. How does Tal make progress?

Answer: Return the exchange, for which White gets a safe king and begins to pick off
black pawns.
35 Rxg7+!
Extra material can be interpreted either as ammunition, to be stockpiled for a future
battle, or as wealth, to be spent and invested. In this case, Tal goes with the latter
interpretation.
35 ... Kxg7 36 Qxc5
 Exercise (critical decision): Black has a choice of covering either his
c6- or a7-pawn. He can play 36 ... R8e7, or 36 ... R8e6. Only one of the
moves offers any hope of salvation. Which one should he play?

36 ... R8e6?
Tal’s relentless pressure finally demands its cumulative toll, either through exhaustion,
time shortage, or a combination of both. Simagin covers the wrong pawn.
Answer: Black had to try 36 ... R8e7!. The trick to surviving time pressure is to grab
hold of a single essential reality, in a world of hallucination and shadow. In this case
Black’s simple organizing principle is: blockade on the second rank. After 37 Qxc6 Rf7 it
isn’t so easy for White to find a puncture what is essentially a closed circle. Black’s two
remaining pawns remain safe for now and he can blockade passed ‘c’ and the kingside
passer on the second rank. I’m not so confident Black can actually draw this, but this line
was his only hope.
37 Qxa7+
Now White’s a-pawn is ready to roll forward.
37 ... Kg6 38 Qa8 Kf6 39 a4
Black is curiously out of synch, helpless to halt the advance of the a-pawn.
39 ... Ke5 40 a5 Kd5 41 Qd8+! Ke4
Simagin avoids Tal’s filthy little cheapo 41 ... Kc5?? (now Black’s king is unable to
escape the queen’s dragnet) 42 Qd4+ Kb5 43 Qb4+ Ka6 44 Qb6 mate.
42 a6 Kf3
Black desperately attempts to concoct a non-existent mating net around White’s king.
43 a7 Re2 44 Qd3+
Of course, Tal isn’t likely to fall for the idiotic 44 a8Q?? Rxh2 mate.
 44 ... R2e3

Exercise (combination alert): Psychologically, it’s very difficult to spot a mate if our
position is an easy win, since we don’t bother to look. The black king’s back aches from
carrying the weight of an impossible defensive burden. Here, we come to the rarest of
places, where the great Mikhail Tal missed a simple mate. Do you see it?

45 Qxe3+ 1-0
The wealthy are more easily able to elude the law’s punishments, when compared to
poor criminals. The pangs which stab us the most are hasty decisions, now impossible to
undo. “I shall assume command of the realm, while you take up your new duties as ‘Vice-
Monarch,’” says the queen, as she informs her brother of his demotion. The combination’s
solution remains incommunicably apart from Tal’s mind, since he fell into a robotic
‘everything-wins’ mode. Of course, Tal’s move had the same effect as mate, since it induced
resignation.
Answer: Aesthetically superior would be 45 Qf1+! Ke4 (Black’s king, like Moses,
dreamed of leading his people to the Promised Land, yet perished before seeing it himself)
46 Qf5 mate.

Game 6
M.Tal-A.Tolush
USSR Championship, Moscow 1957
King’s Indian Defence

1 c4
Talk about pressure. Tal, Tolush and Bronstein were all tied for first in this final round
of the USSR Championship. Bronstein only managed to draw, while Tal eliminated his only
other rival with this win. IM Tibor Karolyi theorizes that Tal played 1 c4 to avoid Tolush’s
Nimzo-Indian. In any case, Tolush didn’t look at home in the King’s Indian.
1 ... Nf6 2 Nc3 g6 3 e4 d6 4 d4 Bg7 5 f3 e5!?

This move order is a bit odd and reduces Black’s options.

Question: Why isn’t Black harmed from the fact that he loses castling rights?

Answer: It’s an ending, so Black’s king is better off in the centre anyway. Also, White
must deal with his d4 hole. Black stands at least equal in that position. Normal would be 5
... 0-0. after which Black has a choice of ... e5, ... c5, and ... Nc6, ... a6, and ... Rb8 set-ups.
6 Nge2 Nbd7 7 Bg5
This must have come as a surprise for Tolush, who probably expected 7 Be3.
7 ... c6
Black can later play for ... a6 and ... b5.

Question: Shouldn’t Black toss in ... h6?

Answer: On 7 ... h6?! White simply retreats with 8 Be3 followed by d5, Qd2, g4, h4 and
queenside castling. Black will regret tossing in his ‘free’ ... h6 move, since it allows White
easier access to pry open Black’s king.
8 Qd2 0-0 9 d5
Instead, after 9 0-0-0 Qa5?! (Black is better off with 9 ... a6) 10 Kb1 a6 11 Nd5! Qxd2
12 Ne7+ (this move ensures the gain of the bishop-pair, and Black’s good bishop with it) 12
... Kh8 13 Rxd2 Nb6 14 dxe5 dxe5 15 c5 Nbd7 16 Be3 Re8 17 Nxc8 Raxc8 18 Nc1 Bf8 19
Nb3 Kg7 20 g3!, Bh3 is coming and Black ended up in a miserable ending, V.Korchnoi-
P.Thipsay, Kolkata 2000. This game is annotated in Korchnoi: Move by Move.
9 ... c5?

A serious strategic error and a violation of the principle: Don’t close the centre when
attacked on the wing. Black should go for 9 ... cxd5 10 cxd5 a6 when even then, I don’t
really like his chances after 11 g4.
10 g4 a6 11 Ng3
Tal may already be dreaming of a future Nf5 sacrifice. His move clamps down on
Black’s possible ... b5 breaks. Black landed in an awful Sämisch set-up, since he is unable
to even dream about an ... f5 break, and he self-clogged the queenside. Meanwhile, White
can leisurely build with h4 and h5.
11 ... Re8
It’s a bad sign when Black prepares to hunker down with ... Nf8 in a KID.
12 h4 Qa5 13 Bh6!?
Generally a swap of dark-squared bishops helps Black, since that is White’s good
bishop. However, Tal feels that such positional considerations evaporate, since the threat of
checkmate supersedes. I’m not so sure. A more strategically minded player would toss in 13
h5 and leave Bh6 for a more opportune moment.
13 ... Nf8
Black may just survive a line like 13 ... Bxh6 14 Qxh6 Kh8! (preparing to eject the
queen with ... Ng8 next) 15 h5 Ng8 16 Qd2 Nf8 17 g5 Qd8, intending ... f6 next, with some
counterplay.
14 h5 Qc7
Perhaps Tolush originally intended 14 ... Bd7?, but then noticed that White has 15 Nf5!.
The knight can’t be accepted and White has a strategically won game.
15 Bd3 b5!
Black needs a distraction.
 16 0-0-0!?
Of course Tal is not to be bought off and appeased with a line like 16 cxb5 c4 17 Bxg7
Kxg7 18 b6 Qxb6 19 Bxc4. White has an extra pawn, while Black has only faint hopes to
generate something along the open b- and c-files.
16 ... bxc4 17 Bb1!?
Tal refuses to be distracted, so he slides his bishop to b1 to keep queenside lines as
closed as possible. 17 Bxc4 also favours White.
17 ... Bh8
Now Nf5 becomes all the more tempting. Tolush probably found the line 17 ... Bxh6! 18
Qxh6 Kh8 19 Nf5 Bxf5 20 exf5 Ng8 21 Qe3 gxf5 22 Bxf5 f6 23 Ne4 rather unappetizing for
Black. Yet this looks better than what he got in the game.
18 Rdg1
In obvious preparation of Nf5.
18 ... Rb8 19 Nf5

A sacrifice so obvious that it’s not even worthy of an exclamation mark.

Question: What is so great about the move? White’s knight may be

intimidating, yet Black is by no means obliged to accept the offer.

Answer: As we all understand: if the defender is unable to chase away an entrenched

enemy piece, then this plants the seed for future combinations for the opponent.
19 ... N6d7?!
Black’s only hope was 19 ... Rb7! bolstering defences along the second rank, while
preparing to double rooks on the b-file later on. However, after 20 Bc2! (preparing to
reactivate the bishop via a4) 20 ... Rd8 21 Ba4 Rb4 22 Bc6 Black has no useful plan other
than to shuffle, while White builds for mate on the other side of the board.
 Exercise (planning): Tal missed a clear path to dismantle
Black’s kingside defences. How would you proceed here?

20 Bg5?!
Answer: Tal missed the line 20 hxg6! (step 1: open the h-file) 20 ... fxg6 21 Bxf8! (step
2: clear h6 for the knight) 21 ... Nxf8 22 Nh6+ (forcing Black’s king to a fatally awkward
post) 22 ... Kg7 23 f4! (step 3: crack open the central dark squares) 23 ... exf4 (23 ... Qb7
24 f5 is also hopeless for Black) 24 Qxf4 Qe7 25 Rf1 Rb7 26 g5! (step 4: play g5, after
which there is no good defence to the threat of Qf6+ and gxf6 mate) 26 ... Nd7?? (this
covers one mate, while allowing another) 27 Qf7+ Qxf7 28 Rxf7 mate.
20 ... Bg7!
Tal may have expected 20 ... Rb7?? 21 Nh6+ Kg7 22 hxg6 fxg6 23 f4 with a crushing
attack, or 20 ... Bf6? 21 Bxf6! Nxf6 22 Qh6 Bxf5 23 exf5! (threat: g5, followed by f6) 23 ...
Kh8 24 hxg6 fxg6 25 fxg6 Ng8 26 Qg5 Qg7 27 Rg2 intending to double rooks along the h-
file, which is decisive.
21 Nxg7
So the f5-knight’s threats were not so empty. Tal jumps on the opportunity to eliminate
Black’s best defender of his kingside dark squares.
21 ... Kxg7 22 Bh6+
The young Tal only focused on the delivery of checkmate. Actually, come to think of it,
he did the same thing in his middle and late years as well! A more positioned approach
would be 22 Bc2 Rb4 23 Kb1 a5 24 hxg6 fxg6 25 Ba4 Ba6 26 Qh2 Reb8 27 Rg2 Kg8 28
Ka1 Nb6 29 Bc6 Qg7 30 f4! Nbd7 31 f5 with a crushing bind.
22 ... Kg8 23 f4?!
Tal’s two great chess faults were lack of patience and overly optimistic assessments.
This move – an example of both negative traits – unnecessarily allows Black’s worthless
pieces use of e5. He should revert to the slower plan mentioned above with something like
23 Bc2 Re7 24 Kb1 a5 25 Ka1 Ba6 26 Ba4.
23 ... exf4!
Principle: Open the position when attacked on the wing. Black is just strategically
busted if he allows White f5.
24 Qxf4 Qd8!
Feeding another defender to the kingside. 24 ... f6? 25 Bxf8 Nxf8 26 g5 fxg5 27 hxg6!
offers White a strong attack. 27 ... gxf4?? is impossible, due to 28 gxh7+ Kf7 29 h8Q with
mate to follow.
25 hxg6
25 Qxd6 allows Black to close kingside lines for now with 25 ... g5!.
25 ... Nxg6!
Tolush defends like a shifty cockroach who skilfully avoids yet another stomp: 25 ...
fxg6? 26 Qxd6! (now there is no reason not to grab d6) 26 ... Rb6 27 Qg3 Qe7 28 Bf4 is just
a strategically lost game for Black.

26 Qh2!?
Once again, Tal isn’t tempted by the d6 bait.

Question: Why isn’t d6 hanging in this line?

Answer: Tal didn’t want to swap queens. If 26 Qxd6 Nge5! (threat: ... Rb6) 27 Na4
Rb4! 28 Bc2 (and not 28 Nxc5?? Rb6! and White’s queen is trapped) 28 ... Nd3+ 29 Bxd3
Ne5! (zwischenzug!) 30 Qxd8 Nxd3+ 31 Kc2 Rxd8 32 Rh5! Rd6 33 b3 Bd7 34 Nc3 White
still remains better due to his superior structure, but in this version Black’s forces attained
some degree of activity, and more importantly, he isn’t in danger of getting mated.
 26 ... Nde5 27 Bf4?
The bishop approaches its square with inflated expectations. If your revelation mutates,
then it isn’t a real revelation. Every once in a while Tal would struggle with conflicting
impulses and make a move which makes a natural positional player cringe. One note of
caution though: Karolyi thinks this was an inputting error in the score sheet and that Tal
actually retreated his bishop to the correct d2-square. I’m assuming the database score is
correct. If not, I owe both Tal and Tolush an apology for the question marks I handed both of
them. White is still in command after 27 Bd2.
27 ... Nf8?
The players’ last two moves have been as erratic as a steel ball within a pinball
machine. This contradictory – almost oxymoronic – idea, pulls Black in the opposite
direction of his wishes (of course, this is all under the assumption that the database score is
correct!). Now White’s attack once again becomes potent.
After 27 ... Nxf4! I’m certain Tal intended 28 Qxh7+ Kf8 29 Qh6+ Ke7! (Black’s king
runs for cover to the queenside; a move based on the philosophy: it isn’t possible to execute
a king in absentia) 30 Qxf4 Kd7! (the king, who never performed a single bold action in his
life, loves to regale the ladies of the court with his thrilling, yet completely imaginary,
escapades) 31 Bc2 Kc7 32 g5 Bd7. The advantage changed hands, since White’s kingside
attack has been stymied, while Black retains some queenside pressure, and his knight rules
e5.
28 Qh6!
Tal decides to further provision before launching his final assault. Threat: Bg5, intending
Bf6.
28 ... Neg6?
The knight zigs, when it would have been wiser zagging. Black had to try 28 ... Qb6 29
Rg2 Rb7, although 30 Bxe5! Rxe5 31 Rf1 Ng6 32 g5 Qd8 33 Rf6 a5 (33 ... Qf8 34 Qxf8+
Kxf8 35 Rxd6 is also busted for Black) 34 Rh2 wins, since if 34 ... Nf8 35 Bc2 Qe7 36
Rhf2 Ng6 37 Nd1 Qf8 (37 ... Qe8 38 Rh2 Nf8 39 Rxd6 is also completely hopeless for
Black) 38 Ne3 Qxh6 39 gxh6 Rh5 40 Rxd6 with a won ending for White.
29 Bg5! f6
Black undertakes elaborate precautions against White’s coming attack, yet they aren’t
enough. He blockades e5 with plans to later occupy the square with a piece. However, the
word “later” presupposes that he survives his current ordeal and that there will be a “later”!
 Exercise (combination alert): The following is one of Tal’s most endearing
combinations. Find one powerful move and White’s attack flares to decisive levels:

Answer: Clearance, with dual purpose:

1 The sleeping b1-bishop is granted entry into the attack.
2. e4 is cleared for White’s knight.
30 e5!!
The e5-square – the sole gatekeeper of Black’s king – is carelessly brushed aside, as Tal
shatters the blockade longitudinally. Of White’s two bishops, it was always the dark-
squared twin who was the favourite, with the b1-bishop the outcast. Now all that has
changed. Tal said that after he made his move, Tolush refused to make eye contact for the
remainder of the game.
30 ... Rxe5
Avoiding 30 ... fxg5?? (it isn’t easy to be an honest person when your position is mired
in want) 31 Bxg6 Re7 (or 31 ... hxg6 32 Qh8+ Kf7 33 Rh7+! Nxh7 34 Qxh7+ Kf8 35 Rf1+
Bf5 36 gxf5 and Black is mated) 32 Ne4!, which is decisive, since 32 ... Nxg6 allows 33
Nf6+ Kh8 34 Qxh7+ Rxh7 35 Rxh7 mate.
31 Bxg6 Rb7!
Tolush finds the only move:
a) 31 ... hxg6?? 32 Qh8+ Kf7 33 Rh7+ Ke8 34 Bxf6 Qb6 35 Na4 Qb4 36 Bxe5 Qxa4 37
Bxd6 forces mate.
b) 31 ... Nxg6?? 32 Qxh7+ Kf8 33 Qxg6 Rxg5 34 Rh8+ Ke7 35 Rh7+ Kf8 (the king’s
face is one of a man who has just been told unwelcome news) 36 Qf7 mate.
c) 31 ... Rxg5?? 32 Bxh7+ Kf7 33 Ne4 Re5 34 Bg8+!! Ke8 (34 ... Kxg8 35 g5! Kf7 36
gxf6 forces mate) 35 Nxf6+ Ke7 36 Qg7 mate. “I am not a perfect person,” confesses the
queen. Of course what she actually means is: “I have done some terrible things, and if
allowed, will continue to do more terrible things.”
32 Ne4!
Tal just flourished in such chaotic positions. The final attacker is added, and he doesn’t
bother to move either of his hanging bishops.
32 ... fxg5?!
Black offers greater resistance with 32 ... Rxg5 33 Bf5! Rgg7 34 Bxc8 Qxc8 35 Nxd6
Qd8 36 Nxb7 Rxb7 37 g5 Ng6 (37 ... f5? 38 g6! mates) 38 gxf6 Qxf6 39 Qd2 with a winning
position for White.
Instead, 32 ... Rxe4 33 Bxe4 fxg5 34 Rf1 Rg7 35 Rf6 transposes to the game’s
continuation.
33 Rf1!
Threat: Rxf8+!, as well as Nf6+.
33 ... Rxe4
A terminally ill person has no need of possessions. Black must hand over the exchange,
since 33 ... Nxg6?? 34 Nf6+ Kh8 (or 34 ... Kf7 35 Qxh7+ Kf8 36 Qg8+ Ke7 37 Rh7 mate)
35 Qxh7+! (the sight of an exceptionally beautiful woman invariably muddles your writer’s
mind) 35 ... Rxh7 36 Rxh7 is mate.
34 Bxe4 Rg7 35 Rf6!
Triple purpose:
1. Targeting d6.
2. Opening the possibility of doubling rooks on the f-file.
3. Preventing ideas like ... Ng6.
35 ... Bxg4

Exercise (combination alert): Black’s last move was an error in an

 already lost position. Tal missed a forced mate. Try and find it:

36 Rhf1?!
Answer: Annihilation of defensive barrier. Tal rejected 36 Bxh7+!! (the bishop demands
a tithe much higher than the traditional 10%) 36 ... Nxh7 37 Rg6! (this may be the move Tal
missed; White threatens g7, as well as h7) 37 ... Qe7 38 Qxh7+ Kf8 39 Qh8+ Kf7 40 Rxg7+
and mate in three moves.
36 ... Nd7 37 Rxd6 Qe7
Attacking White’s bishop ...
38 Rxa6!
... which Tal ignores, by playing on Black’s weak back rank.
38 ... Kh8
38 ... Qxe4?? walks into the back-rank mate 39 Ra8+, after which Black only has spite-
blocks.

Exercise (combination alert): White to play and force mate:

Answer: Annihilation of defensive barrier/deflection.

39 Bxh7! Nb8
Would-be defenders move in ever-widening orbits around Black’s king. 39 ... Rxh7??
allows 40 Ra8+ with mate in two moves.
40 Bf5+
Stronger is 40 Ra8! which leaves Black helpless.
40 ... Kg8
 Exercise (critical decision): Should White take Black’s
hanging bishop? If not, then what should he play?

Answer: Black’s bishop is poisoned.

41 Be6+!
41 Bxg4?? walks into Black’s trap: 41 ... Qe3+ 42 Kd1 Qd3+ 43 Ke1 Qg3+ 44 Rf2
Qg1+ 45 Rf1 Qg3+ with perpetual check.
41 ... Bxe6 42 Rxe6 1-0
The spectators gave Tal a standing ovation, for winning the USSR Championship, but
perhaps more for producing an attacking masterpiece. 42 ... Qd7 43 Qh5 Rf7 44 Rh1 Rh7 45
Re8+ forces mate in three moves. After the game, Tolush, who was Spassky’s coach, told
his student: “Today, I lost to a genius.”

Game 7
B.Gurgenidze-M.Tal
USSR Championship, Moscow 1957
Modern Benoni

1 d4 Nf6 2 c4 c5 3 d5 e6 4 Nc3 exd5 5 cxd5 d6 6 Nf3 g6 7 e4

Today, top players generally play 7 Bf4 or 7 Nd2.
 7 ... Bg7

Question: Why did Tal avoid 7 ... Bg4 which allows him to swap away
his traditional problem piece in the Benoni, his light-squared bishop?

Answer: A cautionary guideline: the trouble is 7 ... Bg4?! fails to achieve its swapping
goal after 8 Qa4+! Bd7 (8 ... Nbd7? 9 Nd2 leaves the g4-bishop dangling, with no safe
squares; 9 ... Bh5 10 h3 g5 11 e5! dxe5 12 Nde4 Nxe4 13 Nxe4 is a winning position for
White) 9 Qb3 b6 10 a4. Black’s bishop is completely misplaced on d7, since it blocks his
queenside development, I.Nikolaidis-L.Rama, Ilioupolis 1995.
Black can also play the move order 7 ... a6! 8 a4 and only now play 8 ... Bg4! which
prevents White’s annoying Qa4+ ideas.
8 Be2 0-0 9 0-0 Re8 10 Nd2 Na6
Slightly more common is 10 ... Nbd7.
11 Re1
It isn’t clear if the rook belongs on e1. More solid is 11 f3 Nc7 12 a4 Nd7 13 Kh1 b6 14
Ra3 Qe7 15 a5 when White pursues queenside ambitions, S.Mamedyarov-V.Ivanchuk,
Khanty-Mansiysk 2010.
11 ... Nc7 12 a4 b6 13 Qc2 Ng4!?
Tal’s reckless knight comes running at full speed, like a puppy who hears her name
called. It isn’t easy to creep closer to our secret goal, without inadvertent communication of
its intent. It is only because of your writer’s kindly nature that I give Tal’s last move an
interesting mark, rather than the dubious it actually deserves!
 Exercise (critical decision): Tal’s provocative last move gives
White a choice: swap away Black’s knight with 14 Bxg4, or chase it
away with 14 h3. One line gives White a strategic advantage, while
the other allows Black a strong attack. Which move should he play?

14 h3?
This is no time for vacillation. Gurgenidze walks into a vicious attack. Is Tal going to
back down and retreat his knight? Not likely!
Answer: White secures a clear strategic advantage by immediately removing the
intruder with 14 Bxg4! (a laying out of facts tends to unsettle a guilty conscience; this move
ruins Tal’s trap and turns his 13 ... Ng4?! into a dubious idea) 14 ... Bxg4 15 Nc4 when Bf4
is coming and d6 is vulnerable. Also Black’s g4-bishop just dangles on the square, not sure
just where he is going. Following 15 ... Bc8 16 Bf4 Bf8 17 Qd3! (intending Qg3) 17 ... Ba6
18 b3 White exerts unpleasant pressure on d6, while preparing an eventual e5 break.
14 ... Nxf2!!
Annihilation of king’s cover. Unlike many of Tal’s other sacrifices, this one is
completely sound. By now in the book, Tal’s motivation pattern is well established. His
motto: taunt, then engage. Did Gurgenidze really believe Tal would meekly retreat with 14
... Ne5? 15 f4 Nd7 after which White gained time, as well as the advantage?
15 Kxf2
Now White’s fatally weakened king draws his enemies near to him, as if with a
beckoning finger.
15 ... Qh4+
15 ... Bd4+ 16 Kg3 Be5+ 17 Kf2 Qh4+ 18 Kf1 Bd4 transposes to the game continuation.
16 Kf1
 No choice, since his e1-rook requires protection. 16 g3?? walks into 16 ... Bd4+ 17 Kg2
(or 17 Kf3 Qh5+ 18 Kg2 Bxh3+ 19 Kh2 Bg4+ 20 Kg2 Qh3 mate) 17 ... Qxh3+ 18 Kf3 Re5!
(the quickest path to mate; 18 ... Bg4+ 19 Kf4 g5+ 20 Kxg5 Be3+ 21 Kf6 Qh6 is also mate)
19 Nc4 Bg4+ 20 Kf4 g5 mate.
16 ... Bd4
Threatening mate on f2, which forces White’s next move.
17 Nd1

Exercise (combination alert): White’s fragile hope of remaining

alive has about the same chance as a flower, after a freshly
fallen snow storm. How would you continue Black’s attack?

Answer: Queen sacrifice.

17 ... Qxh3!
The first rule of a prosecutor is to never ask the witness a question to which he or she
doesn’t already know the answer. This is bargaining from a position of strength, since it
isn’t a real queen sacrifice. Black threatens mate again, this time on h1.
17 ... Bxh3?! lacks the same force as Tal’s move and allows White to fight on after 18
Ra3 Bf5 19 Ne3.
18 Bf3
A wounded veteran friend once told me that you don’t actually feel the pain during the
battle, since you don’t have the time to waste on it. White avoids 18 gxh3?? Bxh3 mate.
18 ... Qh2 19 Ne3 f5!
 Tal ruthlessly opens central lines, casting his f-pawn into a supporting role in the attack.
20 Ndc4
After 20 exf5?? Bxe3 and White is slaughtered.

Question: Why can’t White’s king make a run for the queenside with 20 Ke2?

Answer: He won’t make it across the hostile centre: for example,, 20 ... Bxe3! 21 Kxe3
Nxd5+ 22 Ke2 (22 Kd3?? Nb4+ forks king and queen; 22 Kf2 Qh4+ is decisive as well) 22
... Nb4 23 Qb3+ d5 24 Kd1 Ba6! (White is paralysed, with his king stuck in the middle of
the board) 25 Nf1 Qe5 26 Bd2 Bc4 27 Qc3 Qxc3 28 bxc3 Bb3+ 29 Ke2 Nc2 30 Kf2 fxe4
31 Be2 Nxa1 32 Rxa1 d4 when Black’s rook and four pawns overwhelms White’s two
minor pieces.
20 ... fxe4 21 Bxe4 Ba6!
Pinning the c4-knight, while preparing to bring out his last undeveloped piece, his a8-
rook.
22 Bf3 Re5
The human move, preparing to double rooks on the e-file, which adds further pressure to
White’s troubled e3-knight. The comp found 22 ... Rf8! 23 Ke2 Rxf3 24 Kxf3 Re8! 25 Bd2
Qh4 26 Rh1 Rf8+ 27 Ke2 Rf2+ 28 Kd3 Qf4 29 Rae1 Bxe3 forcing mate.
23 Ra3 Rae8 24 Bd2
The comp doesn’t like this move, but everything else loses as well.
 Exercise (combination alert): Continue Tal’s attack:

Answer: Pin.
24 ... Nxd5! 25 Bxd5+
And not 25 Nxd5?? Qg1 mate.
25 ... Rxd5! 26 Ke2
No time to grieve for the fallen, since if 26 Nxd5?? Qh1 mate. After the king’s murder,
the authorities consider his sister a “person of interest”, which is police speak for: “person
who committed the crime, but we don’t yet have enough evidence to arrest and convict”.
26 ... Bxe3! 27 Rxe3
 Exercise combination alert: Black attackers continue to
ransack the white king’s home. How did Tal end the game?

Answer: Removal of key defenders.

27 ... Bxc4+! 0-1
First the sermon, then comes the bishop’s favourite part: the collection plate. 28 Kd1 (or
28 Qxc4 Qxg2+ 29 Kd1 Qxd2 mate) 28 ... Rxe3 29 Rxe3 Qg1+ 30 Re1 Qf2 (threat: ... Bb3)
31 Kc1 (White’s king clings to his hope of escaping to the queenside, the way a drowning
person clings to a plank of wood in the middle of a Tsunami) 31 ... Bb3 32 Qc3 Qxe1+! 33
Bxe1 Rd1 is mate. Games like this make me want to chuck the Slav and take up the Benoni!

Game 8
M.Tal-A.Koblencs
Riga 1957
Sicilian Defence

1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 Nc6 6 Bg5 e6

The Richter-Rauzer line of the Sicilian.
7 Qd2 Be7 8 0-0-0 0-0 9 Nb3
This move sets up Bxf6 ideas. 9 f4 is more commonly played today.
9 ... Qb6
 Question: Doesn’t this move block Black’s b-pawn?

Answer: It does. As compensation, Black’s queen vacates d8 for the rook, in

preparation for a future ... d5 break. Less accurate is 9 ... Qc7 10 Bxf6 gxf6 (R.Bellin-
N.Oud, Guernsey 1989; I don’t see compensation for the pawn after 10 ... Bxf6?! 11 Qxd6
Qb6 12 Qg3) when White’s coming attack looks rather scary after 11 h4.
10 f3
This move solidifies e4, while setting up an English Attack style plan with g4 and h4 to
follow. 10 Be3 Qc7 doesn’t bother Black, since after 11 Nb5 Qb8 White is unable to
capture d6, while Black regains tempi with ... a6 and ... b5. One of Black’s ideas of placing
his queen at b6 is that 10 Bxf6? just hands Black a valuable bishop after 10 ... Bxf6 and
White is unable to grab d6, since 11 Qxd6? is met with 11 ... Qxf2, with a strategically won
game for Black.
10 ... a6 11 g4
11 Bxf6 offers Black full compensation after 11 ... Bxf6! 12 Qxd6 Rd8 13 Qc5 Rxd1+
14 Kxd1 Qc7 15 Qe3 b5 16 f4 b4 17 Na4 e5. Black gained several tempi for his attack,
owns the bishop-pair and also dominates the dark squares, which more than compensates
White’s extra pawn, V.Jansa-V.Babula, Lazne Bohdanec 1995.
11 ... Rd8
This move covers d6, which eliminates White’s Bxf6 ideas. Now White must be on
constant alert for ... d5.
12 Be3
This move actually sends Black’s queen to the square she wants to go to. In return, White
clears the path for g5.
12 ... Qc7 13 h4?!
 Inaccurate. White should take the steam out of Black’s ... d5 with the immediate 13 g5!.
13 ... b5?!
13 ... d5! takes advantage of the white queen’s unfavourable position: 14 exd5 Nxd5 15
Nxd5 Rxd5 16 Qe2 Rxd1+ 17 Qxd1, M.Hebden-N.Giffard, Paris 1988. Black’s queenside
attacking chances look more promising than White’s on the other side after 17 ... b5.
14 g5
Now we transpose back into White’s correct move order.
14 ... Nd7 15 g6!?

Tal takes trespass of his coach’s territory lightly, by offering a pawn to rapidly open
kingside lines. A more cautious route would be 15 Kb1 b4 16 Ne2 with mutual chances,
V.Anand-A.Muir, Barcelona 1993.
15 ... hxg6?!
This recapture doesn’t look traversable in full safety, and offers White huge attacking
compensation. 15 ... fxg6! follows the principle: If your king is endangered, then capture
away from the centre.16 h5 gxh5 17 Rxh5 Nf6 18 Rh1 was N.Short-A.Muir, Gibraltar 2004.
Maybe it’s stylistic, but I don’t believe in White’s full compensation for the pawn after 18 ...
Ne5.
16 h5 gxh5 17 Rxh5
So White gets open g- and h-files for the pawn – a reasonable deal for both sides.
17 ... Nf6 18 Rh1
Intending Qh2.
18 ... d5
Black follows the principle: Meet a wing attack with a central counter, yet in this case
it fails to halt White’s attack. Maybe Black should try 18 ... Ne5 19 Qh2 Ng6. Even then, I
prefer White’s chances after 20 Bd3, intending f4 and Rdg1. If 20 ... e5? White gets a
winning attack after 21 Nd5!, which wins material, since 21 ... Nxd5?? walks into 22 Qh7+
Kf8 23 Qh8+! Nxh8 24 Rxh8 mate.
19 e5?!
This not-so-artful fabrication throws away the advantage. White has a winning attack
after 19 Bf4! (intending Qh2) 19 ... Bd6 (19 ... e5 20 Qh2 Kf8 21 Qh8+ Ng8 22 Rg1 Bf6 23
Bh6! wins) 20 Bxd6 Qxd6 21 f4! when he threatens both Qh2 and also e5. Black is busted
after 21 ... Ng4 22 e5 Qc7 23 Bd3 f5 24 exf6 Nxf6 25 Rdg1 with a powerful assault to
follow.
19 ... Nxe5
19 ... Qxe5?? loses to 20 Bf4 Qf5 21 Bd3, winning heavy material.
20 Bf4?!
Correct was 20 Qh2 Kf8 21 Qh8+ Ng8 22 Bd4 f6 23 Rh7 Bd6 24 Bd3 when White may
have enough attacking compensation for the pawn.
20 ... Bd6 21 Qh2

White’s mating threat is easily controlled.

21 ... Kf8
Instead:
a) 21 ... Nd3+?? is met with the cool 22 Kb1! when, suddenly, Black is faced with both
Qh8 mate, and also Bxd3.
b) 21 ... Ng6! 22 Bxd6 Qxd6 23 Qg1 b4 when White’s kingside attack has been halted
and I see no compensation for the missing pawns.
22 Qh8+?!
This is actually a waste of time if Black finds the correct response.
22 ... Ng8?
22 ... Ke7! gives Black a winning position, since g7 is taboo. If 23 Qxg7?? (23 Qh3 is
forced, but White is losing after 23 ... b4 24 Ne2 a5 when Black has both the extra pawn and
the stronger attack) 23 ... Rg8 24 Qh6 Nd3+! wins.
 23 Rh7
Now Tal gets some compensation for the pawn, since he forces a loosening of Black’s
king’s cover.
23 ... f5
Black threatens both ... Ng6 and also ... Nd3+, which picks off an important white
attacker.
24 Bh6!
Tal eliminates both threats by taking advantage of two pins to increase the pressure on
g7.
24 ... Rd7

Exercise (combination alert): It appears as if Black is winning. After all, he

threatens both ... Ng6 and also ... gxh6. How did Tal keep his attack alive?

Answer: Clearance/double attack.

25 Bxb5!!
Long suffering material poverty breeds criminality. White threatens both Bxd7 and also
Rg1. After Tal’s last move, the impression that Black stands better is dispelled in a single
stroke.
25 ... Rf7?
Koblencs, who decides to deal with the problem with a sideways crab approach, goes
astray. His head must have been spinning about now. Other options were:
a) 25 ... axb5? is one of those presents which look wonderful when wrapped, and not so
great when opened: 26 Rg1! Ng4 27 Nxb5 Bf4+ 28 Kb1 Qe5 29 Bxf4 Qxf4 30 fxg4 fxg4 31
Nd6!! Re7 32 Rgh1! (threat: Qxh8+!) 32 ... Qxd6 33 Rf1+ Ke8 34 Qxg8+ Kd7 35 Rf8 Kc7
36 Rh8 Qa6 37 Rd8 g3 38 Qf8 Rd7 39 Qc5+ Kb7 40 Qb4+ Kc7 41 Qc3+ Qc4 42 Qxg3+
Kb7 43 Qf2 and White has a winning attack, despite the reduced material.
b) 25 ... Ng6! is Black’s only move: 26 Nd4! (threatening e6) 26 ... Re7 27 Bxg7+! Rxg7
28 Bd7!! (interference) 28 ... Qxd7 29 Nxe6+! (overloaded defender( 29 ... Qxe6 30 Qxg7+
Ke8 31 Nxd5 N8e7 32 Nf6+ Kd8 33 Qf7! Qe5 (or 33 ... Qxf7? 34 Rxf7 Kc7 35 Ne8+ and
White is winning the ending if he picks up another piece) 34 Kb1 Rb8 35 b3 Rb7 with a
dead even assessment by Komodo here, which for us humans, translates to completely
unclear.
26 Rg1!
Adding a fourth attacker to g7, while preventing ... gxh6 and ... Ng6.
26 ... Ra7 27 Nd4

Tal threatens Bxa6!, when he could have played the move without preparation: 27
Bxa6!!. Only a comp can find such a move. Now if 27 ... Bxa6 (27 ... Rxa6?? is crushed by
28 Rgxg7) 28 Nd4! (going after e6) 28 ... Bc8 29 Ndb5! Qd7 30 Nxd6 and Black can’t
afford to abandon control over g7, since 30 ... Qxd6 loses to 31 Bxg7+.
27 ... Ng4!?

Question: This interference looks overly desperate.

Does Black have a calmer defensive idea?

Answer: I agree. When we bribe our opponent with concessions, we may be

inadvertently funding our own future assassination. Black’s best defence may lie in 27 ...
Bc5! 28 Rgxg7 Rxg7 29 Rxg7 Qxg7 30 Bxg7+ Rxg7 31 Qh4 Nf7 32 Bd3 Ne7. Black has a
rook and minor piece for the queen – generally considered to be not enough – yet he may
survive since his position is compact and weakness-free.
28 fxg4 Be5?
This logical defence should lose. Correct was 28 ... Bf4+! 29 Kb1! (or 29 Bxf4 Qxf4+
30 Kb1 Qxd4 31 Ne2 Qd2 32 Rgh1 Rf6! 33 Bd3 e5 and Black is surviving the attack) 29 ...
Bxh6 30 g5 Bxg5 31 Rxg5 Qf4 32 Rgh5 (threatening the Qxg8+! cheapo again) 32 ... Rfc7
33 Rh4 Qf2 34 a4! axb5 35 Ndxb5 Rxc3 36 bxc3! Qb6 37 Rb4 Re7 when Black has some –
but probably not enough – compensation for the exchange.
However, 28 ... gxh6?? loses to the stunning interference shot 29 Bd7!! Qxd7 30 gxf5
Bf4+ 31 Kb1 Bg5 32 Nxe6+ and the point is Black no longer may capture with his c8-
bishop. White wins.

Exercise (critical decision): White’s candidate moves are 29 Nc6, or 29 Nf3.

Only one of them is immediately decisive. Which one would you play?

29 Nc6?!
Answer: Tal misses the stronger 29 Nf3!! axb5 30 Nh4!.The invasion to g6 is decisive.
29 ... Bxc3?
29 ... axb5! is Black’s final chance to remain alive, and if 30 Nxe5 Qxe5 31 Rf1!. Even
here White has a wicked attack. Black is unable to close lines with 31 ... f4? since it is met
with the shot 32 Rxg7! Qe3+ 33 Kb1 Rxg7 34 Rxf4+ Ke8 35 Rf8+! Kxf8 36 Bxe3, winning.
30 Be3!!
 Tal was the master manipulator who knew how to throw opponents out of synch, by
deftly pitting one piece against another. Threat: Bc5+ and Qxg8+. Tal’s elegant solution is
even stronger than 30 bxc3 which also wins.
30 ... d4 31 Rgh1!
Ah yes, the familiar cheapo Qxg8+! arises again. White’s mating threats feels like the
only tidy corner of an otherwise messy room.
31 ... Rd7
Vacating f7 for his king and eliminating the Rxg8+ threat.

Exercise (critical decision): A single move separates Tal from glory or

obscurity. Our candidate moves: 32 Bg5, 32 g5 and 32 gxf5. Two of them
 win, while the other allows Black’s resurrection. Which one would you play?

32 Bg5??
Tal complicated to confuse his opponents. Sometimes, however, he was the inadvertent
victim of his own complications. On the chess board it is very easy for any one of us to slip
into a kind of psychotic episode, where we completely lose touch with the position’s reality.
Chess writers politely call it “losing the thread of the game”.
Answer: 32 gxf5! wins easily. If 32 ... exf5 33 Qxg8+! Kxg8 34 Bc4+ forces mate.
Answer no.2: 32 g5! is also crushing after 32 ... dxe3 33 Qxg8+! (yet another version of
a now almost tired theme) 33 ... Kxg8 34 g6! (threatening a back-rank mate) 34 ... Rd1+ (the
only move) 35 Kxd1 Qd6+ 36 Bd3 e2+ 37 Kxe2 Qe5+ 38 Nxe5 Kf8 39 Nc6 Ke8 40 Nxa7.
Black’s king escaped mate, at the cost of an overwhelming material deficit.
32 ... axb5 33 R1h6!
Dirty tricks with Rf6+ are in the air.

Exercise (critical decision): Our hearts beat oppressively when we

land in such indecipherable situations. Black’s defensive choices:
33 ... Qxc6, 33 ... d3 and 33 ... Bxb2+. Two of the three save him. Take your pick.

33 ... d3??
Having a save, missing it, and losing, is like a death row inmate who is exonerated
posthumously, a few minutes after the administration of lethal injection. Black finds the path
to the loss.
Answer: 33 ... Bxb2+! 34 Kxb2 Qxc6 and Black threatens mate, starting with..Rxa2+!,
which forces 35 Rf6+! gxf6 36 Bh6+ Rg7 37 Bxg7+ Ke7 38 Bxf6+ Kd6 39 Rxa7 Qc3+ 40
Kb1 Qe1+ with perpetual check.
 Answer no.2: 33 ... Qxc6! is also good enough for a draw after 34 Rf6+! gxf6 35 Bh6+
Rg7 36 Bxg7+ Ke7 37 Bxf6+ Kd6 38 Be5+ Kd5 (the king postures and poses to the hand-
clapping delight of the ladies of the court) 39 Rxa7 Bxb2+! 40 Kxb2 (40 Kb1 Qc3 41 Bf4
Qe1+ 42 Kxb2 Qc3+ is drawn as well) 40 ... Qc3+ 41 Kc1 Qe1+ 42 Kb2 Qc3+ with
perpetual check.
Instead, if 33 ... gxh6?? 34 Bxh6+ Rg7 35 bxc3 wins.
34 bxc3 d2+ 35 Kd1! Qxc6
Black’s threat of mate in one move is too slow.

Exercise (combination alert): We feel the black king’s Hamlet-like

grief and angst. Although White carries a burden of a considerable
deficit in the material side of the ledger, he can force the win. How?

Answer: Clearance.
36 Rf6+!
The parasite makes yet new demands upon the long suffering host.
36 ... Rf7
36 ... gxf6 loses to 37 Bh6+ Rg7 38 Bxg7+ Ke7 39 Bxf6+ Kd6 40 Be5+ Kd5 41 Rxa7,
while if 36 ... Ke7 37 Rxg7+ Kd6 38 Rxe6+! Kxe6 39 Rg6+ Kd5 40 Qd4 mate.
37 Qxg7+ 1-0
The queen frowns in counterfeit concern over her brother’s declining health. After 37 ...
Ke8 38 Rxf7 (the beggarly pittance of defenders is insufficient to save Black’s king from
checkmate) 38 ... Qf3+ 39 Kxd2 Qf2+ 40 Kc1 Qf1+ 41 Kb2 Black’s checks run out (if you
don’t count spite-checks!).

Game 9
L.Aronson-M.Tal
USSR Championship, Moscow 1957
Dutch Defence

1 d4 e6 2 c4 f5

Question: Tal seems to jump from opening to opening.

Isn’t this a mistake, since it’s impossible to master everything?

Answer: Tal, unlike Fischer, wasn’t one of those players who played the same opening
every game. There is a certain kind of player (your writer included) who loves the
adventure of moving from one opening to another to keep his thought process from growing
too mechanical, which I think compensates. If you look at the top 10 players in the world
today, you see that the majority play multiple opening systems, rather than the same lines
each time.
3 Nf3 Nf6 4 Nc3 Be7 5 g3 0-0 6 Bg2 d6
Tal opts for the gnarled Classical Dutch set-up. Now I don’t fully trust this set-up for
Black (although a few of my students play it with success), mainly since I wrote a book on
Bird’s Opening (1 f4), and experienced difficulty equalizing for White, in the chapter on the
Reversed Classical. So I certainly wouldn’t be tempted to play this line as Black, a tempo
down. To us, all butterflies look the same, while a lepidopterist notices species, shadings
and markings, invisible to the lay person. My Classical Dutch-loving students vigorously
defend their beloved line, and claim that Black gets hidden counterplay (seen only by their
discerning eyes!), despite the appearance of inferiority. This game may be a perfect example
of Black’s hidden potential in the Classical Dutch.
7 0-0 Qe8
 Question: What is the idea behind ... Qe8?

Answer: Two ideas behind ... Qe8:

1. Black tries to implement ... e5, with ... Bd8 and ... e5.
2. Black plays for an attack with ... Qh5.
8 Re1
White plays directly for an e4 break, rendering Black’s e6-pawn backward after ... fxe4.
White’s alternatives:
a) 8 b3 a5 9 Bb2 Na6 10 Re1 Qg6 11 e4 Nxe4 12 Nxe4 fxe4 13 Rxe4 with a pleasant
edge for White, V.Chuchelov-A.Spice, Clichy 1995. Note that White’s rook isn’t actually
hanging, since 13 ... Qxe4? self-traps the queen after 14 Nh4 Qxh4 15 gxh4 Bxh4 16 Qh5
Bxf2+ 17 Kh1. Black doesn’t get enough for the queen, since White controls the initiative.
b) 8 Qc2 Na6 (8 ... Qh5, 8 ... Nc6 and 8 ... Qg6 are also played here) 9 a3 Qg6 10 b4 c6
11 c5! Nc7 12 Bf4 Nce8 13 b5! and White presses with his queenside initiative,
V.Malakhov-S.Volkov, Khanty-Mansiysk 2007.
8 ... Qg6
Suppressing White’s e4 break ...
9 e4!

... or maybe not!

9 ... fxe4 10 Nxe4

Question: Can White try 10 Ng5?

Answer: Your suggestion is virtually a Theoretical Novelty, since I see only two games
with this move in my database. White looks better after 10 ... Nc6! (Black avoids White’s
devious trap 10 ... d5? which fails to 11 cxd5 exd5 12 Nxd5! Nxd5 13 Bxe4 Qd6 14 Bxh7+
Kh8 15 Qh5 Nf6 16 Qh4 Ng4 17 Bd3+ Nh6 18 Bf4 Rxf4 19 gxf4 and White threatens Nf7+;
if 19 ... Qf6 20 Qh5 Nc6 21 Qe8+ Bf8 22 Nh7 with a winning position for White) 11 Ngxe4
e5 12 dxe5 dxe5 (12 ... Nxe5?! 13 f4 leaves White in control) 13 Nxf6+ Bxf6, A.Dziedzic-
H.Faber, correspondence 2005. White stands a shade better after 14 Nb5.
10 ... Nxe4 11 Rxe4 Nc6
11 ... Qxe4? 12 Nh4 trapping the queen is similar to note ‘a’ from above, but even worse
for Black, since after 12 ... Qxh4 13 gxh4 Bxh4 in this version White doesn’t even lose his f-
pawn and can play 14 Be3 when Black doesn’t get enough for the queen.
12 Qe2 Bf6
Black hopes to engineer ... e5.
13 Bd2
Getting ready for Bc3, after which Black is unlikely to achieve the freeing ... e5 push.
13 ... e5!?
If he doesn’t play this freeing break now, it’s unlikely he can achieve it later. 13 ... d5?!
weakens and fails to bother White’s rook, who can slide over a square with 14 Rf4.
14 dxe5 dxe5 15 Bc3
Covering against ... Nd4 tricks.
15 ... Bf5
This would be great if not for White’s next move.
16 Nh4! Bxh4 17 Rxh4

Question: Isn’t Black just strategically lost? After all, White owns the
bishop-pair in an open position, and also may later pick up the e5 isolani.

Answer: I like White’s position, but don’t think Black is losing. The battle isn’t as one-
sided as it first appears. The following factors offer Tal’s side counterplay:
1. White’s rook, although ominously posted, is also out of play if it fails to help generate
an attack upon Black’s king. Rooks are notoriously awkward when lifted into a crowded
middlegame. In this case there is no simple route for the stranded rook back to home base.
2. The e5-pawn may be a future weakness, but for now, White must keep close watch for
... Nd4 tricks.
3. Black is fully developed, with active pieces and an open f-file.
17 ... Rae8 18 Qe3

Question: Why did White move his queen to e3?

Answer: Two reasons:

1. White once again covers against ... Nd4.
2. White sets up the strategic threat b5, b5 and Qxa7.
18 ... h6 19 b4!
Going forward with his plan.
19 ... Qf6 20 b5 Nd8 21 Bd5+
Also strong is 21 c5 c6 22 Ra4 cxb5 23 Rxa7 Rf7 24 Ra8 Rd7 25 Re1 Rd3 26 Qc1 Bg4
27 f4 and White earned himself a winning position, Y.Yakovich-S.Dyachkov, Maikop 1998.
21 ... Kh8

22 f4?!
This doesn’t go very far in satisfying White’s desires. With this natural move, White
unwittingly enters a Faustian contract, where his own king becomes more exposed than
Black’s.

Question: It seems a7 hangs. Why didn’t White take it?

 Answer: There are two kinds of players: the majority who play for the initiative, and
then there is the greedy minority who grab material, even at the cost of relinquishing the
initiative to the opponent. I’m embarrassed to admit that your writer belongs in the latter
group. 22 Qxa7!? can be met with 22 ... g5 23 Rd4! c6 24 Bg2 exd4 25 Bxd4 Re5 26 f4 gxf4
27 gxf4 Nf7 28 fxe5 Qg6 29 Qa3 (29 e6+ Kh7 30 exf7?? is met with 30 ... Be4!, after which
White’s king is toast) 29 ... Rg8 30 Qf3 Ng5 and the comps like White. In real life, however,
Black gets practical chances against White’s exposed king.
White still holds the advantage with the tricky line 22 Re1! g5 23 f4!! gxh4 24 fxe5 Qg5
25 e6+ Kh7 26 e7 Rg8 27 exd8Q Qxe3+ 28 Rxe3 Rxd8 29 Re7+ Kg6 30 Rxc7 hxg3 31
Rxb7 Rxd5 32 cxd5 gxh2+ 33 Kxh2, which leaves Black fighting for the draw.
22 ... exf4 23 Qd2! Qb6+ 24 Bd4 Qg6!
With this move Tal sets up his coming idea.
25 Qxf4
Threat: Rxh6+.
25 ... Kh7 26 Qxc7?
White should go for 26 Qd2 c6 27 bxc6 bxc6 28 Bg2 Ne6 29 Bxa7 Ng5 when Black gets
light-square play for the pawn.

Exercise (combination alert): As the conditions of a position change, so do the

requirements. By grabbing the c7-pawn, White used a powerful tool for an insignificant
purpose. He smells sulphurous fumes, yet ignores the impending explosion. The exultation
White feels proves to be both illusory and temporary.
We sense Tal’s hidden agenda (well, not so hidden, since his agenda is his opponent’s
king about 99% of the time!), yet it isn’t easy to identify, in its misty insubstantiality. How
did Tal seize the advantage after White’s last move?
 Answer: Interference/weak back rank.
26 ... Bb1!!
The hunter’s natural exhilaration of pursuit reverses itself upon White, who now
unexpectedly becomes the hunted. “God helps those who help themselves,” says the bishop,
who hopes that his great self-sacrifice will one day lead to his canonization. This move
renders White’s a1-rook useless, while weakening his back rank, which in turn endangers
his king.
27 Be5?!
White can still minimize the damage by bringing his h4-rook back into the game with 27
Rf4! Re1+ 28 Kf2 Rfe8 29 Qxg7+! Qxg7 30 Bxg7 Kxg7 31 Bf3, with better chances to hold
than in the game.
27 ... Ne6!
We sense felonious intent emanating from the knight. Tal confronts White with an
unpleasant choice: he can either chop the knight, which decisively weakens his already
rotting light squares, or he can do nothing and allow the knight entry to g5, after which
White’s king is unlikely to survive. Also strong was 27 ... Nf7! 28 Bxf7 Rxf7 29 Qd6 Qxd6!
30 Bxd6 Re1+ 31 Kg2 g5 32 Rd4 Be4+ 33 Rxe4 Rxe4. Black wins the exchange while
maintaining the initiative. He threatens the c4-pawn, as well as ... Re2+.
28 Qd6
This queen is the meddling mother whose ‘help’ only amplifies her children’s problems.
Nothing else saved White either:
a) 28 Qa5 Ng5 29 Qb4 Bd3 30 Qb2 Rf5 and White is busted. If 31 Bd4?? Rxd5! 32
cxd5 Nf3+ 33 Kf2 Nxh4 34 gxh4 Qf5+ 35 Kg1 Qg4+ forces mate.
b) 28 Qxb7 Ng5 29 Qc7 (29 Qxg7+ Qxg7 30 Bxg7 Re1+ 31 Kg2 Kxg7 is also lost for
White) 29 ... Nh3+! 30 Rxh3 Rxe5! 31 Qxe5 Qb6+! (attackers seething with resentment over
past wrongs, begin a cruelly punitive counterattack) 32 Kg2 Qf2+ 33 Kh1 Qf1 mate.
c) 28 Bxe6 Rxe6 29 Rd4 Rf7 30 Qb8 Qc2 and White is mated.
28 ... Qf5
Threatening mate on the move, while enabling his knight to enter g5 without allowing a
queen swap.
29 Bf4 Ng5
Whenever we look at a litter of puppies or kittens, we get a sixth sense feeling of which
ones are destined to be trouble makers. This knight gives us just such an uneasy feeling.
White is faced with unsolvable threats along the light squares.
30 Qb4
Attacking Black’s troublesome bishop, who decides that moving away from b1 is not
such a big imposition.
30 ... Be4!
Tal eliminates White’s sole defender of his light squares, after which the breach goes
unrepaired.
31 Bxe4
 31 Rf1 Nh3+ 32 Rxh3 Qxh3 33 Bxe4+ Rxe4 34 Qb1 Qf5 35 Bb8 Qxf1+ 36 Qxf1 Rxf1+
37 Kxf1 Rxc4 38 Bxa7 Ra4 is a completely lost ending for White.
31 ... Rxe4 32 Rf1
32 Bxg5?? allows mate after 32 ... Qf2+ 33 Kh1 Re2 34 Rg1 Qf3+.
32 ... Re2 33 Qd6 Rxa2
Threat: ... Qc2 and stronger than 33 ... Qe4, which is met with 34 Qd5 Rg2+ 35 Kh1
Rxa2+ 36 Qxe4+ Nxe4 37 Kg1 g5 38 Rxh6+ Kxh6 39 Bxg5+ Kxg5 40 Rxf8 Nd6 and Black
wins.
34 Qd5 Qc2 35 c5

Exercise (combination alert): There are three different

paths to Black’s win. You only need to find one of them:

Answer: Deflection.
35 ... Rd8!
Tal, of course, is attracted to the flashiest move. Also easily winning are the lines:
a) 35 ... Ne4! 36 Rxh6+ (White deals with both material deficit and king’s safety in
descending order of priorities) 36 ... gxh6 (a free rook tends to be a substantial contribution
to our net worth) 37 Qxb7+ Kg8 38 Qd5+ Rf7 39 Qd8+ Kh7 and White’s checks run out.
b) 35 ... Rf5 36 Qxb7 Rxf4! 37 gxf4 (or 37 Rfxf4 Qxc5+ 38 Kh1 Qc1+ 39 Rf1 Qxf1
mate) 37 ... Ne4! and once again, the interference trick is deadly for White.
36 Bd6
36 Qxd8 isn’t so tempting, since 36 ... Qg2 is mate.
36 ... Re8! 0-1
37 Qxb7 Re1! forces mate.
Game 10
M.Tal-O.Panno
Portoroz Interzonal 1958
Ruy Lopez

When this game was played Tal led the Interzonal tournament by a half-point from Tigran
Petrosian. A normal person plays it safe in such a situation. Tal, as we know, was anything
but normal, and took appalling risks against then reigning world junior champion, yet
miraculously, he didn’t fall off the cliff’s edge.

1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7 Bb3 d6 8 c3 0-0 9 h3
Nd7
Karpov’s Variation of the Ruy Lopez. 9 ... Bb7 is Zaitsev’s (who was Karpov’s coach)
line.
10 d4 Nb6 11 Be3
Alternatively, 11 Nbd2 Bf6 12 Nf1 Re8 13 N1h2 exd4 14 cxd4 Na5 15 Bc2 c5 16 Ng4
Bxg4 17 hxg4 cxd4 18 g5 Be5 19 Nxd4 g6 when Black’s activity may fully compensate for
White’s bishop-pair, V.Topalov-L.Fressinet, Cannes (rapid) 2002.

11 ... exd4!?

Question: Why did Black voluntarily give up control over the centre?

Answer: Black plays for imbalance and plans to follow with ... Na5 and ... c5, hoping
for a kind of Benoni structure. Alternatives:
a) 11 ... Bf6 12 Nbd2 Na5 13 Bc2 Nac4 14 Nxc4 Nxc4 15 Bc1 Re8 16 b3 Nb6 and
Black came out of the opening with a decent position, since the swap of knights benefits his
side, I.Boleslavsky-S.Flohr, Groningen 1946.
b) 11 ... Rb8 12 dxe5 Nxe5 13 Nxe5 dxe5 14 Qh5 Qd6 15 Nd2 Bf6 16 Nf3 Nc4 17 Rad1
Qe7 18 Bg5 g6 19 Bxf6 Qxf6 20 Qh6 Qe7 (most of Tal’s opponents were rightfully paranoid
when they played him; there was nothing wrong with 20 ... Nxb2 21 Rd2 Nc4 22 Bxc4 bxc4
23 Rd5 Rb5 when Black stands at least even) 21 Rd3 f6 22 Nh4 g5? (Black is still okay
after 22 ... Be6) 23 Nf5 Bxf5 24 exf5 Kh8 25 Red1 Rfd8 26 Rxd8+ Rxd8 27 Rxd8+ Qxd8
28 h4 g4? 29 Bxc4 bxc4 30 Kh2! Kg8 31 Qh5 Qd2 32 Qxg4+ Kf8 33 f3 Qxb2 34 Qxc4 Qf2
35 Kh3 1-0, M.Tal-K.Darga, Munich Olympiad 1958.
12 cxd4 Na5!
A new move at the time, intending ... c5 next, which must have come as a surprise for
Tal, who probably expected 12 ... d5?!, which allowed White to take over the advantage
after 13 Nc3 dxe4 14 Nxe4 Bf5 15 d5 Na5? 16 d6! cxd6 17 Bxb6 Qxb6 18 Ng3 Be6 19
Bxe6 fxe6 20 Rxe6 Bf6 21 Rxd6 Qb8, M.Tal-V.Antoshin, USSR Championship, Moscow
1957. At this point White can secure a winning position with 22 Ne4!, and if 22 ... Bxb2 23
Neg5! Bxa1 24 Qb1! winning, since 24 ... g6?? 25 Rxg6+! Bg7 26 Rb6! simultaneously
threatens Black’s king and queen.
13 Bc2 c5 14 e5!?

Question: Isn’t Tal basically undermining his own centre with this last move?

Answer: Kasparov gave the move a dubious mark, while I almost awarded it an exclam
since it forced Panno out of his home preparation – even though objectively the move isn’t
that great for White. Tal simply decided to meet a surprise with a surprise. “There you have
it, Tal’s manner of play: who else could have made such an audacious, provocative move?”
asks Kasparov. Tal’s move gained immediate dividends on the clock, since Panno thought
for over an hour before responding. White has a better chance at an edge with the calmer 14
Nc3, A.Zaitsev-V.Lepeshkin, Moscow 1963.
14 ... dxe5!
The only path to equality. Panno correctly avoids 14 ... cxd4?! 15 Bxd4. This position is
full of danger for Black: for example, 15 ... Nac4 16 Nc3! and if 16 ... Nxb2?? 17 Bxb6
Qxb6 18 Nd5 Qd8 19 Qd4! Nc4 20 Qe4! g6 21 Nxe7+ Qxe7 22 Qxa8 Bf5 23 Qxa6 Bxc2 24
Qxb5 with a hopeless position for Black.
15 Nxe5 Nbc4 16 Qd3
Threatening mate. Of course, Tal isn’t interested in a mere draw after 16 Qh5 g6 17
Nxg6 fxg6 18 Bxg6 hxg6 19 Qxg6+, with perpetual check.
16 ... f5 17 Bb3!? f4!?
Panno chases away the defender of d4 with tempo. Black’s game also looks good after
17 ... Nxb3 18 Qxb3 (18 axb3?? hangs a piece to 18 ... Nxe5) 18 ... cxd4 19 Bxd4 Qxd4 20
Nc6 Qxb2 21 Nxe7+ Kh8 22 Nc3 Ra7 23 Rad1 Rd7. The comp claims White gets full
compensation for the pawn here. I’m not convinced and would take Black if given a choice.

18 Bd2!

Question: Can White win a pawn with 18 Nxc4?

Answer: The line works out in Black’s favour after 18 ... Nxb3! 19 Qxb3 Be6! 20 Bxf4
Bxc4 21 Qe3 Bf6 22 Be5 cxd4, and Black won an important pawn.
18 ... Nxb3!
After 18 ... cxd4? 19 Bxa5 Qxa5 20 Nd2 White wins material.
19 Nc6!?
 “And I’m floating in a most peculiar way, and the stars look very different today.” An
emotional lid is removed, where all forms of restraint are emptied. Kasparov once again
gives this move a dubious mark.

Question: Why, oh why, was Tal attracted to ideas which contained

a decidedly high potential for painful consequences? His last move
looks unsound. Doesn’t Black get enormous compensation if he plays ... Nxa1?

Answer: First, scolding Tal for playing unsoundly, is a bit like yelling at the checkout
clerk at the grocery store, because the price of milk went up. Secondly, the findings can’t be
accurately described as definitive. I put this position on Komodo for about 15 minutes and it
assessed at ‘+0.32’ – a mild ‘+=’ position! I actually disagree with the comp’s assessment
and think Black achieves dynamic equality, but perhaps no more than that.
We are all familiar with the principle of the carrot and the stick. In this case, I feel like
Tal applied the former, offering his young opponent almost unfathomable complications,
rather than 19 Qxb3 Bf6 20 Na3 Qd5 21 Nexc4 bxc4 22 Qxc4 Qxc4 23 Nxc4 Bxd4 24 Bc3
with approximate equality.
19 ... Nxa1!
I think this is Black’s best move. Komodo points out the bizarre alternative 19 ... Nxd4!?
20 Nxd8 Bf5 21 Qf1 Bxd8 22 Bc3 Nc2 23 b3 Nb6 24 Re5 Nxa1 25 Bxa1 Nd7 26 Re1. I’m
the greedy type, and still prefer White’s extra material.
20 Nxd8 Bf5 21 Qf3 Raxd8
21 ... Rfxd8?? is a case of the wrong rook. White wins after 22 Rxe7 Bxb1 23 dxc5!,
and if 23 ... Nxd2 24 Qc3! forces mate.
22 Rxe7 Bxb1 23 Bxf4 Rxd4
Panno has faith in his piece activity. Black can also take on a passed d-pawn with 23 ...
cxd4 24 b3 Nd6 25 Qd1 Rxf4 26 Qxb1 Nf5 27 Re1 d3 28 Qxa1 Nd4 29 Qb2 d2 30 Rd1
Ne2+ 31 Kh2 (or 31 Kf1? Ng3+ 32 Kg1 Re4! 33 Qa1 Ne2+ 34 Kf1 Rc8! 35 f3 Rc1 36 fxe4
Rxa1 37 Kxe2 Rxa2 and Black stands clearly better, since 38 Rxd2?? is a lost king and
pawn ending after 38 ... Rxd2+ 39 Kxd2 a5 40 Kc3 Kf7 41 Kd4 Ke6 42 Kc5 a4 43 bxa4
bxa4 44 Kb4 Ke5, which leaves White’s king too far away) 31 ... Rfd4 (Black threatens ...
b4 and ... Nc3) 32 b4 Rd3 33 f3 Nc3 34 Rxd2 Rxd2 35 Qxc3 Re2 36 Qc6 Rdd2 37 Qc8+
Kf7 38 Qf5+ with perpetual check.
24 Qg4?
After this move Black should be winning. Correct was 24 g3! threatening Qb7, and then:
a) 24 ... Be4 25 Qg4! (25 Rxe4? walks into Black’s trap after 25 ... Nd2! 26 Qa3 Nxe4
when White is busted) 25 ... Bg6 26 b3 Nd2 27 Qe2 Ne4 28 Be5 Rxf2 29 Qg4 Rf1+! 30
Kxf1 Nd2+ 31 Ke2 Rxg4 32 hxg4 Ndxb3 33 axb3 Nxb3 34 Bxg7 with an exceedingly
difficult position to assess. I suspect that White can halt all three black queenside passers
and is the one with the real winning chances, due to Black’s cut off king.
b) 24 ... Nd6 25 Qg4 Nf7 26 Re1 Bd3 27 Rxa1 g5 28 Rd1 c4 29 Re1 h5 30 Qxh5 gxf4
31 Qg4+ Kh7 32 Qh5+ Kg7 33 Qg4+ with perpetual check.
24 ... Bg6
Black now gets too much for the queen.
25 Qe6+ Bf7 26 Qf5 Nc2!
Back into the game. Panno exploits the fact that White’s bishop would hang, if the queen
took the c2-knight.
27 b3

Exercise (critical decision): Black’s candidate moves are 27 ... Bg6

and 27 ... Rd1. One line leads to a clear advantage for Black, while
the other allows White to escape. Which one would you play?
 27 ... Bg6?
The wrong one.
Answer: Black consolidates with 27 ... Rd1+! 28 Kh2 Nd2! (threat: ... Nf1+, inducing
White’s king into discovered check) 29 Bxd2 Rxd2 30 Qxc5 h6 31 Qb6 Nb4 32 Qa5 Rd4
and Black’s two minor pieces and rook are worth more than White’s queen.
28 Rxg7+!
Panno either overlooked or underestimated this shot. Tal correctly rejects 28 Qe6+? Bf7
29 Qf5 and now Black can revert to the superior line with 29 ... Rd1+! 30 Kh2 Nd2!.
28 ... Kxg7 29 Bh6+!
Attraction/overloaded defender.
29 ... Kxh6 30 Qxf8+ Kg5 31 bxc4 bxc4!

Question: Isn’t White just losing? Surely Black’s rook and minor
pieces, plus two passed c-pawns must weigh the game in his favour?

Answer: You underestimate the x-factor, which is Black’s unsafe king. Normally, a lone
queen isn’t enough to bother the enemy king. In this instance, the queen is assisted by her 3:1
kingside pawn majority, which ensures that Black’s king remains insecure. White may
actually stand slightly better in this position, due to this factor.
32 g3?!
When our ‘winning’ idea misfires, it’s as if our bolt of lightning congeals in mid-air.
Some plans require a shotgun approach, spraying buckshot anywhere in the general vicinity
of the target, while others require the pinpoint accuracy of a marksman or markswoman.
This position is a case of the latter.
Dvoretsky suggested 32 f3!! which is also Komodo’s first choice. Further analysis
shows that this is White’s strongest plan. If 32 ... Bd3 (or 32 ... Bf5 33 Kf2 Kg6 34 g4 Rd2+
35 Kg3 and if Black plays 35 ... Bd3? 36 h4 h6 37 Qe7! h5 38 gxh5+ Kf5 39 Qg5+ his rook
falls) 33 h4+ Kg6 34 g4 (now it becomes clear that Black’s king is in serious danger) 34 ...
h5 (34 ... c3?? 35 h5+ Kg5 36 Kh2! Ne3 37 Kg3 Nf1+ 38 Kh3 c2 39 Qxc5+ Kh6 40 Kh4!
threatens mate on g5, and 40 ... Kg7 41 Qe7+ Kg8 42 h6 leaves Black helpless to prevent
mate on g7) 35 Qe8+ Kg7 36 Qe7+ (there is no rush to take the h5-pawn) 36 ... Kg6 (the
king’s hair looks greyer now than it did a few moves ago) 37 g5! c3 38 Qf6+ Kh7 39 Qf7+
Kh8 40 Qxh5+ Kg7 41 Qe8 Nb4 42 h5 c2 43 h6+ Kh7 44 Qf7+ Kh8 45 Qg7 mate.
32 ... Be4!
Threat: ... Rd1+ and ... Rh1 mate Now Black should be okay.
33 h4+!?
Tal still wants to win, rejecting perpetual check after either:
a) 33 Qg7+ Kf5 34 Qf7+ Ke5 35 Qe7+ Kd5 36 Qd7+ Ke5 37 Qe7+.
b) 33 f4+ Kg6 34 g4 Rd1+ 35 Kh2 Nd4 36 Qe8+ Kg7 37 Qe7+ Kg8 38 Qxe4 c3 39
Qd5+ Kg7 and White must take perpetual check, otherwise ... c2 and ... c1Q follows.
33 ... Kg4!?
Out in the open tends to be an unfavourable locale for surreptitious actions. Our
emotional debris can nudge us into inexplicably irrational choices at the board. Maybe
Panno begins to play for a win himself. Probably wiser is 33 ... Kg6 34 g4 Rd1+ 35 Kh2
Nd4! and White has nothing better than to take perpetual check.
34 Kh2!
Threat: Qf4+, followed by Qg5 mate.
34 ... Bf5!
Panno boldly points the finger of culpability to Tal’s weakened light squares, avoiding
the trap 34 ... Kh5?? 35 Qf6 Rd5 36 Qf7+! Kg4 (36 ... Kh6 37 g4! forces mate) 37 Qf4+
Kh5 38 Qxe4 and wins.
35 Qf6!
Tal continually generates threats, which in the long run exhausts his opponent.
35 ... h6!
35 ... a5?? hangs material to 36 Qg5+ Kf3 37 Qxf5+.
 36 Qe5!?

Question: Doesn’t 36 f3+ win a piece?

Answer: It wins a piece, but not the game, after 36 ... Kxf3 37 Qxf5+ Ke3 38 g4 (not 38
Qxc2?? Rd2+ and White loses the king and pawn ending) 38 ... c3 39 g5 hxg5 40 hxg5 Kd2
41 Qxc5 (or 41 g6? Ne3 42 Qxc5 Rh4+ 43 Kg3 Rc4! and White is the one in danger of
losing) 41 ... Rg4 42 Qf2+ Kd3 43 Qf5+ Re4 44 Qh7 Ne3 45 Kg3 (45 g6?? c2 46 g7 Ng4+
47 Kh3 c1Q 48 g8Q Qe3+ (in these multiple promotion situations, the side who gets the first
check is the one who generally forces mate, as in this case) 45 ... Nf1+ 46 Kf2 c2 47 Qd7+
Kc3 48 Qc6+ Kb2 49 Qb6+ Ka1 50 Qf6+ Kb1 51 Qb6+ Ka1 with perpetual check.
36 Qxh6 Kf3 37 Qf8 Ne3! 38 fxe3 Rd2+ 39 Kg1 Rd1+ 40 Kh2 Rd2+ is perpetual check.
36 ... Re4 37 Qg7+
Once again rejecting f3+, which only draws.
37 ... Kf3
It almost feels like the vultures descend upon White’s king, to pick his carcass clean. In
reality, it is still Black who has more to lose.
38 Qc3+ Ne3!?
 This move only partially resolves Black’s concerns. Panno is probably also going for
the full point. Objectively better is 38 ... Kxf2! 39 Qxc2+ Kf3 and White has nothing better
than to take the draw.
39 Kg1!
If playing for checkmate is the heart of White’s hopes, then it barely beats. He still faces
nearly insurmountable obstacles to his goal. Tal once again dodges the draw, after 39 fxe3
Rxe3 40 Qxc4 Re2+ 41 Kh1 Re1+ 42 Kh2 Re2+.
39 ... Bg4?!
This subsequent idea somewhat invalidates Panno’s earlier 38 ... Ne3 winning try. After
this inaccuracy, White begins to generate real winning chances. Correct is 39 ... Be6! 40
fxe3 Rxe3 41 Qf6+ Ke2 42 Qxh6 Rxg3+ 43 Kh2 Re3 44 Qh5+ Kd2 45 Qxc5 c3 46 Qd6+
Ke2 47 Qc5 Bxa2 48 h5 Bb1 49 h6 Be4 50 Qc4+ Kd2 51 Qd4+ Ke2 52 Qc4+ and once
again the game ends in perpetual check.
40 fxe3 h5 41 Qe1!
 Exercise (critical position): Tal threatens mate on f2. Black
can play either 41 ... Rxe3, or back the rook up with 41 ... Re6.
Only one of the moves draws. Which one should Black play?

41 ... Rxe3?
The fog of time pressure muddles our brains, erasing both dimension and distance in our
mental mirage.

Question: What time pressure? This is move 41!

Answer: True, but Panno had not written down his score for the last few moves and
didn’t know that he had reached time control. So for all practical purposes, he remained in
time pressure! In this case, the obvious move loses.
Answer to Exercise: Black draws with GM Ludek Pachman’s fortress suggestion, 41 ...
Re6!!. Komodo, along with all my other comps, is fooled here into giving White a
completely winning assessment. The key, of course, is to avoid losing the a-pawn, which
deprives White of a passed pawn. The position is actually drawn, even if Black loses both
c-pawns. For example, 42 e4 (the best try, but it shouldn’t matter) 42 ... c3 43 Qxc3+ Kxe4
44 Qxc5 Kf3 45 Qc3+ Ke4 46 Kf2 Kf5 and Black will simply make king moves, while it
becomes clear that White’s imprisoned king is no more than a minor character in the play.
42 Qf1+ Ke4
And not 42 ... Kxg3?? 43 Qf2+ Kh3 44 Qh2 mate.
43 Qxc4+ Kf3 44 Qf1+ Ke4 45 Qxa6!
Now White gets a passed a-pawn, which renews his winning chances.
45 ... Kd4?
 The comp points out an almost impossible to see draw with 45 ... Rxg3+ 46 Kf2 Rf3+!!
(this hidden drawing line was also pointed out by Vladimir Vukovic) 47 Kg2 Kd5 48 Qa8+
Kd4 49 a4 c4 50 Qd8+ Ke3 51 a5 c3 52 Qe7+ Kd3 53 a6 Re3 54 Qd6+ Ke2 55 Qc5 Bf3+
56 Kg3 Ba8+ 57 Kf4 Rf3+ 58 Ke5 Kd2 59 a7 Re3+ 60 Kf6 c2. The fact that Black managed
to shepherd his c-pawn just one up from the promotion square is corroboration that his
strategy has been a success: 61 Qa5+ Rc3 62 Qg5+ and White must take perpetual check.
46 Qd6+!
Forcing Black’s king in front of the c-pawn.
46 ... Kc4 47 a4 Re1+ 48 Kf2 Re2+ 49 Kf1 Ra2 50 Qa6+ Kd4 51 a5 c4 52 Qb6+ Kd5
After 52 ... Kd3 53 Qb1+ Rc2 54 a6 Bf3 55 a7 c3 56 Qb5+ Kd2 57 Qe5 (now Black is
in zugzwang) 57 ... Bh1 (or 57 ... Rc1+ 58 Kf2 Bh1 59 Qe2 mate) 58 Qxh5 Rc1+ 59 Kf2
Rb1 60 Qe2+ Kc1 61 h5 c2 62 h6 White’s passers prevail.
53 a6 Ra1+ 54 Kf2 c3
Likewise, 54 ... Ra2+ 55 Ke3 Ra3+ 56 Kf4 Rf3+ 57 Kg5 Rb3 58 Qa5+ Kd4 59 a7 wins.
55 a7 c2 56 Qb3+ Kd6

Exercise (combination alert): Find one accurate move and White forces the win:

Answer: The check forces Black’s king to an unfavourable square, no matter where he
moves.
57 Qd3+! 1-0
Panno resigned here, seeing: 57 ... Ke7 (amazingly, Black’s king has no good squares:
for example, 57 ... Ke5 58 Qc3+ pops the rook or 57 ... Ke6 58 Qe4+ Kf7 59 Qxc2! Rxa7
60 Qh7+ and the rook falls all the same) 58 Qxc2! and Black is unable to play 58 ... Rxa7??
since 59 Qc5+ picks off the rook. When the government taxes us, they do two things to us
which we deeply dislike:
 1. They take our money.
2. They make us do maths.
Note too that 57 Qxc2?? is an awful blunder, since Black can safely play 57 ... Rxa7!
which draws, since White has no way to win the rook.

Game 11
Y.Averbakh-M.Tal
USSR Championship, Riga 1958
Modern Benoni

1 d4 Nf6 2 c4 e6 3 Nc3 c5 4 d5 exd5 5 cxd5 d6 6 e4 g6 7 Be2 Bg7 8 Nf3 0-0 9 0-0 Re8
10 Qc2
Averbakh later regrets this decision. 10 Nd2 is a superior method of protecting e4.
10 ... Na6
Tal’s predilection was to develop his knight on a6, rather than the more traditional d7-
square in Benonis. In this instance the a6 post is doubly tempting, since White must watch
out for ... Nb4 tricks.
11 Bf4
Black’s traditional tender spot in Benoni structures is d6. Instead, 11 Re1 prevents Tal’s
coming idea, but maybe Averbakh didn’t want to prevent it, since Tal’s idea is semi-sound at
best.
11 ... Nb4 12 Qb1 Nxe4!?
As usual with Tal, discretion loses the argument.

Question: The sacrifice’s output looks substandard to me.

This idea can’t possibly work, can it? Black doesn’t regain the piece.
 Answer: Tal just couldn’t help but make the move which appealed to emotion, over
reason. When he engaged in his signature speculative sacrifices, he in a sense, propounded a
theory before he actually formed one. I think it’s actually a semi-sound sacrifice, heavily
mitigated by the fact that White is severely tested in the coming few moves, where one slip
could cost him the game.
Objectively better is a line like 12 ... Nh5 13 Bg5 f6 14 Be3 f5 15 Bg5 Bf6 16 Bb5 Rf8
17 Bh6 fxe4! 18 Nxe4! (not 18 Bxf8?? exf3 19 Bh6 Bf5 20 Qd1 a6 and White’s light-
squared bishop is fast running short of safe squares; after 21 Bc4 b5 22 Bb3 fxg2 23 Re1 c4
24 a3 Nd3 25 Bc2 Nxb2 26 Qd2 Bxc2 27 Qxc2 Nd3 Black has too many pawns for the
exchange, as well as a strong attack) 18 ... Bg7 and Black looks fine.
13 Nxe4 Bf5 14 Nfd2 Nxd5 15 Bxd6?
When we lack accurate data, we tend to connect non-existent dots. Tal’s gamble pays
off. When we see a great player blundering away the game, it tends to dull our own pain
when we do the same, since it shows us that incompetence isn’t unique to us alone.
He should have played 15 Bg3!. No matter how hard Black stresses the system to its
limits, White’s position fails to collapse: for example, 15 ... Bh6 (in his annotations, Tal
said he planned 15 ... Qe7? 16 Bf3 Rad8 when, according to Tal, he “has an active position
as his central pawns threaten to advance”; Komodo disagrees and thinks White is just
winning after 17 Re1) 16 Bb5! Bxd2 17 Bxe8 Qxe8 18 Nxd6 Bxb1 19 Nxe8 Bc2 20 Nd6
(threat: Rad1, bagging two pieces for the rook) 20 ... Nb4 21 Nxb7 Nd3 22 Nd6 Nxb2 23
Rfc1! Bxc1 24 Rxc1 Ba4 25 Rxc5 and White stands better in the ending, since his knight
dominates and he also owns the superior bishop of opposite colours.

Exercise (combination alert): White’s greedy capture costs him his extra
material, after which is busted. How did Tal win back his material?
 Answer: Pin/overloaded defenders.
15 ... Nf6!
Once imaginary figments suddenly turn into very real dangers for White, who must return
the piece and remain a pawn down. I wonder if Averbakh expected Tal to fall for the trap 15
... Qxd6? 16 Nxd6 Bxb1 17 Nxe8 Rxe8 18 Bb5! and White wins material.
16 Bf3
Not much choice. After 16 Nxf6+?? Qxf6 17 Bd3 Qxd6 18 Bxf5 Qxd2 19 Bc2 Rad8
Black dominates. If 20 Rd1?? Black has 20 ... Qxd1+! 21 Qxd1 Rxd1+ 22 Rxd1 Re2 23 Bb3
Bd4 and White has no chance in the ending.
16 ... Nxe4 17 Nxe4 Bxe4 18 Bxe4 Qxd6

Question: Black won a pawn and also owns the better bishop.
Does White have drawing chances, due to the bishops of opposite colour?

Answer: The technical phase is not so easy for Black to convert, since the opposite-
coloured bishops complicate the winning task. However, Black can play upon the principle:
Bishops of opposite colours favour the attacker. In this case Tal is unlikely to win an
ending, and must go for a direct attack on White’s – for now – well guarded king.
19 Qc2
19 Bxb7? only benefits Black after 19 ... Rab8 20 Bf3 Rxb2.
19 ... Re7
Preparing to double rooks on the open e-file.
20 Bf3 Rae8 21 Rad1 Bd4
On its invulnerable d4 post, Black’s bishop is disdainfully aware of his superiority over
his f3 counterpart.
 22 a4 b6 23 b3
Averbakh hopes to erect a light-squared barrier to Black’s queenside pawn majority.
23 ... Re5 24 Rd2!
White regains some measure of composure after his disastrous opening. It’s not easy to
come up with a good move when dealing with the charmless sameness of seemingly eternal
defence. White intends Re2, removing a pair of rooks from the board.
24 ... h5!

Question: What is Tal’s idea behind this pawn push?

Answer: It isn’t easy to thread the position’s intricacies. In his annotations, Tal gives
this move without comment, yet it contains real depth. The idea is to play ... h4!, then he can
follow with two potential plans:
1. Play ... Be5, which forces White to weaken either his king, with g3, or his dark
squares with h3.
2. Play ... Qf4, followed by ... Be5, threatening mate on h2, after which a g3 weakening
is forced.
25 Re2 Rxe2 26 Bxe2 h4!
Continuing with his plan.
27 Kh1
After 27 Bb5 Re6 28 Qd3 Qf4 29 g3 Qf6 30 Kg2 Re5 Black continues to make progress
against White’s king.
27 ... Qf4! 28 g3
And not 28 Bc4?? (this fails to raincoat White from the coming thunderstorm) 28 ... Be5!
(Black ignores White’s threat) 29 Qxg6+ Kf8 and White must give up a piece to prevent
mate or queen loss.
 28 ... Qf6
White must be on high alert for ... Qc6+.
29 Qd1
So that White can play Bf3 if necessary.
29 ... Rd8
Threat: ... Bxf2, with a discovery on White’s queen.
30 Bg4

Exercise (combination alert): White intends to meet 30 ... Bxf2

with 31 Qe2, pinning the bishop. Find Tal’s shocking response:

Answer: Annihilation of defensive barrier/attraction/removal of the guard. The most

devious way to set a trap is to feign capitulation. Tal’s embezzlement scheme continues to
drain White of pawns, as he willingly enters what looks like a fatal pin.
30 ... Bxf2!!
Having spotted a geometric anomaly, Tal ignores White’s pin! The bishop, who lures his
victims with his seraphic sweet expression, dreams of annihilation of the church’s enemies,
which is synonymous with his own enemies.
31 Qe2 Rd2!
Now we see the darker side of Tal’s devious combination. The rook is taboo.
32 Qe8+
32 Qxd2?? walks into mate after 32 ... Qc6+. Black’s rook may be the executioner, but
his queen is the one who gives the order.
32 ... Kg7 33 gxh4 Qd4 34 Bh3 Qd3!?
Tal gets fancy. The simple 34 ... Qxh4 is stronger.
35 Bg2?!
 White puts up more of a fight with 35 Qe5+ Kh7 36 Qe7 Qxb3 37 h5 Kh6, but Black
wins in the long run anyway.

Exercise (combination alert): How did Tal simplify and win another pawn?

Answer: Simplification/ weak back rank.

35 ... Rd1! 0-1
Following 36 Qb5 (the only move) 36 ... Rxf1+ 37 Bxf1 Qe4+ (the queen circles in a
geosynchronous orbit) 38 Bg2 Qxh4 39 Qe2 Qf6 White will never be free of either his
material deficit or his insecure king, and as we all understand, never is a really long time.

Game 12
B.Spassky-M.Tal
USSR Championship, Riga 1958
Nimzo-Indian Defence

When I interviewed Boris Spassky for the San Diego Union Tribune in 1987, I never forgot
his advice: “Study your losses with great care!”. When my students want to show me only
their wins, I repeat Boris’ advice and ask them to show me their losses. When we refuse to
analyse our losses, we essentially tell ourselves: “Look, I lost. Let it be. How I lost is
inconsequential.” This game – a masterpiece – is probably one of the most painful losses in
Spassky’s glorious career, since he had Tal on the ropes and failed to find the knockout
punches at the key moments.

1 d4 Nf6 2 c4 e6 3 Nc3 Bb4 4 a3

The structurally dubious Sämisch variation, which is in my opinion the riskiest of
White’s options in the Nimzo-Indian.
4 ... Bxc3+ 5 bxc3 c5 6 e3
6 f3 is also commonly played here.
6 ... Nc6 7 Bd3 e5

Today, the more noncommittal 7 ... 0-0 is played more often.

8 Ne2
White prepares f3 next.
8 ... e4!?
A move based upon the philosophy: if a rival country is weak in attacking resources,
then you can choose to ignore its declaration of war. This move scores well for Black, since
it effectively defangs White’s attacking ambitions. On the down side, Black’s e-pawn
becomes vulnerable to ideas like Ng3 and Qc2.
9 Bb1 b6
Preparing ... Ba6 and ... Na5.
10 Ng3
White levies his strivings upon the e4-pawn; Black does the same with the c4-pawn.
10 ... Ba6 11 f3!?
 Spassky offers a pawn.

Question: Why would White offer a pawn, when he can just chop e4 instead?

Answer: Spassky was one of the few players who was unintimidated by Tal in sharp
positions. The game is duller after 11 Nxe4 Nxe4 12 Bxe4 Bxc4 13 Bd3 Bxd3 14 Qxd3 d5
and Black equalized, A.Martin-J.Plaskett, Hastings 1984/85.
11 ... Bxc4!?
Tal, fearful of Spassky’s initiative, declines.

Question: Why can’t Black accept with 11 ... exf3?

Answer: White gets attacking chances for the pawn after 12 Qxf3 Bxc4 13 Nf5 0-0 14
e4 cxd4 15 cxd4, Mu Ke-Wei Yi, Beijing 2012. Now Black should continue 15 ... Re8 16
Be3 d6 when White owns a powerful centre, a dominating knight on f5, and bishop-pair.
Still, I have my doubts about White’s full compensation for the pawn, mainly since
Spassky’s king is unable to castle.
12 Nf5?!
Threat: Nd6+ and Nxc4. Spassky overpresses. Better is 12 fxe4 0-0 13 e5 Nd5 with
dynamic equality.
12 ... 0-0
The unplayed 12 ... d5! may be an improvement: 13 fxe4 (if 13 Nxg7+ Kf8 14 Nf5 Rg8
and Black took over the initiative) 13 ... g6 14 e5 gxf5 15 exf6 Qxf6 16 Qf3 Ne7 17 Ba2
Bd3, which leaves White begging for a repetition draw after 18 Bb1 that Black is by no
means forced to accept. Indeed, he can play on with 18 ... Qh4+ 19 Qf2 Qe4 20 Bxd3 Qxd3
21 Qd2 Qc4 when Black rules the light squares.
 13 Nd6 Bd3!?
This move eliminates White’s bishop-pair. Komodo prefers Black’s position after 13 ...
Ba6! 14 fxe4 Ne8 15 Nf5 g6 when White’s attacking chances are reduced somewhat by the
fact that he is still unable to castle.
14 Bxd3 exd3 15 Qxd3 cxd4 16 cxd4 Ne8
Challenging White’s d6 outpost.
17 Nf5 d5
We reach a dynamically equal position (although the comps all like White).
18 a4
Threat: Ba3.
18 ... Nd6 19 Nxd6 Qxd6 20 Ba3 Nb4 21 Qb3 a5

Question: Doesn’t White stand better? After all Black’s knight

is pinned and White may win a pawn after 0-0, Rab1 and Bxb4.

Answer: White has no way to win a pawn, since in the line you gave, Black responds
with the plan ... Rfc8 and ... Rc4.
22 0-0 Rfc8 23 Rac1 Qe6
Tal evades the pin.
 24 Bxb4?!

Question: Why would you say this move is dubious? White

shattered Black’s queenside pawns and may well win the b4-pawn.

Answer: Black’s b4-pawn is alive and well. Not only that, a4 is actually weaker than
b4, as the proceeding moves will show.
24 ... axb4 25 Kf2
By covering e3, Spassky threatens b4.
25 ... Qd6
Protecting b4, while attacking h2.
26 h3 Kf8?!
When we veer from the correct plan, we watch a dubbed foreign film, where the actor’s
lip movements are slightly out of synch with the words spoken. Tal’s last move is an
inaccuracy, which throws away his advantage, since his king moves to a square where b4
may fall with check. Tal could have seized a clear advantage with 26 ... g6!, and if 27 Rc2
Rxc2+ 28 Qxc2 Qd7! when Black pays only cursory attention to his b4-pawn, and instead,
concentrates on White’s weakened a-pawn. If 29 Ra1 Rc8 30 Qb2 Rc4 with a bind for
Black, who seized the only open file, while keeping the b4-pawn alive.
27 Rc2 Rxc2+ 28 Qxc2 g6 29 Rc1
The big difference in this version is that Spassky seized control over the c-file.
29 ... Qd7 30 Qc6 Qxc6 31 Rxc6 Ra6
31 ... Rxa4 32 Rxb6 Ke7 is an approximately even ending.
32 a5! b3!
Tal decides to amend his b-pawn incongruity (which in the past, offended his otherwise
pristine position) into an asset. So far life moved at a languid pace for two such aggressive
players. That is about to change. Suddenly, the game turns into a mutual queening race.
33 axb6
Certainly not 33 Rxb6?? Rxb6 34 axb6 b2. Black promotes, while White doesn’t.
33 ... b2 34 b7 b1Q 35 Rc8+ Kg7 36 b8Q
So Spassky is now up a pawn, which is a temporary state, since Tal has a forcing
variation to regain it.
36 ... Ra2+ 37 Kg3 Qe1+ 38 Kh2 Qxe3
The game is even and should end in a draw – but not when both sides play for the win!
39 Rg8+

Exercise (critical decision): All the ledgers look balanced and the game should be
drawn, conditionally upon Tal finding the correct next move. Black’s king can move to either
f6 or h6. Only one of the squares allows him to draw. Which one?

39 ... Kf6?
It’s hard to remain composed when faced with oblivion. Black’s king steps on a trap
door, and the floor goes out from under him. With this move, Tal throws away his most
important asset: king safety.
Answer: Correct is 39 ... Kh6! 40 Qf8+ Kh5 41 Qxf7 Rxg2+! 42 Kxg2 Qd2+ 43 Kg3
Qg5+ with perpetual check.
40 Qd6+ Qe6 41 Qf4+ Qf5 42 Qd6+ Qe6 43 Qg3!
No draw. Spassky correctly recognizes that Black’s king is in more danger than his own.
43 ... Qe3 44 h4!
Cutting off g5 as an escape route for Black’s king, while offering d4.
 44 ... Re2

Question: Can Black get away with 44 ... Qxd4?

Answer: Nyet! That move walks into a forced mate after 45 Qg5+ (a student tried 45
Qd6+ Kf5 and now the comp found 46 Qd7+ Kf6 47 Qd8+ Kf5 48 Qg5+ Ke6 which
transposes to the main variation, where White forces mate) 45 ... Ke6 46 Re8+ Kd6 47
Qe7+ Kc6 48 Rc8+ Kb5 49 Qb7+ Qb6 50 Qxd5+. Black’s rook hangs and he is also mated.
45 Qd6+ Qe6 46 Qf4+ Qf5 47 Qh6!
Spassky wrings every drop of initiative from his position, while Tal’s king emits the
terminal feeling of Death reaching out for him. Now Qg7+ is a serious threat.
47 ... Ke7
Avoiding 47 ... Qxf3?? 48 Qg5+ Ke6 49 Re8+ Kd6 50 Qd8+ Kc6 51 Qc8+ Kb5 (51 ...
Kd6 52 Qa6+ wins the rook) 52 Qc5+ Ka4 53 Ra8+ Kb3 54 Qa3+, which wins Black’s
queen or mates.
48 Qf8+ Kf6 49 Qg7+ Ke7 50 Ra8!
 50 ... Qd7
Covering against Ra7+. Instead:
a) 50 ... Qf4+ 51 Kh3 Qf5+ 52 Kg3 and Black runs out of checks. If he tries 52 ...
Rxg2+?? (52 ... Qd7 is correct) 53 Kxg2 Qc2+ 54 Kh3 Qf5+ 55 Kg3 White’s king escapes,
while Black’s doesn’t.
b) 50 ... Qxf3 51 Qf8+ Kf6 (or 51 ... Kd7 52 Ra7+ Kc6 53 Qc5 mate) 52 Ra6+ wins.
51 Qf8+ Kf6 52 Ra6+ Re6 53 Qh8+ Ke7 54 Ra8! Re1 55 Kg3?!
Komodo weaved a mating net with the amazing line 55 Rc8!! h5 (a student suggested the
line 55 ... Qd6+ 56 g3 Re2+ 57 Kh1 Re1+ 58 Kg2 Rg1+!, which looks like perpetual check,
yet the comp says: ‘No draw!’: 59 Kxg1 Qxg3+ 60 Kf1 Qxf3+ 61 Ke1 Qe3+ 62 Kd1 Qd3+
63 Kc1 Qe3+ 64 Kb2 Qe2+ 65 Rc2 Qb5+ 66 Kc3 Qa5+ 67 Kd3 Qa3+ 68 Kd2 Qb4+ 69
Ke2 Qb5+ 70 Kf2 and, whew, no more checks!) 56 f4! Re4 57 f5!! gxf5 58 Qf8+ Kf6 59
Kg1! Rxh4 (or 59 ... f4 60 Qh8+ Ke7 61 Ra8! Re6 62 Qb8 Qd6 63 Qf8+ Kf6 64 Qh6+ Kf5
65 Qxh5+ Ke4 66 Qf3+ Kf5 67 Ra5 Kf6 68 Rxd5 and Black is busted) 60 Rc6+! Kg5 (after
White’s series of blows, Black’s king wonders if all his teeth are still intact; instead, 60 ...
Qxc6 61 Qh6+ wins the queen) 61 Qg7+ Kf4 62 Kf2! Rg4 63 g3+ Ke4 64 Qe5+ Kd3 65
Qe3 mate.
55 ... h5 56 Kf2 Re6
 Exercise (combination alert): Despite the reduced material, Black’s king is in
grave danger. We sense the presence of a combination, just below the threshold
of consciousness. It’s almost impossible for a human to find, but give it a shot:

57 Rc8?!
Answer: 57 Qb8!!, threatening Ra7, wins. If 57 ... Kf6 58 g4! (now White’s kingside
pawns assume a key role) 58 ... hxg4 59 fxg4 Re4 60 Qh8+ Ke7 61 Qf8+ Kf6 62 Ra6+ Re6
63 g5+ Kf5 64 Qa3! threatens Qf3 mate, and if 64 ... Kg4 (alternatively, 64 ... Rxa6?? 65
Qf3+ Ke6 66 Qf6 mate, while 64 ... Qc7 65 Qh3+ Ke4 66 Rxe6+ fxe6 67 Qf3+ Kxd4 68
Qe3+ Kc4 69 Qc1+ picks off Black’s queen) 65 Qf3+ Kxh4 66 Qf4+ Kh5 67 Ra1 leaves
Black helpless.
57 ... Rd6?!
57 ... Rb6 was necessary.
58 Qf8+ Kf6
 Exercise (combination alert): Once again, Spassky is given the opportunity
to find a forced win. This time you may have a better chance to find it,
since you are more familiar with the tactical themes from the last exercise:

59 Re8?!
Answer: It’s critical that White include his pawns in the mating attack. Therefore correct
is 59 g4! hxg4 60 fxg4 Rb6 61 Kf3!!. This move is unbelievably difficult to find. Even
White’s king participates in the hunt of Black’s king: 61 ... Rb1 (61 ... Rb3+ 62 Kf4 Rb1 63
Qh8+ Ke7 64 Kg5!! costs Black his queen) 62 Qh8+ Ke7 63 Qe5+ Qe6 64 Qc7+ Qd7 (or
64 ... Kf6 65 Qd8+ Qe7 66 Qh8+ Ke6 67 Qe5+ Kd7 68 Rc7+ Kd8 69 Qxe7 mate) 65 Qc5+
Qd6 (65 ... Ke6 66 Rc6+ wins the queen) 66 Re8+ Kd7 67 Qc8 mate.
59 ... Re6 60 Qh8+ Kf5 61 Qh6!
Spassky is relentless. He now threatens mate on g5.
61 ... Kf6
Avoiding both 61 ... Rxe8?? 62 Qg5+ Ke6 63 Qe5 mate and 61 ... Qxe8?? 62 Qg5 mate.
62 Qh8+ Kf5
Unbelievably, the position is even, despite appearances. Spassky soon grows frustrated
and blunders.
63 Rd8 Qc6 64 Rc8 Qa6! 65 Kg3?
Now the players’ fortunes reverse and the advantage swings to Black. The game is even
after the correct 65 Rc2.
65 ... Qd6+!
 Exercise (critical decision): From this point, Tal launches
an all-out assault on Fort Spassky. White’s king can move
to f2 or h3. He survives only one of the squares. Which one?

Answer: 66 Kh3!
In the wild, alertness to approaching predators is a key survival factor of any species. If
66 Kf2?? (“There are few – if any – in the world, capable of out-thinking me,” theorizes the
king, as he taps his forehead knowingly) 66 ... Qb4! (threatening mate on e1; now the white
king’s stability is one of a provisional quality, soon to alter to its opposite) 67 Kg3 Qe1+ 68
Kh3 Qh1+ 69 Kg3 Re2! and, astonishingly, White has no checks to save himself from mate,
except spite checks on e5 or f6, which hang his queen.
66 ... Re1
 Exercise (critical decision): No time to rest. Black threatens mate on h1.
White can play either g3 or g4+. Only one of the moves holds the draw:

67 g3?
After this move, white defenders flee in terrified disarray, like a crowd from a burning
theatre. This move allows Tal a forced win.
Answer: Spassky draws a pawn down ending with 67 g4+! Kf4 68 Rc3! hxg4+ 69 fxg4
Re3+ 70 Rxe3 Kxe3 71 Qc8! Kxd4 72 h5 gxh5 73 gxh5 Qf6 74 Qg4+ Ke5 75 Qe2+ Kd6 76
Qa6+ Ke7 77 Qa3+ Ke6 78 Qe3+ Qe5 79 Qb6+ Ke7 80 h6 (White’s h-pawn offers just
enough play to hold the game) 80 ... Qf5+ 81 Kh2 Qe4 82 Kg3 d4 83 h7 Qxh7 84 Qxd4,
which is a theoretical draw according to my all-seeing, all-knowing tablebase. Admittedly
Black still retains practical chances to win the game.
67 ... Rg1?
Tal misses the decisive 67 ... Qa6! 68 g4+ Kf4 69 Qh6+ Kxf3 70 Rc3+ Ke4! (once
again, White is out of checks) 71 Kg3 Qf1 72 Qd2 Qg1+ 73 Qg2+ Kxd4 74 Rf3 hxg4 75 Rf2
Re3+ 76 Kxg4 f5+ 77 Kg5 (or 77 Kf4 Re4+ 78 Kg5 Rg4+) 77 ... Rg3+!, which wins the
queen.
68 f4 Re1?!
68 ... Ke4! still creates great difficulties for White after 69 Kh2 Rd1 70 Qe5+ Qxe5 71
dxe5 Kf3 72 Rc7 Rd2+ 73 Kh3 Kf2! (Black’s attack continues!) 74 Kh2 (74 Rxf7?? Kg1!
forces mate next move) 74 ... d4 75 Rxf7 d3 76 f5 Kf1+ 77 Kh1 gxf5 78 e6 Re2 79 Rxf5+
Ke1 80 Rxh5 d2 81 Rd5 d1R (always underpromote when you can!) 82 Rxd1+ Kxd1 83 g4
Rxe6 84 Kg2 Ke2 85 Kg3 Ke3 86 g5 Ke4 87 Kg4 Ra6 88 h5 Ra1 89 h6 Rg1+ 90 Kh5 Kf5
and White’s pawns are halted.
69 Rc2?
 After this error, White makes a once difficult defensive task impossible. White still has
drawing chances after 69 Re8!.
69 ... Qe6!
After multiple false starts, Tal hits upon a winning plan. However, he decides to look
past pragmatic considerations. Capablanca, Fischer and Karpov would have played the far
clearer 69 ... Qa6! 70 Qc8+ Qxc8 71 Rxc8 Rd1, which is an easy win in the rook and pawn
ending for Black.
70 Rf2
70 Qc8 Qxc8 71 Rxc8 Rd1 transposes to the lost rook and pawn ending from the above
note.
70 ... Rh1+ 71 Kg2
Alternatively, 71 Rh2 Ke4+ 72 Kg2 Rxh2+ 73 Kxh2 Kf3 74 Qe5 Qc6 75 Kh3 Qc4 76
Qe1 Qc8+ 77 Kh2 Qc2+ 78 Kh3 Qg2 mate.
71 ... Qe4+ 72 Rf3

Exercise (combination alert): Previous conjecture morphs

into hardened certainty. The bowstring is taut. All which is
required of us now is to let the arrow fly. How did Tal force mate?

Answer: His king must assist in the mating attack.

72 ... Kg4!
Answer no.2: A student suggested 72 ... Qe2+! 73 Rf2 Rh2+ 74 Kxh2 Qxf2+ 75 Kh3
Qf1+ 76 Kh2 Kg4 77 Qc8+ f5 78 Qc2 Qh3+ 79 Kg1 Qxg3+ which also wins.
73 Qc8+ f5! 0-1
Oh, the sweet bliss of the perfect geometric fit! And so ends one of the most amazing
endings I have ever seen. Your normally verbose writer has inexplicably run dry of
metaphors to describe my astonishment. After 74 Kxh1 (or 74 Qc3 Rh3! when the threat of
... Rxg3+ forces mate after 75 Kf2 Rh2+ 76 Kg1 Qe2 and mate in two moves) 74 ... Kh3!
White’s position resembles a ransacked home, after the burglars overturned virtually every
item, in search of cash, jewellery and the Sony Playstation4. 75 Qc3 Qb1+ mates in two
moves.

Game 13
M.Tal-G.Fuster
Portoroz Interzonal 1958
Caro-Kann Defence

I met the Hungarian/Canadian Geza Fuster at the 1975 (well, I think it was 1975!) Canadian
Open, where the veteran IM complained to my father that he lost four games on time. That I
found very odd, since Fuster won both the Canadian and U.S. blitz championships in the
1950’s. I guess age slows us all down.

1 e4 c6
If I were paired with Tal, 1 ... c6! or 1 ... e6! would be my first choices.
2 d4 d5 3 Nc3 dxe4 4 Nxe4 Nd7
Smyslov’s system of the Caro-Kann, which is about equally solid as its slightly more
popular counterpart 4 ... Bf5.
5 Nf3
5 Ng5 is the most popular and most challenging line for Black today: 5 ... Ngf6 6 Bd3
(or 6 Bc4 e6 7 Qe2 Nb6 which is another dangerous offshoot for Black) 6 ... e6 7 N1f3 Bd6
8 Qe2 h6 9 Ne4 Nxe4 10 Qxe4 Qc7 11 Qg4 Kf8. I have been on the black side of this
position many times. Visually, the position looks quite awful for Black, but this is an optical
illusion, since Black will soon enjoy a coiled spring effect, chasing White’s queen around.
5 ... Ngf6 6 Nxf6+
6 Ng3 is the alternative.
6 ... Nxf6 7 Bc4 Bf5
IMs don’t fall for 7 ... Bg4?? 8 Bxf7+! Kxf7 9 Ne5+ Kg8 10 Nxg4 with an extra pawn
and a winning position for White.
8 Qe2
Tal prepares to castle long, rather than go for the safer 8 0-0 e6 9 c3 Be7 10 Bd3 Bg6 11
Bf4 0-0 12 Ne5 a5 13 Nxg6 hxg6. The game is even, despite White’s edge in space and
bishop-pair, G.Guseinov-G.Kamsky, Khanty-Mansiysk 2013.
8 ... e6 9 Bg5 Be7 10 0-0-0 h6
Safer is 10 ... Bg4 11 h3 Bxf3 12 Qxf3 Nd5 13 Bxe7 Qxe7 with an equal position, which
Fischer managed to win, R.Fischer-T.Petrosian, Bled 1961.
11 Bh4 Ne4?!
Right idea (to simplify); inaccurate implementation. This move doesn’t feel properly
aligned with the opening’s protocol. Correct is 11 ... Nd5! 12 Bxe7 Qxe7 13 Ne5 0-0-0
when it will be difficult for White to achieve anything meaningful with his extra space,
J.Estrada Nieto-L.Eperjesi, Budapest 1995.
12 g4!

Principle: Create confrontation when leading in development.

12 ... Bh7
Wisely avoiding:
a) 12 ... Bxg4?? 13 Bxe7 Bxf3 14 Qxf3 leaves Black down a piece.
b) 12 ... Bxh4?? 13 gxf5 Nxf2 14 fxe6 Nxd1 15 exf7+ Kd7 16 Rxd1 with a winning
attack for White.
13 Bg3!?
Tal isn’t satisfied with a mere strategic edge after 13 Bxe7 Qxe7 14 Bd3 Nd6 15 Bxh7
Rxh7 with a touch of space for White, as well as time, since Black’s h7-rook is misplaced
and needs to be brought back into play.
13 ... Nxg3?!
It’s tempting to pick off the bishop-pair, but not if it opens a key file.
14 fxg3?!
 Question: Shouldn’t Tal open the h-file instead?

Answer: I think taking with the h-pawn is inherently stronger. Tal just can’t resist
opening the f-file, even though this means a devaluation of his structure. After 14 hxg3!
Black’s king is stuck in the centre, since castling kingside looks too risky with the open h-
file.

Question: Why can’t Black prepare to castle long with 14 ... Qc7?

Answer: White makes good use of his open h-file with 15 g5! Bf5 (15 ... hxg5? allows
16 Ne5 Bf6 17 Rh2! and the pin along the h-file is decisive; if 17 ... Kf8 18 Rdh1 Kg8 19
Nxf7! is crushing) 16 gxh6 Rxh6 17 Rxh6 gxh6 18 Ne5 with a clear advantage for White.
14 ... Qc7 15 Ne5 Bd6
Covering f7, which enables Black to castle long.

Question: Is castling kingside suicidal for Black?

Answer: I wouldn’t say suicidal, but I also wouldn’t call it wise – especially if Tal is
your opponent. White generates attacking chances after 15 ... 0-0?! 16 h4 Rad8 17 g5! (it’s
worth a pawn to open the h-file) 17 ... hxg5 18 hxg5 Bxg5+ 19 Kb1 Qe7 20 Rh5 (preparing
to double rooks on the h-file) 20 ... Bh6 21 Ng4 Bg6 22 Nxh6+ gxh6 23 Rxh6 and White
regained the sacrificed pawn, while keeping attacking chances alive.
16 h4 f6?
Defence of a slightly inferior position is a monotonous occupation, which tests our
patience. This entanglement induces baroque consequences for Black. Instead, 16 ... 0-0-0??
loses on the spot to 17 Nxf7!. Correct is the simplifying 16 ... Bxe5! (this is the only path to
restoration) 17 dxe5 Rd8 18 Rxd8+ Qxd8 19 Rd1 Qa5 20 Qe3 0-0 21 Rd7 Qb4 22 Qc3 Qb6
23 b3 Rd8 and Black should hold the game.

Exercise (combination alert): Tal’s next move is similar to the criminal,

thinking out loud his next ‘job’, within earshot of two police officers
drinking coffee in the adjacent booth of the diner. What did Tal play?

Answer: Of course Tal sacrificed! For the piece Black’s king is trapped in mid-board,
while the centre opens around him.
17 Bxe6!
I suspect that Tal may have attempted to dissect a frog with a butcher’s cleaver in high
school biology class. The one beneficial thing about shock, is that the trauma imprints itself
so deeply in our mind, that we actually learn from it and don’t repeat our mistake. In this
case, the lesson is: don’t allow Tal to keep your king trapped in the centre!

Question: It feels to me like this sacrifice belongs in the realm

of speculation, rather than verified fact. Is this sacrifice sound?

Answer: Not only is it sound, the comps claim White is winning. There is no way Tal
would play the milquetoast 17 Nf3? 0-0-0! 18 Bxe6+ Kb8. Black picks up g3, after which
his bishop-pair gives him the advantage.
17 ... fxe5 18 dxe5 Be7
Of course, Black allows White’s e5-pawn to remain on the board, since this clogs the e-
file – for now. 18 ... Bxe5?? 19 Rd7 Qb8 20 Re1 Bg8 (or 20 ... Bxg3 21 Bf7+! Kxd7 22
Qe7+ Kc8 23 Be6 mate; the bishop rudely pantomimes himself playing a violin at the black
king’s protests) 21 Qc4 Bh7 22 Bf5 Bg8 23 Qc5 forces mate.
19 Rhf1!
White also reaches a winning attack after 19 Rd7! Qxd7 20 Bxd7+ Kxd7 21 e6+ Kc8 22
Rd1. Black has no good way to deal with the coming Rd7. If 22 ... Rd8? 23 Qe5! Rxd1+ 24
Kxd1 Bf6 25 Qd6 Bg8 when the simplest is 26 Qf8+ Kc7 27 Qxa8.
19 ... Rf8 20 Rxf8+
Also strong is 20 g5! Rd8 21 gxh6 Rxf1 22 Rxf1 Bc5 23 Rf7 Qxf7 24 Bxf7+ Kxf7 25
Qh5+ Ke6 26 Qg5 Be7 27 Qg4+ Bf5 28 Qc4+ Rd5 29 hxg7 Bh7 30 Qf4 Bg8 31 g4 Rxe5 32
Qf8!. White wins one of the bishops, with an easy conversion.
20 ... Bxf8
Black hopes to have time for ... Rd8 next.
21 Qf3!
21 Rd7? is mistimed: 21 ... Qxd7 22 Bxd7+ Kxd7 23 e6+ Kd6 and Black stands no
worse, and may even hold the advantage.
21 ... Qe7
Not 21 ... Rd8?? 22 Rxd8+ and Black lacks a good recapture.

22 Qb3?
There is no way to load bullets into the chamber of an imaginary gun. This move mottles
an otherwise brilliant idea. Tal achieved partial success, yet the process remains
incomplete. He sets up a deep trap – in a sense milking his premise – when in fact he has a
forced win here. His big weakness: he wanted too much, after which he risked getting
nothing.
Stronger was the immediate 22 Bd7+! Qxd7 23 Rxd7 Kxd7 24 Qf7+ Be7 25 e6+ Kd6 26
c4! (threat: c5+!; attempts to buy the queen’s affections with the gift of g7 is haughtily
refused) 26 ... c5 27 Qf3! Kc7 28 Qf4+ Kc6 29 Qe5 Rg8 30 Qd5+ Kb6 31 Qd7, and if 31 ...
Bf6 32 Qd6+ Ka5 33 Qxc5+ Ka4 34 Qb5 mate.
 22 ... Rb8??
That which we are unable to heal, we tend to hide. When Tal generated seemingly
endless petty threats, often there arose a cumulative effect, where the opponent inexplicably
blundered. Now Tal’s dreams are fulfilled beyond his wildest expectations. Black had to try
22 ... b5! 23 g5 h5 24 a4 a6 25 Rf1 Be4 26 Bf7+ Kd7 27 Qe3! Qxe5 (27 ... Bd5 28 Qb6!
Kc8 29 Bxh5! Be6 30 Qxc6+ Kb8 31 Qb6+ Kc8 32 Rf7! wins, since Black is unable to play
32 ... Bxf7?? 33 Bg4+) 28 Bg6 Bc5 29 Qxe4 Qxe4 30 Bxe4 Rf8 when he should hold the
draw, due to the bishops of opposite colours.
23 Bd7+!
Tal had seen much deeper into the position than his opponent.
23 ... Qxd7
Black must part with the queen, since after 23 ... Kd8?? 24 Bf5+ White regains the piece
with an easily won game.
24 Rxd7 Kxd7
The shadows fail to offer concealment for Black’s king.
25 Qf7+ Be7

Question: It feels as if White’s lone queen meets Black pieces with

the stern resolve of Joan of Arc, facing her inquisitors. Isn’t White
just losing here, since rook and two bishops are too much for just one queen?

Answer: You may have to revise your previous assumptions. I was hoping you would
say that! The position is a forced win for White. Let’s pause for an exercise and try to work
it out.

Exercise (combination alert): White to play and force the win:

 Answer:
Step 1: Push the e-pawn forward, forcing Black’s king to an unfavourable square,
whether he moves up or down.
26 e6+ Kd8
Avoiding 26 ... Kd6?? 27 Qf4+ when we see why Tal played his earlier Qb3?, which
was intended to lure Black’s rook to b8.
Step 2: Take the g7-pawn, after which Black’s h7-bishop is trapped.
27 Qxg7 1-0
27 ... Be4 28 Qe5! is a fatal double attack. The queen reprimands black bishop and rook,
neither of whom can believe their collective four ears.

Game 14
M.Bobotsov-M.Tal
Varna 1958
King’s Indian Defence

1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4 e4 d6 5 f3

The Sämisch King’s Indian.
5 ... 0-0 6 Nge2
6 Be3 suppressing ... c5 is more commonly played here: 6 ... e5 7 d5 Nh5 8 Qd2 Qh4+ 9
g3 Nxg3 (at this point, I thought my opponent had blundered) 10 Qf2 (10 Bf2?? hangs
material to 10 ... Nxf1!; three unfortunates fell into this trap in my database) 10 ... Nxf1 11
Qxh4 Nxe3.

Compare this line with Tal’s sacrifice. Black picks up the c4-pawn and gets two minor
pieces and two pawns for the queen, with reasonable practical chances – although at the
IM/GM level, the comps have worked it out to an advantage for White. In C.Lakdawala-
S.Trump, Montreal, 1976. I grossly underestimated Black’s initiative, thinking the benefits
of Black’s sacrifice are not commensurate with the high price tag. Of course, I went on to get
clobbered.
6 ... c5 7 Be3 Nbd7
7 ... Nc6!? is a much disputed theoretical sacrifice: 8 dxc5 dxc5 9 Qxd8 Rxd8 10 Bxc5
Nd7 11 Ba3 Nde5 12 Nf4 Nd4 and Black’s control over the central dark squares offers
compensation for the pawn, J.Granda Zuniga-P.Glavina, San Fernando 1991.
8 Qd2
Instead, 8 d5 Ne5 9 Ng3 h5 10 Be2 h4 11 Nf1 e6 leads to a Benoni-like position, with
mutual chances, R.Ponomariov-M.Carlsen, Medias 2010.
8 ... a6 9 0-0-0!?

One must be a person of vast self-confidence to provoke Mikhail Tal! Far safer are 9 d5
and 9 Rd1.
9 ... Qa5 10 Kb1
Now Black must watch out for Nd5 tricks ...
10 ... b5!
... which Tal totally ignores! Of course, a Benko-like Gambit is infinitely more tempting
when your opponent castles long.
11 Nd5!?
“After what happens next, it is possible to put a big, fat question mark by this move, but
how many of us would have fallen into Tal’s trap?” writes IM Andrew Martin. I would add:
the line is actually playable for White, but only if you are a comp! From a human
perspective, White’s moves are too difficult to find over the board, after Tal’s coming
sacrifice.
Martin suggests 11 dxc5 before playing Nd5. Even here Black gets loads of
compensation and practical chances after 11 ... dxc5 (not 11 ... Nxc5?! 12 Bxc5 dxc5 13
Nd5 Qxd2 14 Nxe7+ Kh8 15 Rxd2 and I don’t believe in Black’s full compensation) 12
Nd5 Nxd5! 13 Qxa5 Nxe3 14 Rc1 Nxc4 15 Rxc4 bxc4 16 Nc3 Rb8 17 Bxc4 Ne5 18 Be2
Be6 19 Rd1 Nc6 20 Qa3 Nb4. Black has only two minor pieces for the queen, yet may have
full compensation, since all his supercharged pieces take direct aim at White’s king. In fact,
if given a choice, I prefer Black’s position.

11 ... Nxd5!
Tal, a discriminating shopper when it comes to sacrifice, knows a good deal when he
sees one. A long standing idea, accepted by precedence and faith, can be upended by a
single suggestion from a person with an open mind. Apparently this wasn’t Tal’s idea.
Instead, he followed a suggestion which flowed from the imagination of Rashid
Nezhmetdinov, one of the greatest attackers and combinational players ever.

Question: Rashid who? Why is it I have never heard of Nezhmetdinov?

Answer: Many chess players haven’t, since he never rose to world-class status. But if
you want to improve in your ability to generate initiative, attack and see combinations, then
you would be well advised to study Nezhmetdinov’s games. A few pro-Nezhmetdinov
endorsements: “Rashid Nezhmetdinov is a virtuoso of combinational chess” – David
Bronstein; “Nobody sees combinations like Rashid Nezhmetdinov” – Mikhail Botvinnik;
and “His games reveal the beauty of chess and make your love in chess not so much the
points and high placing, but the wonderful harmony and elegance of this particular world” –
Mikhail Tal.
Tal was not a retreater and wouldn’t even consider the anti-initiative move 11 ... Qd8?
12 Nxf6+ Bxf6 13 dxc5 dxc5 14 Nc3 bxc4 15 Bxc4 e6 16 Na4 Rb8 17 Bb3. I don’t think b2
will be so easy to attack and I prefer White’s position, since he has readily available pawn
targets and an open centre, which enhance his chances for a successful defence.
12 Qxa5 Nxe3

Question: How can such a sacrifice be sound?

Black only got two pieces for the queen.

Answer: A temptingly original idea should be weighed with its utility. In this instance, I
think the sacrifice holds up, even under the comps’ ruthless scrutiny. They assess at nearly
even – a very bad sign for the material up side. Let’s frame our thoughts by first gathering
information. Black gets the following for his queen:
1. Two pieces, but since White’s rook and c4-pawn are simultaneously attacked, Black
picks up at least one more pawn.
2. Black’s unopposed dark-squared bishop rules its diagonal, which ends at White’s
king.
3. Queenside lines open rapidly, after which b2 may come under heavy assault.
Conclusion: From a practical perspective, White’s position is virtually impossible for a
human to defend.
13 Rc1
13 Rd3 has been tried too. White didn’t survive that one either: 13 ... Nxc4 14 Qe1 Rb8
15 h4 cxd4 16 Nxd4 Nc5 17 Rd1 Na4! and White found himself under heavy assault,
R.Beyen-G.Klompus, correspondence 1964.
13 ... Nxc4 14 Rxc4?!
White gives immediate voice to his anxiety, with the thought: “No matter how high the
pain, the wound must be cauterized, before I bleed out.” Ah yes, as I recall in the seven
steps of grief, first comes denial, then later anger. We busy ourselves with the little problem
in front of us, at the peril of ignoring the long-term problem of a far greater threat. In this
instance, White eliminates a dangerous attacker, at the too-high cost of allowing the b-file to
open.
White had to try 14 Qe1 cxd4 15 Nf4, K.Bulski-R.Barski, Warsaw 2012. Black should
continue 15 ... Nde5 16 Nd3 Be6 when we note an almost imperceptible renewal of White’s
defensive chances. Komodo assesses at dead even – although I would take Black’s side
every time, if given a choice.
14 ... bxc4 15 Nc1?!
The knight is misplaced on c1. He puts up greater resistance with 15 Nf4 Rb8 16 Bxc4
Bxd4 17 b3 Ne5 18 Be2 although even here, I hate White’s position.
15 ... Rb8
Target: b2.
16 Bxc4 Nb6
Tal didn’t want to lose a tempo after 16 ... Bxd4 17 Nb3 Bg7.
17 Bb3
The problem is the bishop is unstable on b3, vulnerable to a future ... c4. 17 Bxa6 Bxd4
18 Nb3 was his last chance. After 18 ... Be5 19 Bb5 Be6 both ... c4 and ... Nc4 are in the
air, and I don’t see White surviving.
17 ... Bxd4 18 Qd2 Bg7 19 Ne2
Hoping to have time for Nc3, which eases the pressure on b2.
19 ... c4 20 Bc2

Exercise (combination alert): We note a snaking constellation of attackers around

White’s king. At this stage, within his mind, Tal must have been grinning evilly and rubbing
his hands together, like a silent movie villain. Find the key attacking move:

20 ... c3!
 Answer: Clearance/line opening. This move openly illuminates the extent of the damage
to the white king’s safety.
21 Qd3
21 Nxc3?? is crushed by the simple double attack 21 ... Nc4. Would-be defenders
discreetly disperse upon the c-pawn’s entrance. Past decisions made in the heat of passion
are then repentantly amended in the light of the sober present. I will bet that at this stage
Bobotsov deeply regretted his 14 Rxc4?!, opening the b-file for Tal.
21 ... cxb2
One glance and we can see that White’s king won’t survive.
22 Nd4 Bd7 23 Rd1 Rfc8
Martin compares this position to a Dragon Yugoslav Attack, gone terrible awry for
White.
24 Bb3
Bobotsov clenches tightly to his final hope, which is to gum up the b-file.
24 ... Na4!
The knight augments his prestige by climbing to a higher square, where it targets c3,
which forces White to reopen the b-file by exchanging on a4.
25 Bxa4
Well, White’s hope didn’t last all that long. 25 Ne2?? fails miserably to 25 ... Bb5 26
Qd2 Bxe2 27 Bxa4 (27 Qxe2?? walks into the fork 27 ... Nc3+) 27 ... Bxd1 28 Bxd1 Rc1+
29 Qxc1 bxc1Q+ 30 Kxc1 with an extra rook for Black.
25 ... Bxa4
Threat: ... Bxd4, and if the queen recaptures, then ... Bc2 mate.
26 Nb3 Rc3!
There are multiple paths to the win, one of which is 26 ... Bxb3! 27 axb3 Rb5 (threat: ...
Ra5 and ... Ra1 mate) 28 f4 (intending e5) 28 ... Rbc5 29 e5 dxe5 30 fxe5 Bxe5 and White
loses the fight for c1, the node of both sides’ contention. There is no defence to the coming ...
Rc1+.
27 Qxa6
 Exercise (planning): Black has two methods to force a win. Find one of them:

Answer:
Step 1: Eliminate the knight, a key defender of c1.
27 ... Bxb3!
Answer no2: Also decisive is 27 ... Bb5! 28 Qa7 Rbc8 29 Rd4 Rc2 30 Qb6 Bd3!, and
if 31 Rxd3 Rc1+ 32 Nxc1 bxc1Q mate. “I continue to hear errant stupidity spouted from that
direction,” says the queen, pointing to her brother. She adds: “If it continues to irritate me, I
will silence it at its source, by ordering my brother’s wagging tongue cut out from his
mouth.”
28 axb3
Step 2: Double rooks on the c-file, after which there is no defence to the coming check
on c1.
28 ... Rbc8! 29 Qa3 Rc1+ 30 Rxc1 Rxc1+ 0-1
After 31 Ka2 it would be inaccurate to state that White’s defence missed by a hair’s
breath, since Black has no less than three separate mates in one, the prettiest of which is to
underpromote with 31 ... b1B mate. The bishop eyes White’s king the way a cult member
gazes upon an adored leader.

Game 15
M.Tal-NN
Riga (simul) 1958
Nimzo-Indian Defence

Question: Why show a simul game in a best games collection?

 Answer: Tal routinely created amazing combinations in a simul setting, and I wanted to
show an example. In the late 1980’s I had a student named Jack Miller, who had the good
fortune to play Tal in a simul. Jack played the game of his life and looked as if he was about
to secure a world champion’s scalp. But no, Tal had seen deeper and came up with an
absolutely astonishing resource to pull out the win. I lost contact with Jack several decades
ago and if I had that game, I would have used it instead of this one, which is no less
amazing.

1 d4
Another classic isolani game: 1 c4 e6 2 Nf3 d5 3 d4 Nf6 4 Bg5 Be7 5 Nc3 0-0 6 e3
Nbd7 7 Bd3 c5 8 0-0 cxd4 9 exd4 dxc4 10 Bxc4 Nb6 11 Bb3 Bd7 12 Qd3 Nbd5 13 Ne5
Bc6 14 Rad1 Nb4 15 Qh3 Bd5 16 Nxd5 Nbxd5 17 f4! Rc8 18 f5 exf5 19 Rxf5 Qd6.

Exercise (combination alert): How did Botvinnik dismantle Black’s position?

Answer: Removal of the guard: 20 Nxf7! Rxf7 21 Bxf6 Bxf6 22 Rxd5 Qc6 23 Rd6! Qe8
24 Rd7 1-0, M.Botvinnik-M.Vidmar, Nottingham 1936. This game is annotated in
Botvinnik: Move by Move.
1 ... Nf6 2 c4 e6 3 Nc3 Bb4 4 e3 0-0 5 Bd3 d5 6 Nf3 c5 7 0-0 cxd4 8 exd4 dxc4 9
Bxc4
 So we reach a classical isolani position, from a Nimzo-Indian move order. I have been
in Black’s position several times, but always from a Caro-Kann, Panov-Botvinnik move
order.
9 ... Nbd7
I usually play 9 ... b6, keeping both d7 and c6 options open for the b8-knight.
10 Bg5 Be7?!

Question: Doesn’t this move violate the principle: Don’t move

the same piece more than once in the opening, without good reason?

Answer: It’s a book move which indeed violates this principle. Black would be better
served playing for the hanging pawns structure with 10 ... Bxc3 11 bxc3 Qc7 followed by ...
b6 and ... Bb7, completing development.
11 Qe2 Nb6 12 Bb3 Bd7 13 Rad1 Nbd5
Alternatively, 13 ... Bc6 14 Ne5 and then:
a) 14 ... Nfd5 15 Nxc6 bxc6 16 Bc1 Nd7 17 Rd3, M.Euwe-G.Fuster, Beverwijk 1953.
Black looks only slightly worse after 17 ... N7f6.
b) 14 ... Nbd5 15 Rfe1 Bb4 16 Bd2 Bd6 17 Bc2 Nxc3 18 Bxc3 Bd5 19 Bb1 Rc8 20
Qd3 g6 21 Bd2 Re8 22 Bg5 Be7 23 Qh3 Nh5 24 Be3 Bg5 saw Black defend well and
equalize, D.Anderton-I.Clark, British League 2004.
c) 14 ... Bd5 15 Bc2 Nfd7 16 Bf4 Nxe5!? 17 dxe5 Bg5 18 Qh5 h6 19 Bg3 Qc7 20 h4
Bd8 21 Rd4! (intending to swing the rook over to g4; Black is already hopelessly lost) 21 ...
Qc5? 22 Rg4 f5 23 exf6 Rxf6 24 Be5 Bc6 25 Rxg7+. Botvinnik played isolani positions to
perfection, M.Botvinnik-A.Budo, Leningrad 1938.
14 Ne5 Re8?
By bolstering e6, Black weakens f7, which is even more important. I would try 14 ...
Bc6 15 f4! intending f5, after which White’s chances still look more promising.
15 Rfe1!?
Tal’s last move betrays his lifelong fixation on his opponents’ kings. He prefers to attack,
rather than win a pawn with 15 Nxd5 Nxd5 16 Qf3! Rf8 17 Bxd5 exd5 18 Nxd7 Qxd7 19
Bxe7 Qxe7 20 Qxd5 with an extra pawn for White. Tal may have rejected this since major
piece endings are notoriously difficult to win for the pawn up side.
15 ... Bc6 16 Rd3!
The rook is Tal’s designated assassin. This is one of those ‘or else!’ warning moves. Tal
prepares to swing the rook over to either g3 or h3. One of my students suggested 16 f4 h6 17
Bxf6 Bxf6 18 f5 exf5 19 Qc4? intending to apply pressure to d5 and f7. The idea is flawed
and Black escapes after 19 ... Bxe5 20 dxe5 Qb6+ 21 Kh1 Ne3! 22 Qxf7+ Kh8 and White is
the one in trouble.

16 ... Nxc3?

Question: Didn’t Black’s last move stabilize d4, and also increase
the attacking influence of the b3-bishop, both which only benefit White?

Answer: Yes, and yes! No single factor of Black’s strategic trouble is fatal. It requires
the confluence of two or more to sink his (or ‘her’, since I don’t know NN’s gender) game.
After this move Black’s situation grows intolerable. Black has to try 16 ... Rc8 17 Rh3 g6
and then pray he doesn’t get mated.
17 bxc3 Nd5
Or 17 ... Bd5 18 c4 Be4 and White still gets his combination. I can’t show it to you,
otherwise I give away the answer to the combination alert!
 The import of Black’s error has yet to register. The paranoid mind detects the presence
of unseen enemies, some perhaps real, and many more which are simply figments of a
diseased imagination. Tal took full advantage of this human trait. But he liked it even better
when his opponents trusted him, as in this case.

Exercise (combination alert): White’s pieces survey Black’s king surreptitiously,

the way my wife Nancy does when I enter the kitchen, just in case – God forbid –
I am tempted to eat meat, dairy, sugar, gluten, any processed food, or a list of
about a dozen other forbidden food items. How did Tal respond?

18 Nxf7!
Answer: Annihilation of defensive barrier/attraction. The knight, like an action movie
hero, stylishly spins, aims and fires.
18 ... Kxf7
After the kick to the king’s ‘private’ zone, it is highly unlikely he will ever father
children.
19 Qxe6+ Kf8
When the king dares to look up, he sees his sister’s wrathful face.
 Exercise (combination alert): The first combination alert was
rather easy, and I suspect most of you saw Tal’s knight sacrifice
on f7. Now comes the hard part. How did Tal continue his attack?

Answer: The answer is obscured in a fog of paradox. How often do you see the
attacking side sacrifice a piece, and then follow by retreating a developed piece back to its
square of origin?
20 Bc1!!
Intending Rf3+ and if Black blocks with ... Bf6, then Ba3+ is a killer.
Answer no.2: Also effective is 20 Bxe7+ Rxe7 21 Rf3+ Ke8 22 Qf7+ Kd7 23 Rxe7+
and:
a) 23 ... Qxe7 24 Bxd5 with a winning ending for White.
b) 23 ... Nxe7 24 Qe6+ Kc7 (or 24 ... Ke8 25 Rf8+! Kxf8 26 Qf7 mate) 25 Rf7 wins.
20 ... Bf6
Instead, 20 ... Qd7 (the queen unsuccessfully attempts to arbitrate the squabble) 21 Rf3+
Bf6 22 Ba3+ Re7 23 Qxd7 Bxd7 24 Bxd5 is an easy win for White.
21 Ba3+ Re7 22 Re4!
Threat: Rde3. There is no rush to capture e7.
22 ... Ke8
22 ... h6 23 Rde3! leaves Black helpless, since the defenders are as frozen in place as
department store mannequins.
23 Bxe7 Nxe7
23 ... Bxe7 24 Rf3! threatening Rf7 is decisive.
24 d5 Bb5
 Exercise (combination alert): Black’s reclusive king continues to inhabit his Great
Expectations room, where he broods over past wrongs. Unfortunately for him,
there are more wrongs waiting to emerge. White to play and force mate:

Answer: Clearance/pin/attraction/interference.
25 d6!
Tal callously avoids saving his d3-rook, the way a man steps around a sleeping
homeless person on the city pavement.
25 ... Bxd3 26 d7+!
This pyramidal social structure does an unexpected headstand, as the lowly d-pawn
orders Black’s queen to a fatal square. The pretty attraction/interference shot clogs the black
king’s access to d7.
26 ... Qxd7 27 Qg8 mate 1-0
“We are called to speak for the weak, for the voiceless, for victims of our nation and for
those it calls enemy,” declares Black’s king, who then is embarrassed to realize he just
accidentally quoted Martin Luther King. Poor NN must have been sadly shaking his head
and contemplating the chain of unfortunate events which led him here.

Game 16
J.Donner-M.Tal
Zurich 1959
Modern Benoni

1 d4 Nf6 2 c4 c5 3 d5 e6
My editor GM John Emms called the Benoni Tal’s “meat and potatoes” line, which he
played and understood better than anyone in the world, in his era. To my mind, a player who
willingly entered into Tal-territory like this is akin to the ancient sailor who decides to enter
uncharted waters, knowing full well that they may be infested with sea monsters. I have a
feeling Tal’s results would have fallen if all his opponents weaponized boredom and played
either the London or Colle System on him!
4 Nc3 exd5 5 cxd5 d6 6 Nf3 g6 7 e4 Bg7 8 Bd3 0-0 9 0-0
Today, 9 h3 is the prevailing interpretation of what is considered White’s best shot at an
advantage.
9 ... a6
The immediate 9 ... Bg4! is more accurate.
10 a4
White can fix his slightly inaccurate last move with the correct 10 h3!.
10 ... Bg4!

Question: What is so great about Tal’s last move?

Answer: My friend GM Yasser Seirawan – after rudely beating me in a Benoni –

revealed his Benoni principle: If Black achieves one swap of minor pieces, he equalizes;
two swaps and Black stands better. Theory, back in the 1950’s didn’t recognize this fact,
misinterpreting the correct plan, the way a book is misinterpreted from one language to
another, by an incompetent translator.
11 h3 Bxf3 12 Qxf3 Nbd7 13 Bf4
With 13 Qd1 Re8 14 Re1 Rc8 15 a5 c4 16 Bc2 White covers the b3-square, as well as
his e-pawn, A.Aleksandrov-J.Gdanski, Warsaw 2008. I prefer Black after 16 ... Rc5 17 Ra4
Rxa5 18 Rxc4 b5 19 Rd4 Ra1.
13 ... Qc7 14 Qe2
 Donner wants to suppress both ... c4 and ... b5 ideas, at the cost of placing his queen on
the same file as Black’s e8-rook.
14 ... Rfe8
Nobody is comfortable when a rook stares at our queen on the same file, even when the
queen is temporarily shielded by a pawn.
15 Bh2
Instead, 15 a5 c4!? (15 ... Rac8 is probably better) 16 Bxc4 Nh5 17 Be3 Bxc3 18 bxc3
Rxe4 19 Bd3 Re5 20 c4 Nf4 21 Qc2 Nxd3 22 Qxd3 Rc8 23 Rfc1 Nc5 24 Qc3 looks a bit
better for White, whose dark-square power outweighs his weak c-pawn and Black’s c5
outpost.
15 ... Rac8
Intending ... c4.
16 Bc4
White returns the bishop-pair to maintain the blockade of c4. On the down side, this
means yet another swap, which benefits Black. I think White was better off playing a line
like 16 Bc2 c4 17 a5.
16 ... Ne5 17 f4!?

White clings to his plan of an e5 break, and makes the extravagant assertion that he is the
one with the initiative, despite a growing body of evidence which suggests an opposite
view. His move is made with the philosophy: sometimes we must compromise structure to
kindle piece activity. A single cell amoeba is attracted to an unfamiliar source of light and
heat. Moral: living organisms are creatures of curiosity. It’s very difficult to hold back from
such an aggressive gesture. Sometimes we fight fiercely for some trivial point, just to prove
a point – in this case Donner tells Tal: “I plan to push you around.”

Question: From your tone, you disapprove of White’s thematic looking move.
 What is wrong with it? White simply begins a central/kingside attack.

Answer: How to draw the line between ‘not enough’ and ‘too much’? To my mind
White’s would-be kingside initiative/attack just barely ekes out its survival, like a
rattlesnake who needs to get lucky and find a careless mouse. I think the move weakens
more than generates attacking chances, and unnecessarily places White under future strategic
obligations of defending e4.
17 ... Nxc4 18 Qxc4 Nd7

Question: Who stands better?

Answer: I prefer Black, whose bishop is better than White’s.

Question: What about the fact that White controls the centre?

Answer: I think this factor is balanced by Black’s chances to overextend and undermine
with ideas like ... Qb6!, followed by ... Qb4!. Also, remember Seirawan’s Benoni principle.
Black achieved a swap of two pairs of pieces, which significantly reduces White’s chances
for a successful kingside assault.
19 Rfe1?!
White’s intended e5 break will never happen. He is better off playing 19 a5.
19 ... Qb6! 20 Rab1 Qb4!
This move breaks down White’s blockade of c4.
21 Qf1 c4
Just like that, White’s lifeless position feels as inert as the passed out drunk, still
clutching his drained bottle of rum. It felt like Donner made natural moves and is now
mysteriously strategically busted. Is the Benoni really just a forced win for Black in Tal’s
hands?
22 Re2

Exercise (critical decision): Analyse 22 ... Bxc3 and its aftermath. Is it worth it
for Black to give up his dark-squared bishop to damage White’s structure?

22 ... b5
Tal retains his advantage with this move, but he could have had more. Both parties are
guilty of dark-square bias of the Benoni bishop. White’s last move was a mistake and in this
case Black is justified in handing over the bishop.
Answer: Why is it that an idea’s actuality is sometimes so utterly alien from our deluded
romantic notions? 22 ... Bxc3! is a psychologically difficult move to make, since we all
worship that dark-squared Benoni bishop. We all observe and process data, and our
interpretation can vary wildly, from stylistic bias. In this instance, there is only one correct
answer. After 23 bxc3 Qxc3 24 e5 (24 Rxb7? only makes matters worse for White after 24
... Nc5 25 Rb1 Nxe4 26 Rd1 f5 when Black’s pieces rule the board) 24 ... Nc5 25 Rd1 f5!
26 Kh1 Qb3 Black’s c-pawn is about to run down the board and White is busted. If 27 exd6
Rxe2 28 Qxe2 Ne4 and Black dominates.
23 axb5 axb5 24 Kh1 Bxc3!
 It soon becomes clear that the bishop’s purposes are not so innocent. Now Tal sees the
correct idea, and White’s position convulses into a terminal level.
25 bxc3 Qxc3 26 Rxb5 Qd3!
Dual purpose: Black clears the way for his c-pawn to run down the board, while
attacking White’s e-pawn. Defensive preparation is not in White’s power anymore.
27 Qe1
White can offer slightly more resistance with 27 e5 Nc5, but the result is still not in
doubt.
27 ... c3 28 Rb1 Nc5 0-1
That was almost too easy! We all love it when our position runs on cruise control, where
the next move is as obvious as your last one. I risk belabouring the point when I mention that
White is just as busted now, as we was about seven moves ago. Donner, recognizing that his
pieces are an inherently dysfunctional unit, resigned, having seeing the line 29 Qd1 (29 e5
c2 30 Rc1 Nb3 wins) 29 ... Rxe4! 30 Qxd3 (30 Rc2 Re1+ 31 Qxe1 Qxc2 is also hopeless
for White) 30 ... Nxd3 31 Rxe4 (if you are a person of scant financial means, then you may
as well borrow money, since you don’t ever intend to pay it back) 31 ... c2 32 Rf1 c1Q 33
Rxc1 Rxc1+ 34 Bg1 and now Black forks with 34 ... Nf2+.

Game 17
M.Tal-S.Gligoric
Bled/Zagreb/Belgrade Candidates 1959
King’s Indian Defence

1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4 e4 d6 5 f3 0-0 6 Be3 e5 7 Nge2 c6 8 d5 cxd5 9 cxd5 a6 10

Qd2 Nbd7 11 g4 h5 12 h3 Nh7 13 h4?
We all make such careless moves, which are dishes that smell wonderful, yet when the
details come to light, are as devoid of taste as unsweetened oatmeal. To Tal, who decides to
test a shady hypothesis, the conducting of an attack was an inexact science, where sometimes
he would lazily just head in the general direction of his opponent’s king, without bothering
about details. In this case his last move constitutes a serious inaccuracy. White should play
either 13 Rg1 or 13 0-0-0.

Question: I don’t get it. Tal’s last move looks completely thematic to me.
Isn’t White playing for g4 almost a distinguishing characteristic of the line?

Answer: Yes, but g4 must be properly timed. White’s last move – at first what looks like
a no-brainer – upon closer inspection, turns out to be a brainer! Let’s turn your question into
an exercise:

Exercise (planning): Tal’s last move was a strategic inaccuracy. If you find
Black’s correct plan, you extract a serious concession from White. How?

Answer: Step one: swap on g4.

13 ... hxg4! 14 fxg4 Nhf6?
Gligoric flubs step two, missing a promising opportunity with 14 ... Nb6!. When one is
up to no good (and I mean Tal, of course!), the thing he or she fears most, is exposure to the
light. Black threatens both ... Bxg4 and also ... Nc4. White has only unpleasant options at
this point: 15 Ng3 (I’m almost certain Tal would have handed over his g-pawn, rather than
allow his dark-squared bishop to be swapped away with 15 g5 Nc4 16 Qc1 Nxe3 17 Qxe3,
which is also clearly better for Black) 15 ... Bxg4 16 h5 and White only has fishy
compensation for the pawn – which in Tal’s hands was still very, very dangerous!
15 Bh3
 Tal allows his opponent the same idea. He can avoid Black’s coming idea with 15 g5
Nh5 16 0-0-0.
15 ... Nb6
Strategic threat: ... Nc4. However, this time the move lost potency. When we see the
correct idea, but a move late, it’s a little like eating cheese a day past the expiration date –
probably still edible, but clearly past its prime.
16 Bg5!
This covers against both threats.
16 ... Nc4 17 Qd3!

17 ... Qc7!

Question: What was wrong with chopping on b2?

Answer: 17 ... Nxb2? allows White to greatly increase pressure on f6 with 18 Qf3! b5
19 h5! with a strong attack (White should avoid the greedy 19 Rf1?! Qa5! 20 Bxf6 Bxf6 21
Qxf6 b4 22 Nd1 Nd3+ 23 Kd2 Nc5 24 Qf3 Bd7 when Black’s attacking chances fully
compensate his missing piece).
18 b3 Na3 19 Rc1 Bxg4
At first, I thought this move was an error, which compromised the entire operation.
Having examined it closer, I’m not so sure. Gligoric’s move walks into Tal’s devious trap,
but his position stinks if he avoids the trap as well! For example:
a) 19 ... Nxg4 alertly detects Tal’s subterfuge, yet walks into other problems: 20 Nb1
Qb6! 21 Nxa3 Nf2 22 Qf3 Nxh3 23 Rxh3 Qb4+ (23 ... Bxh3?? is met with the zwischenzug
24 Nc4 Qb4+ 25 Bd2 Qc5 26 Qxh3 with a winning position for White) 24 Bd2 Qxa3 25
Rg3 Qxa2 26 Rc7.
b) 19 ... Qb6 (this insipid line may be Black’s best) 20 Qf3 Nh7 21 Be3 Qa5 22 0-0
Nb5 23 Na4 Nd4 24 Nxd4 exd4 25 Bf4 Qd8 26 Qg3 Nf6 27 Rce1 and White threatens
Bxd6, as well as Bg5, renewing the pin. Black looks to be in trouble in this line as well.
20 Bxf6! Bxe2
Forced.
21 Kxe2 Bxf6

Exercise (combination alert): Tal had foreseen this position

and planned a devilish trap which wins material. How?

Answer: Double attack/clearance.

22 Nb1! Qa5 23 b4!
This clearance/double attack shot wins Black’s knight. I don’t know about you, but I find
that I normally solve combinations when on a tactics website, and then go blind with the
same combinational pattern while playing a tournament game. I think the two have the
similar relationship of shooting at bull’s eyes at a gun range, when compared to shooting at
the enemy on the battlefield.
23 ... Qxb4 24 Nxa3 Bxh4
It’s not easy to keep our wits about us, after we walk into an opponent’s shot. When we
are forced into a concession, we become the punished third grader attempting to keep the
resentment out of his voice, as he answers his interrogating teacher. Black gets three pawns
for the piece – not so bad. His trouble comes not from the material side of the ledger, but
from a lack of king safety. So wealth in this position comes not from material, but from
initiative and attack, all owned by Tal.
25 Nc4
Tal understandably wants to bring the offside knight into play. The comp gives the line
25 Bf5 Bg5 26 Rb1 Qd4 27 Rbg1! Bf4 (or 27 ... Qxd3+ 28 Kxd3 Be7 29 Nc4 b5 30 Nb6
Ra7 31 Nc8! Rc7 32 Rc1! Rb7 33 Bg4 and White dominates) 28 Qxd4 exd4 29 Nc2 Be5 30
Kd3. Black’s d4-pawn falls and he only gets two pawns for the piece, rather than the three
he expected.
25 ... b5 26 Nb6!
The knight can’t be challenged easily and actually worms its way eventually towards
Black’s king.
26 ... Rad8 27 Bf5!
Targeting g6, while uncovering on Black’s bishop. Of course the bishop is immune due
to the mate threat.
27 ... Bg5
27 ... Bf6?? is met with 28 Nd7, winning material.
28 Rcg1 Qb2+ 29 Kf3 Bf4 30 Rg2 Qb4 31 Nd7!
The world tends to be a beautiful place, for one with a steadily growing bank account.
Tal threatens the f8-rook, which if moved to e8, allows a knight fork on f6.
31 ... Rc8
31 ... Kg7 is met with 32 Nxf8 Rxf8 33 Bg4.

Exercise (critical decision): Black threatens ... Rc3. White can

play either 32 Nf6+, or sacrifice with 32 Bxg6. Both methods
win, but one is far more efficient. Which one would you play?

32 Nf6+
Answer: This still wins, but immediately decisive is 32 Bxg6!! fxg6 (32 ... Rc3?? is
refuted by 33 Bf5+ Bg3 34 Rxg3 mate) 33 Rxg6+ Kf7 34 Rf6+ Kg7 (or 34 ... Ke7 35 Rh7+
Kd8 36 Nxf8 with mate to follow) 35 Rg1+ Kh8 36 Rxf8+ Rxf8 37 Nxf8, which leaves
Black down a rook with no reason to play on.
 32 ... Kg7 33 Nh5+ Kg8 34 Bxc8 Rxc8 35 Rc2
White is forced to return a piece, since ... Rc3 had to be dealt with.
35 ... Rxc2 36 Qxc2 Qa3+!
The only way to continue playing. 36 ... gxh5?? allows White to force mate after 37
Qc8+ Kh7 38 Qf5+ Kg7 39 Rg1+ Kf8 40 Qc8+ Ke7 41 Qc7+ Kf6 42 Qd8. When the queen
sees any display of kindness, she wants it stopped.
37 Qb3 Qxb3+ 38 axb3 gxh5

Exercise (critical decision): Is 39 Rxh5 White’s

best move? If not, then what should he play?

Answer: Black’s a-pawn is infinitely more valuable than his h-pawn.

39 Ra1!
Tal’s move is far more accurate than 39 Rxh5? a5 40 Rh1 Bd2, after which Black can
put up a fight.
39 ... h4 40 Kg4 Be3 41 Rxa6 Bc5
Black must protect his base on d6.
42 Kxh4 f5
The circles of Black’s downward spiral grow tighter and tighter. This is, of course,
desperation. Gligoric’s newly minted e-pawn isn’t going anywhere.
43 exf5 e4 44 b4!?
This isn’t really necessary, but as we all know, Tal loved to sacrifice – effectively or
not! Easily winning is 44 Kg5 Kf7 45 f6 e3 46 Rc6 e2 47 Rc7+ Kf8 48 Re7 and the e-pawn
is halted.
44 ... Bxb4 45 Kg5 Kf7
45 ... e3 46 Kg6 Kf8 47 f6 Ke8 48 f7+ Ke7 49 Ra8 forces mate.
 46 Ra7+ Ke8 47 f6 1-0

There are no grounds for hope. 47 ... Bc5 48 Rb7 b4 49 f7+ Kf8 50 Kg6 and there is no
remedy to the coming check on the back rank.

Game 18
M.Tal-R.Fischer
Bled/Zagreb/Belgrade Candidates 1959
King’s Indian Defence

When a tactician and a strategist meet, it becomes a battle between magic and science. Tal
dominated Fischer in the late 1950’s when he was in his prime, and when Fischer hadn’t
reached his peak yet.

1 d4!?
When it comes to our openings, it’s human nature to never be satisfied with what we
know right now. An inner voice keeps whispering into our ears: “More!”.

Question: 1 d4 was unusual for Tal. Why did he decline

to open with 1 e4 and face down Bobby’s Najdorf?

Answer: Tal was the clearly stronger player in 1959, but not in the opening. Fischer
knew the subtleties of the Najdorf better, and Tal wanted to test Fischer’s King’s Indian
instead. If I were Tal’s coach, I would have advised him to stick with 1 e4!, since Tal’s
‘expertise’ with 1 d4 was vestigial – meaning a few misplayed games in his teens!
1 ... Nf6 2 c4 g6 3 Nc3 Bg7 4 e4 d6 5 Be2 0-0 6 Nf3 e5 7 d5
Tal enters Petrosian’s system of the KID.
 7 ... Nbd7
7 ... a5 is played more often today.
8 Bg5
If White gets time for Nd2 and g4, he secures a bind.
8 ... h6 9 Bh4 a6

Intending ... Qe8, ... Nh7 and ... f5. This move was book at the time the game was
played, while today, this system is considered a tad too slow for Black, and Fischer’s move
would probably be accompanied with a ‘?!’ mark. Today, Black usually goes for a line like
9 ... g5 10 Bg3 Nh5 11 h4 g4 12 Nh2 Nxg3 13 fxg3 h5 14 0-0. Black’s control over the dark
squares compensates for the weakening of f5 and his dark squares, V.Kramnik-G.Kasparov,
Linares 1994.
10 Nd2
Denying Black a ... g5 and ... Nh5 option, since now h5 is covered.
10 ... Qe8 11 0-0
Tal banks on his queenside chances and decides to castle short, giving Fischer a clear
target of attack. Also playable are 11 b4; and 11 g4.
11 ... Nh7
Making way for ... f5.
12 b4 Ng5
 Question: This looks awkward. Why not just play 12 ... f5 immediately?

Answer: The move is premature. White responds with 13 exf5! and Black must
recapture with his rook, losing time and giving White absolute control over e4. If Black
plays 13 ... gxf5? he loses material to 14 Bh5!.
13 f3
Supporting e4 and giving his dark-squared bishop air on f2. In the same tournament,
Petrosian played 13 Rc1 f5 14 f3 Qe7 15 Kh1 Nf6 16 c5! Nh5 17 c6 b6 18 exf5 gxf5 19 g3!
(threat: f4) 19 ... Bf6 20 f4 Ng7 21 Nc4 exf4 22 gxf4 b5 23 Nd2 Ne4 24 Bxf6 Rxf6 25 Bf3
a5 26 a3 axb4 27 axb4 Rg6 28 Ndxe4 fxe4 29 Bxe4 and White secured a winning position,
T.Petrosian-S.Gligoric, Bled/Zagreb/Belgrade 1959.
13 ... f5 14 Bf2
Taking a bead on the c5-square.
14 ... Qe7
Theory ends here, and I feel that White’s queenside chances eclipse Black’s kingside
attacking chances.
15 Rc1
White intends to apply intense pressure to Black’s queenside with a coming c5 and Nc4.
15 ... Nf6 16 c5
White achieves his thematic pawn break.
16 ... Bd7 17 Qc2
Covering e4, thereby enabling Nc4. Also promising is 17 exf5 gxf5 18 Nc4.
17 ... Nh5
Somehow I get the feeling that this attacking show is representational of Fischer’s wish,
rather than his objective ability to generate an actual attack. After 17 ... fxe4 18 fxe4 Ng4 19
Bxg4 Bxg4 20 Nc4 Black’s pieces on g4 and g5 are somewhat in the way of his kingside
attack, while White’s queenside attack proceeds smoothly.
18 b5!?

When we give up an addiction, what we don’t lose is our desire to return to it. This is a
typical Tal decision, where he is quick to pick a fight, impatiently striking, rather than
building. His last move slightly throws off kilter his position’s delicate settings. Objectively
stronger is 18 Kh1 Nf4 19 Bc4 (intending c6) 19 ... Kh8 20 Be3 with strong queenside
pressure for White.
18 ... fxe4
After 18 ... dxc5 19 bxa6 b6 20 Bc4 (threat: d6+) 20 ... Kh8 21 a4 Nf4 22 a5 White’s
attack is faster than Black’s kingside counterplay.
19 Ndxe4 Nxe4 20 fxe4
The alternative was 20 Nxe4 Bxb5 21 Bxb5 axb5 22 cxd6 cxd6 23 Rb1 Qd7 24 Rb3
Nf4 25 Qd2 g5, intending ... g4. It feels to me like Black equalized.
20 ... Nf4 21 c6
The queenside opens, but at the same time, Fischer begins to generate kingside play.
21 ... Qg5
Do you ever get that creepy feeling when you are alone, that you are being watched?
Well, White’s king feels that way right now. Black threatens both mate on g2, and also keeps
open the option of ... Nh3+ and ... Nxf2, picking off White’s supporter of the dark squares.
22 Bf3
Tal frets over defence of g2, as if over a feverish baby.
22 ... bxc6 23 dxc6?!
This move allows Black’s kingside play to flare. Correct was the counter-intuitive
zwischenzug 23 h4!!. We are taught to avoid making weakening pawn moves around our
king, yet chess is full of such anomalies. After 23 ... Nh3+ 24 Kh2 Qe7 25 Bg3! Nf4 26 dxc6
Be8 27 a4 Bf6 28 Bxf4! exf4 29 Nd5 Qd8 30 b6! Bd4 (threatening mate on h4; if 30 ... Bxh4
31 Kg1 Bg3 32 Nxc7 Qh4 33 Rfd1 and Black’s attack is too slow) 31 Kh3! cxb6 32 Bg4!
suddenly c7 is coming and Black is busted.
23 ... Bg4!

Going after White’s best defender. Black stands no worse here. Fischer correctly
avoided 23 ... Nh3+? 24 Kh1 Nxf2+ 25 Qxf2 Be6 26 b6 cxb6 27 Qxb6. Black’s kingside
attack is no more, while White’s passed c-pawn decisively ties Black down.
24 Bxg4?!
Tal was never comfortable on defence. He should play 24 Be3 axb5! 25 Kh1 Ra3.
Black’s pieces generate tremendous activity and he stands no worse.
24 ... Qxg4 25 Be3 axb5 26 Bxf4!?
Tal expels the intruding knight from his boundaries, at the cost of handing Fischer control
over the dark squares, and adding Black’s f-pawn as a potential attacker.
26 ... exf4 27 Nxb5
 Exercise (critical decision): Black stands at the threshold
of either disaster or revelation. He can defend c7 with 27 ... Rf7,
or he can sacrifice the pawn with 27 ... f3. Which one would you play?

27 ... Rf7?!
More military campaigns have been doomed to failure due to a faulty supply line, than to
defeat on the battlefield. This is no time to hedge or equivocate. He should forget about
defence of c7.
Answer: Black desperately needs to replenish troop levels on the kingside with 27 ...
f3!. In such positions, we should be gripped by an attacking fever. Now if 28 Nxc7? Rab8
(threat: ... Rb2) 29 Rf2 (29 Rb1?? is met with 29 ... Rxb1 30 Rxb1 Be5! 31 Nd5 Bxh2+! 32
Kxh2 Qh4+ 33 Kg1 f2+ 34 Kf1 Qh1+ 35 Ke2 f1Q+ and Black wins) 29 ... Rb2 30 Qc4+
Kh8 31 g3 Rxf2 32 Kxf2 Qh5! and White is in deep trouble, since 33 h4?? is met with 33 ...
Qa5! with deadly threats on d2 and c7.
28 Qc4!
This move disables Tal’s defensive apprehensions. Now all is secure in White’s
position, since he prevents f3, while threatening Nxc7.
28 ... Rc8?!
Now his counterplay is destroyed and what remains is merely an inferior copy of the
original intent. Fischer, resigning himself to a bleak destiny, goes passive and begins to lose
the thread of the position, as the initiative swings to White. Black’s rook needs to remain on
the a-file to try and slow White’s passed a-pawn. Therefore necessary was 28 ... Kf8!.
29 Rf3!
No more ... f3 tricks, rendering Black’s attack into a hastily drawn portrait, abandoned
before completion.
 29 ... Be5 30 Rcf1?!
Tal gets distracted by g3 ideas. He stands clearly better after 30 a4!.
30 ... Kg7 31 a4
At last.
31 ... Ra8
An admission that his earlier ... Rc8 was inaccurate.
32 Kh1 Qg5?
Correct was 32 ... Qh5! intending ... g5 and g4. The game should end in a repetition
draw after 33 Rh3 Qg4 34 Rhf3 Qh5.

Exercise (combination alert): Black’s last move was

an error. How did Tal seize a decisive initiative?

Answer: Pin. Black lacks access to ... g5 and now he gets worked over, based on the
weakness of f7.
33 g3! Raf8 34 gxf4 Bxf4?
This move enhances the pin on the f-file. After 34 ... Rxf4?? 35 Nxc7 a mega fork on e6
can’t be prevented, so Fischer had to try 34 ... Qg4.
35 Nd4!
Now the monster e6 fork is unavoidable.
35 ... Qh4
After 35 ... Re7 36 Ne6+ Rxe6 37 Qxe6 Qe5 38 Qxe5+ dxe5 39 a5 Black survives the
attack, only to enter a lost ending. White’s a-pawn easily secures the win.
36 Rxf4!
A gifted storyteller stretches out the tale to simultaneously enthral and tease his
audience. Simpler was 36 Ne6+ Kh7 37 Rxf4 and wins, since 37 ... Rxf4 is met with 38
Nxf8+ Kh8 39 Rxf4 Qxf4 40 Nxg6+.
36 ... Rxf4 37 Ne6+ Kh8
37 ... Kh7 38 Nxf8+ transposes to the above line.

Exercise (combination alert): How did Tal finish the game?

38 Qd4+!
Zwischenzug/pin. Black’s king is forced to self-pin.
38 ... R8f6
Also losing are:
a) 38 ... Kg8 39 Qg7 mate.
b) 38 ... R4f6 39 Nxf8.
c) 38 ... Qf6 39 Rxf4 Qxd4 40 Rxf8+ Kh7 41 Nxd4 leaves White up a rook and a knight.
39 Nxf4
Threat: Nxg6+.
39 ... Kh7
Black threatens both ... Rxf4 and also ... g5.
40 e5!
A decisive line opening, where Tal punches a gaping hole through the barrier, while
conveniently protecting his knight.
40 ... dxe5
40 ... Rf7 41 exd6 g5 42 Qd3+ Kg8 43 dxc7 is also hopeless for Black.
41 Qd7+ 1-0
Fischer accepts the terms of surrender, refusing to engage in a hopeless fight after 41 ...
Kg8 42 Qd8+. (“Your stupidity almost absolves you from your crimes ... almost,” says the
queen to her brother) 42 ... Kf7 43 Qxc7+ Kg8 44 Qd8+ Kf7 (the Talking Heads would
agree: “Same as it ever was”; Black’s king commutes back and forth, from g8 to f7, on a
regular basis) 45 Qd7+ Kg8 (now the king’s infirmities confine him to his home) 46 c7 and
White promotes.

Game 19
M.Tal-V.Smyslov
Bled/Zagreb/Belgrade Candidates 1959
Caro-Kann Defence

“This game was seemingly a paradox: one of the greatest masters of his time was unable to
withstand a bluff – a semi-correct, risky attack, arising literally out of nothing!” wrote
Kasparov of this game.

1 e4 c6 2 d3!?

Question: Tal wasn’t a closed games player. Why would

he pick such a line against a strategist like Smyslov?

Answer: My guess is that Tal didn’t want to play his normal line 2 d4 d5 3 Nc3 dxe4 4
Nxe4 Nd7, since this line is named after Smyslov, who pioneered it and undoubtedly
understood its intricacies better than Tal.
2 ... d5 3 Nd2 e5 4 Ngf3 Nd7?!
Today, this move is known to be an inaccuracy. Correct is 4 ... Bd6!, and if 5 d4 exd4 6
exd5 cxd5 7 Nxd4 Nc6 8 N2f3 Nge7 is a position which can be reached too via a Tarrasch
French move order.
5 d4!
 Now the opening divides into a separate sub-species. Tal gets his open game after all,
and a favourable one at that, since he leads in development.
5 ... dxe4

Question: Why can’t Black play 5 ... exd4, just as you gave in the above note?

Answer: The problem is Black committed his knight to d7 prematurely and is misplaced
for a Tarrasch French formation after 6 exd5 cxd5 7 Nxd4 when Black gets an inferior
version of the Tarrasch French, due to the d7-knight placement.
6 Nxe4 exd4!?
This move violates the principle: Avoid breaking the central tension without good
reason.
Black may be better off going into emergency mode with 6 ... Ngf6 7 Nxf6+ Qxf6 8 Bg5
Qe6, J.Hickman-N.Hutchinson, Gibraltar 2006. 9 dxe5 can be met with 9 ... f6 10 Bf4 fxe5
11 Bg3 Be7 when Black, although still worse, is better off than in the game’s continuation.
Instead, 6 ... f5?! intending ... e4, next is overly optimistic: 7 Neg5 e4 8 Bc4! Nh6 (8 ...
exf3? 9 Bf7+ Ke7 10 0-0! gives White a winning attack) 9 Ne6 is quite awful for Black.
7 Qxd4

Black is in serious trouble:

1. He lags behind in development in an open position.
2. White soon menaces Nd6+, which gives him the bishop-pair in the open position.
7 ... Ngf6 8 Bg5 Be7 9 0-0-0?!
A vow made over 1,000 times (i.e. I’m going to lose weight!) tends to diffuse its
potency. Tal always went with the emotion-based response, when the position demands a
Mr. Spock-like response, free from stylistic bias. This move, which glides away from the
topic at hand, throws away the advantage. Once again, we see Tal’s weakness crop up: He
wants too much from a position and ends up with nothing. He gains a serious advantage with
the continuation you, me and every other player in the world would make with 9 Nd6+ Bxd6
10 Qxd6 Qe7+. Here lies the rub. An ending, Tal’s least favourite part of the game!
However, after 11 Qxe7+ Kxe7 12 0-0-0 White owns the bishop-pair and a development
lead in the ending. Black will be hard pressed to hold the game.

Question: Why would Tal avoid such an obviously favourable line?

Answer: Sometimes we are inclined to discount the data, simply because we don’t like
the conclusion to which it points. The problem is Tal sometimes irrationally avoided
endings – even good ones, if he could get a messy middlegame position instead! Of course,
Tal would argue: not every position is conducive to a straightforward application of logic,
and it makes sense to play to one’s strengths, rather than enter the opponent’s domain. And
it’s hard to argue with success. Maybe Tal just found technical endings boring, which is a
danger, since boredom’s presence tends to spur us into a dangerous enlivening mechanism.
9 ... 0-0 10 Nd6 Qa5
10 ... Nb6 11 Nxc8 Qxc8 12 Re1 Re8 13 Bd3 Nbd5 also looks okay for Black.
11 Bc4 b5
Both parties feel they are the ones entitled to ownership of the initiative. This is a little
reminder to Tal that his own king may soon come under attack via a pawn storm.
12 Bd2!
 Question: Why did Tal retreat his bishop?

Answer: He had two ideas in mind with the retreat:

1. Tal plans to switch the bishops to c3, where it takes aim at Black’s king.
2. Tal vacates g5 for a knight to launch his kingside attack.
12 ... Qa6!?
Smyslov goes for the full point! 12 ... Qb6 should lead to a pretty forcing draw after 13
Nf5 Re8 (13 ... Bc5 14 Qf4 bxc4 15 Bc3 Bb4 16 Qg5 g6 17 Nh6+ Kg7 18 Nf5+ Kg8 19
Nh6+ is drawn) 14 Rhe1 Bc5 15 Rxe8+ Nxe8 16 Bxf7+! Kxf7 17 Ng5+ Kg8 18 Bb4!! Bxd4
19 Ne7+ Kf8 (19 ... Kh8?? 20 Nf7 mate) 20 Ng6+ Kg8 21 Ne7+ with perpetual check.
13 Nf5! Bd8!?
Ambitious. Smyslov continues to play for the full point.

Question: Doesn’t 13 ... Bc5 win a piece?

Answer: It wins a piece, but not the game, which ends in a draw after 14 Qf4 bxc4 15
Bc3 Re8 16 Qg5 Bf8 17 Nh6+ Kh8 18 Nxf7+ Kg8 19 Nh6+.
14 Qh4!
 The previous mild tremors begin to turn into violent eruptions. In emergency situations,
we prioritize to immediacy. One idea is free-associated with its next step. When my son was
in pre-school, he would constantly come home with a sheet of construction paper, with glued
uncooked pasta in strange patterns only he understood. I would invariably think: “Italian
food sounds good tonight.” Tal finds the only viable move, where he sacrifices a bishop for
a single tempo, rather than:
a) 14 Bb3? c5 15 Qh4 c4 16 Qg5 Nh5! 17 Qxh5 cxb3 18 a3 bxc2 19 Rde1 Nf6 20 Qg5
Bxf5 21 Qxf5 g6 and White’s king faces serious trouble, while Black’s remains safe.
b) 14 Bd3? Qxa2, and if 15 Bc3?? c5 16 Qh4 b4! 17 Bxf6 Nxf6 when White is busted.
14 ... bxc4 15 Qg5
Queen and knight give g7 a significant look.
15 ... Nh5!
Not:
a) 15 ... Ne8? 16 Qxd8 Qxa2 17 Bc3! (Tal said he intended this move, rather than to go
for the favourable ending after 17 Ne7+ Kh8 18 Qa5! Qxa5 19 Bxa5 when Black struggles
in the ending, despite his (temporary) extra pawn) 17 ... Ndf6 (17 ... Nef6?? 18 Rxd7! Bxd7
19 Nh6+! Kh8 20 Qxf6! forces mate) 18 Ne7+ Kh8 19 Qa5 (the best move, although Tal may
have opted for the more complex 19 Rhe1 with advantage there, as well) 19 ... Qxa5 20
Bxa5 with the same good ending for White.
b) 15 ... g6 16 Bc3 Qxa2 17 Nh6+ Kg7 (17 ... Kh8?? loses instantly to 18 Rxd7) 18
Nf5+ is once again drawn.
16 Nh6+ Kh8 17 Qxh5 Qxa2
Not 17 ... gxh6?? 18 Bc3+ f6 19 Qxh6 winning, since 19 ... Rf7 20 Rhe1 Black is mated.
18 Bc3
Control over the kingside dark squares is a necessary precondition for White’s attack.
18 ... Nf6??
 Black’s defence, once thought unimpeachable, is found to have a giant hole in it. I
learned from long and painful experience that positional players who attempt to out-combine
gifted tacticians are like the audience at a magic show, determined to discover the
magician’s sleight-of-hand, yet usually fail to do so.
Correct was 18 ... Bf6! 19 Ng5! Bxg5+ 20 Qxg5 f6 21 Qg3 Nc5! (21 ... gxh6?? 22 Rhe1
leaves Black’s king helpless against the coming assault) 22 Nf7+ Kg8 23 Nh6+ is once
again perpetual check.

Exercise (combination alert): Smyslov just blundered on his last move,

mixing up a ‘must do’ with a ‘must not do’. How did Tal respond?

Answer: Queen sacrifice/weak back rank/smothered mate.

19 Qxf7!
That which was once just conjecture, now forms into absolute certainty. Black has no
saving ideas. This loss must have been frustrating for Smyslov, since his opponent’s
randomly crazy ideas become all the more infuriating when they happen to work!
19 ... Qa1+
19 ... Rxf7?? 20 Rxd8+ is mate in two moves.
20 Kd2 Rxf7
Otherwise:
a) 20 ... Qxd1+ 21 Rxd1 Rxf7 22 Nxf7+ Kg8 (the king’s constant weeping doesn’t
inspire confidence in his defenders) 23 Nxd8 leaves White up a piece.
b) 20 ... Re8?? (grief can make us do crazy things) 21 Qg8+! Nxg8 22 Nf7 is mate.
21 Nxf7+ Kg8 22 Rxa1 Kxf7
White is up a clean exchange in the ending and Smyslov could have resigned here.
23 Ne5+ Ke6 24 Nxc6 Ne4+ 25 Ke3 Bb6+ 26 Bd4! 1-0
More accurate than 26 Kxe4 Bb7 when Black can play on for a while. Tal deservedly won
the brilliancy prize for this game.
Chapter Two
World Champion and 1960-1970
In this chapter, Tal emerges as world champion, having defeated the legendary Mikhail
Botvinnik. He did the unthinkable, outplaying the Lord of Logic strategically, starting with
the first game of the match. “A genius will educate itself,” wrote Mary Wollstonecraft. And
Tal did just that. Before the match, ‘WWSD?’ was Tal’s guiding principle of chess, which
translates to: ‘What would Satan do?’

In the first chapter of the book, Tal deliberately steered for positions so geometrically
disorienting, that it felt to some of his opponents that they accidentally stumbled into the
middle of a fairy tale. But with the stakes so high, with a world championship title on the
line, we see Tal at his most well behaved (although he clearly misbehaved in some of the
games of the match!), uncharacteristically playing it coy, pretending to be a strategist.

Here Botvinnik’s (black) game lies in ruins, down a pawn (normally, Tal felt that the
taking of material was beneath his dignity if he sensed that he might instead deliver
checkmate), with punctured dark squares and a vulnerable king on d7. Then to show that it
wasn’t a fluke, Tal did the same thing over and over again, winning one strategic game after
another – but always with an undercurrent of deeply embedded tactical themes running
through it. Later in the match he even opened with 1 Nf3!? and entered a sleep-inducing Reti
Opening, outplaying his great rival strategically once again.
 Now a true positional player would have swapped queens here, but Tal still retained his
aversion to swaps of any kind and slid his queen back to e1. His quiet opening play was still
at odds with his sharp handling of the middlegame. He continued to squeeze Botvinnik and
emerged with the full point.
I don’t want to mislead the reader in to thinking Tal suddenly transformed into Ward
Cleaver. When he sensed that he was being outplayed, he reverted to his old con-artist self:

On his last move, Tal played ... Nf4!!??. The double exclam is for its psychological
shock value, while the double question mark is an objective evaluation of the sacrifice. Who
else but Tal would take a gamble this large when leading the match by a point? The game
erupted in complete chaos and Botvinnik, with his clock running low, was unable to
consolidate and lost.
Game 20
M.Tal-M.Botvinnik
World Championship (Game 1), Moscow 1960
French Defence

After this, the first game of the 1960 world championship match, Tal revealed that he was
shocked to win it, since he “had the charming habit of beginning a tournament with a
defeat!”.

1 e4 e6 2 d4 d5 3 Nc3 Bb4
Botvinnik’s specialty, the strategically unbalanced Winawer French.
4 e5 c5 5 a3 Bxc3+ 6 bxc3 Qc7
Today, the Poisoned Pawn line of the Winawer is reached via the move order 6 ... Ne7
(the death of an old idea gives birth to a new one) 7 Qg4 Qc7 (7 ... 0-0 is also a major line
which is nightmarishly complicated) 8 Qxg7 Rg8 9 Qxh7 cxd4 10 Ne2 Nbc6 11 f4 Bd7 12
Qd3 dxc3. This is a tabiya position, which sets the later stage where modern day GMs mine
their computers in search of a fashionable trifle, in their hope to discover a theoretical
novelty around the 26th move. In a sense, such positions are endless construction projects
which are never destined to reach completion. I remind my theory-obsessed students: when
everyone thinks alike, a single person (i.e. a writer of chess opening books!) represents the
entire community. But, of course, they never listen to me and just keep on studying
fashionable theory obsessively!
7 Qg4
Tal, of course, opts for White’s sharpest option, rejecting the more controlled 7 Nf3.
7 ... f5!?
In Botvinnik’s era, this was the book move. 7 ... Ne7 8 Qxg7 transposes to the modern
version of the Poisoned Pawn line.
 8 Qg3
Covering against ... cxd4 and ... Qc3+.

Question: After 8 exf6 Nxf6 who is favoured? White has an

open position for his bishops, while Black leads in development.

Answer: Principle: A development lead tends to outweigh bishop-pair in open

positions. White scores just under 40% with this line and Black already stands better.
8 ... Ne7?!
Now Black gets an inferior version of the Poisoned Pawn line.

Question: What is the difference between this way and the modern main line?

Answer: Two differences, both which favour White:

1. In this inferior version, f5 is blocked with a pawn, rather than open for a black piece.
2. Having played ... f5, Black is unable to implement a future ... f6 prying mechanism.
Black should play 8 ... cxd4! 9 cxd4 Ne7! (not 9 ... Qxc2? 10 Bd2 Qc7 11 Rc1 Nc6 12
Bb5 a6 13 Bxc6+ – principle: opposite-coloured bishops favour the attacker, which is
obviously White, in this case – 13 ... bxc6 14 Ne2 and Black’s extra pawn provides little
solace for her wretched position, full of dark-square punctures and grossly lagging
development, A.Bodnaruk-A.Balaian, Belgrade 2013) 10 Bd2 0-0 11 Bd3 b6, intending ...
Ba6, is okay for Black.
Another good path is 8 ... Nc6 9 Nf3 cxd4 10 cxd4 and now is the time for 10 ... Nge7
11 Rb1 (11 Qxg7? Rg8 12 Qxh7 Nxd4! is in Black’s favour) 11 ... 0-0. In this version Black
gets some play down the c-file, and also has the plan of ... Na5 and ... Nc4, occupying c4,
S.Karjakin-V.Ivanchuk, Monaco (rapid) 2011.
 9 Qxg7!

Tal willingly enters the maze, even when facing a theoretically better prepared
opponent. In this instance he walks headlong into Botvinnik’s feared opening preparation. Of
course, today, when an opponent walks into our prep, it’s our comps who defeat them. So to
us, the game then has a wonderfully larcenous feel of a student who cheated on an exam and
got away with an A+ grade. Believe it or not, Botvinnik was probably unprepared for this
move, since it wasn’t played at the time of the game.
Tal’s choice happens to be White’s best move, and a clear improvement over Smyslov’s
safe choice, 9 Bd2 (White’s normal response at the time) 9 ... 0-0 10 Bd3 b6! (preparing to
swap off his bad bishop via the a6-square) 11 Nh3 Ba6 12 Nf4 Qd7 13 h4 Bxd3 14 cxd3
Nbc6 15 Be3 cxd4 16 cxd4 Rac8 and Black’s chances are no worse, since his healthy
queenside pawn majority compensates for White’s kingside attacking chances, V.Smyslov-
M.Botvinnik, Moscow 1957.
9 ... Rg8 10 Qxh7 cxd4 11 Kd1
Tal’s move covers against Black’s threat to check on c3 and pick off the a1-rook. I
would go for 11 Qh5+! Ng6 12 Ne2 Nc6 13 f4 Bd7 14 Qf3 (14 cxd4?? is met with 14 ...
Nxd4! when White’s knight must watch over a queen check on c3) 14 ... dxc3 15 Qxc3 0-0-0
16 Be3 and Black’s position is inferior to a normal Poisoned Pawn French, since his knight
is misplaced on g6, and he lacks an ... f6 prying mechanism, J.Palsson-H.Camilleri,
correspondence 1997.
11 ... Bd7!
Botvinnik’s prepared novelty. The bishop eyes the a4-square, where it takes a bead on
c2.
12 Qh5+ Ng6
Tal was off form in the 1961 rematch, due to health concerns. The following win was
one of his happier days, in an otherwise miserable rematch: 12 ... Kd8!? (Botvinnik’s
attempted improvement, which perhaps took Tal by surprise; Black’s king voluntarily
forgoes castling rights to avoid the irritating pin of his knight on g6) 13 Nf3?! (after 13 cxd4!
Qc3 14 Rb1 Qxd4+ 15 Bd2 Qxf2 16 Nf3 Nbc6 17 Rxb7 White stood clearly better, due to
control over the dark squares, the safer king and passed h-pawn, V.Baklanov-R.Kwiecinski,
correspondence 2008) 13 ... Qxc3 14 Ra2 (awkward but necessary, as 14 Rb1?! Ba4 is a
problem for White, since he is unable to play 15 Bd3?? due to 15 ... Qxd3+) 14 ... Nbc6 15
Rb2 Kc7 16 Rb5?! (Tal sets up a trap, but he should actually play 16 Rb3).

Now Black stands clearly better after 16 ... Qa1!. Instead, play proceeded 16 ... Rh8? 17
Qxh8! Rxh8 18 Bb2 (White regains the queen) 18 ... Qxf3+ 19 gxf3 Ng6? (Black’s chances
are no worse after 19 ... Rh4! 20 Rb3 Nxe5 when he gets a pawn for the exchange, and
formidable centre, while White must deal with the issue of multiple isolanis) 20 h4! (the
passed h-pawn is a big issue for Black) 20 ... Ngxe5 (or 20 ... Rxh4 21 Rxh4 Nxh4 22 f4
Ng6 23 Bc1 and Black is busted in the ending) 21 h5 Nf7?! (Botvinnik should try 21 ...
Nxf3) 22 f4 (Black’s centre is not allowed to roll forward) 22 ... Nd6 23 Rb3! (going after
Black’s d4-pawn) 23 ... Ne4 24 Ke1 Rh6 25 Be2 Be8 26 Rd3 Nf6 27 Bxd4 Nxd4 28 Rxd4
Bxh5?! 29 Rd3! Rh7 30 Rdh3 Bg6 31 Rxh7+ Nxh7 32 Rh6! Nf8 33 Rh8! (Tal finds a forcing
way to win the e6-pawn, after which Black’s game collapses) 33 ... Nd7 34 Rg8! Bf7 35
Rg7 Be8 36 Re7 and game over, since e6 falls. Botvinnik resigned in a few moves, M.Tal-
M.Botvinnik, World Championship (Game 12), Moscow 1961.
Returning to the earlier 12 ... Ng6:
13 Ne2
13 Rb1! looks like an improvement for White: 13 ... d3 (13 ... Nc6 14 Nf3 dxc3 15 Bg5
is also in White’s favour) 14 Bxd3 Nc6 15 f4 0-0-0 16 Qe2 and Black lacked full
compensation for his two missing pawns, P.Roth-F.Winiwarter, Austrian League 1992.
13 ... d3!?
 Played after 37 minutes of thought! When a strategist like Botvinnik suddenly begins to
act erratically (and when I say “erratically”, I mean act like Tal!), inviting complications, he
becomes a person leading a double life, like Walter White, the mild mannered high school
Chemistry teacher/meth dealer, in Breaking Bad.
Such a sudden shift in structure is an indication of the Poisoned Pawn Winawer’s
mutability. Black’s last move was made with the thought: strip the enemy king of his
concealment and you also degrade his protection. The study of chess leads us to a
knowledge of our inner nature. Botvinnik’s was one of always wanting to be in charge, no
matter the cost. He refuses to accept his role as passive bystander and decides upon
precipitous action. Tal gave this move an exclam in his book on the match, which I don’t
believe is merited. I just don’t see this as a remunerative investment. When we engage in a
shaky sacrifice of material, our ‘compensation’ can feel like we lost an item, and then it
returns to us in a dream.
As to Botvinnik’s actual motivation for the sacrifice, I can only dimly imagine. He
tended to shy away from any variation where he lost the initiative, rejecting a line like 13 ...
dxc3 14 Nf4 Kf7 15 g4 Qxe5 16 Nxg6 Rxg6 17 Rg1! with attacking chances and the bishop-
pair for White. Of course, in this one, Black up a pawn, rather than down one, as in the game
continuation. It’s very easy for Black to go wrong here. For example, 17 ... Qd4+? (17 ...
Qf6 is correct) 18 Bd3 Qxf2?? (Black attacks, when he should be defending) 19 Rg3 (threat:
gxf5) 19 ... Kg7 20 Be3 and Black’s queen finds herself trapped.
Botvinnik probably originally intended 13 ... Ba4?, but then saw the response 14 Nf4
Qf7 15 Nxg6 Rxg6 16 Rb1! dxc3 (16 ... b6 17 cxd4 is a strategically won game for White,
who is just up two pawns for nothing) 17 Rxb7! (deflection) 17 ... Qxb7 18 Qxg6+ Qf7 19
Qh6 when White is up a pawn, owns the dark squares, and his h-pawn will soon roll down
the board. Black probably can’t generate enough against White’s king to compensate these
factors.
 14 cxd3 Ba4+ 15 Ke1 Qxe5 16 Bg5!
Tal ensures that Black’s king won’t castle long to safety.
16 ... Nc6 17 d4
Preventing ... d4 ideas, while ensuring his dark-square domination, by fixing Black’s
central pawns on light squares.
17 ... Qc7 18 h4!

Question: How would you assess?

Answer: Black is strategically lost for the following reasons:

1. Black is down a pawn.
2. White’s passed h-pawn anchors his g5-bishop, while remaining a looming threat in
the ending.
3. White rules the dark squares, which outweigh Black’s light-square control.
4. White’s king looks safer than Black’s.
5. White’s h1-rook may now enter the game by way of h3.
18 ... e5!?
Keep in mind that sometimes despair mimics defiance. The comps don’t like this move,
but just because a comp disapproves of an idea empirically, doesn’t mean it won’t work on
a confused human. When our plan is based on an imaginary foundation, the end result never
fits reality’s requirements. In this instance the ignition switch which starts Black’s initiative
is non-existent. It’s infuriating when we recognize our plan’s outline, yet lack knowledge of
the exact mechanism which sets it into motion.
19 Rh3!
One incongruity piles upon another. Tal activates his sleeping h1-rook in unorthodox
fashion. Maybe Botvinnik hoped for the rash 19 dxe5? Ncxe5 20 Rh3 Rh8!, forcing the
awkward 21 Bh6 0-0-0 22 Qg5 Rde8 23 h5 Rxh6! 24 Qxh6 Nf4! 25 Qxf4 Nd3+ 26 Rxd3
Qxf4 27 Rd4 Qh6 28 Rxa4 Qc6!. White is busted, since he must return a full rook, due to the
dual threats ... Qxa4 and ... Qxc3+.
19 ... Qf7!?
We tend to get irritated when our ‘threats’ are ignored, because it is a gesture of the
opponent’s disdain. Botvinnik allows the centre to open, not wanting to be ground down
slowly in a line like 19 ... e4 20 Rb1 Qf7 21 Nf4 b6 22 Nxg6 Rxg6 23 Ba6 Kd7 24 Kf1 Rf8
25 Bf4 Bc2 26 Rb2 Ba4 27 Qe2. White has firm control over the game, while Black remains
counterplayless.
20 dxe5!
Tal picks the correct moment. Principle: Open the position when you own the bishop-
pair.
20 ... Ncxe5
The cost of Black’s decision is twofold:
1. He further opened the game, to the benefit of White’s bishops.
2. Black condemned himself to long hours of servitude to the defence of his now
chronically weakened d5-pawn.
21 Re3
Every white piece is in play. Tal centralizes his rook, threatening f4. The area around
Tal’s king is well policed and his king is secure, despite its central location.
21 ... Kd7 22 Rb1 b6?!
The weakening of a6 plays into multiple variations later on for White. Correct was the
non-weakening 22 ... Bc6.
23 Nf4 Rae8 24 Rb4!
Gaining a tempo, while adding another attacker/defender into the mix. Tal’s rooks
display a marked bias for the lateral. Notice how his pieces operate on the weakened dark
squares. Komodo likes the inhuman continuation 24 Qe2! (intending Qa6) 24 ... Nc4 25
Nxd5!! Nxe3 26 Nxe3 Qg7 (26 ... f4?? loses to 27 Qa6) 27 Rc1 Re4 28 g3 when Bg2 is
coming and Black is quite lost.
24 ... Bc6 25 Qd1!
The queen, having done her job on h5, returns to d1 to pressure d5. Oddly enough, white
king and queen return to their squares of origin, after wandering away earlier, in search of
adventure. All of White’s pieces breathe activity and work in harmony, including the
undeveloped f1-bishop.
25 ... Nxf4
After 25 ... Ng4 26 Rxe8 Rxe8+ 27 Be2 Black has no way to improve his position and
remains lost.
26 Rxf4 Ng6 27 Rd4 Rxe3+
Alternatively, 27 ... f4 28 Rxe8 Rxe8+ 29 Kd2! Kc8 30 h5 Qf5 31 Ba6+ Kb8 32 Bh6
Ne5 33 Bxf4 when a second pawn falls and Black can resign.
28 fxe3
Now Black has no chance of a future ... f4 or ... d4 break.
 28 ... Kc7

Exercise (combination alert): Find one strong move and Black’s game collapses:

Answer: Pin/undermining. Black can’t afford to take on c4, but then again, he can’t
afford not to take it either.
29 c4! dxc4!?

Question: Doesn’t this just lose an exchange?

Answer: When we are losing, it’s hard to shake the feeling that we are a character in a
tragedy, whose end has already been pre-written. Botvinnik theorizes: a gangrenous limb
must be amputated to save the rest of the body. It does lose the exchange, and technically the
move is an error – but not really if everything else loses. For example, 29 ... Kb7 30 cxd5
Bd7 31 a4 when White is up two pawns, while ruling the board, with the bishop-pair,
passed h-pawn and the safer king.
30 Bxc4
The love-smitten bishop will do anything to move closer to the adored presence of the
black queen.
30 ... Qg7 31 Bxg8 Qxg8 32 h5! 1-0
The wispy slip of a pawn, who is the culprit of the commotion in Black’s camp, decides
the game, since White seizes control over f4. The fact that Tal produced a strategic
masterpiece is probably one of those signs of End Times, mentioned in scriptures of all
religions. 32 ... Nf8 33 Bf4+ Kb7 34 Rd8 Qxg2 35 Rb8+ Ka6 36 Qe2+! (once queens come
off the board, Black’s resources border on emaciation) 36 ... Qxe2+ 37 Kxe2 and the passed
h-pawn decides the game’s outcome.
Game 21
M.Botvinnik-M.Tal
World Championship (Game 6), Moscow 1960
King’s Indian Defence

1 c4 Nf6 2 Nf3 g6 3 g3 Bg7 4 Bg2 0-0 5 d4 d6 6 Nc3 Nbd7

In those days, playing ... Nbd7 and ... e5 was Black’s only plan against the g3 King’s
Indian. Today, Black’s alternative set-ups are:
a) 6 ... Nc6 7 0-0 a6 8 h3 Rb8.
b) 6 ... c6 7 0-0 Qa5.
c) 6 ... c5 7 0-0 Nc6 8 d5 Na5.
7 0-0 e5 8 e4 c6 9 h3 Qb6
The queen applies pressure to d4, hoping to resolve central pawn tension. Tal threatens
the cheapo ... exd4, followed by the discovered attack ... Nxe4, winning a pawn.
10 d5
Karpov beat both John Nunn and Judit Polgar with 10 c5 dxc5 11 dxe5 Ne8 12 e6 fxe6
13 Ng5, with compensation for the pawn.
10 ... cxd5
Later 10 ... Nc5 11 Qe2 Qa6 12 Bg5 Na4 13 Nxa4 Qxa4 was seen in R.Hübner-
A.Shirov, German League 1992. White looks just a shade better after 14 b3.
11 cxd5 Nc5 12 Ne1

Once again, Botvinnik is better prepared than Tal in the opening. Tal wrote that he was
surprised by this move, and had only seen 12 Re1 and 12 Nd2.

Question: What is Botvinnik’s idea behind this retreat?

 Answer: He prepares to challenge Black’s well posted knight with Nd3, later on.
Instead, after 12 Qe2 Bd7 13 Be3 Rfc8 14 Rfc1 Qa6 15 Qxa6 bxa6 Black hopes to generate
counterplay down the open b-file, which partially compensates his broken structure,
A.Grischuk-B.Jobava, Khanty-Mansiysk 2009.
12 ... Bd7
Connecting his rooks.
13 Nd3
Tal criticized Botvinnik, not for making this thematic and obvious move, but for
unnecessarily burning up ten minutes on his clock before making the move. One
psychological problem which plagues all positional players is that we hate surprises, and
tend to blow time on our clock quadruple checking our analysis, instead of just making the
move immediately and trusting to fate.
13 ... Nxd3 14 Qxd3 Rfc8!?

This move indicates that Black plans to play on the queenside, which is traditionally
White’s realm in KID structures.

Question: Why wouldn’t he play for ... Nh5, ... f5 and a kingside attack?

Answer: Tal wrote that he didn’t like the look of 14 ... Nh5 15 Be3. Now he only looked
at 15 ... Qd8 16 Qe2 which prevents Black from playing ... f5; exf5 and now ... gxf5 is
impossible, since Black’s knight hangs. This would force ... Bxf5, leaving White in control
over e4. Instead of playing the queen to e2, White has 16 Nb5!, double attacking a7 and d6,
which pretty much forces 16 ... Bxb5 17 Qxb5 when White has both space and the bishop-
pair.
15 Rb1
 Preparing Be3. 15 Qe2 is slightly more accurate, since White’s queen later moves to e2
anyway, so White may save a tempo by avoiding the unnecessary Rb1.
15 ... Nh5!?

Normally, we don’t play on two fronts, but Tal is guided his own rules. Black plays for
an ... f5 break after all.
16 Be3 Qb4!
Keeping an eye on the e4-pawn, while preparing to double rooks with ... Rc4 and ...
Rac8. Tal wouldn’t even consider a passive (not to mention weak) line like 16 ... Qd8? 17
Nb5 Bxb5 18 Qxb5 when White controls all strategic trumps in the position.
17 Qe2
So that if Black plays ... f5; exf5, Tal won’t be allowed to recapture on f5 with a pawn.
17 Qd1! is an interesting new plan, which seems to suppress Black’s counterplay, after
17 ... Rc4 18 Kh2 Rac8 19 Bf3! (not only chasing the h5-knight back, but also preparing
Be2) 19 ... Nf6 20 a3 Qa5, K.Panczyk-A.Wojtkiewicz, Czestochowa 1992. White achieves a
queenside bind, while suppressing Black’s kingside play after 21 Be2 R4c7 22 Bd2 Qb6 23
Bd3.
17 ... Rc4
Tal avoids the strategically suspect line 17 ... f5?! 18 exf5 Bxf5 19 Rbc1 Nf6 20 g4!
(White seizes control over e4) 20 ... Bd7 21 f4! with the initiative for White.
18 Rfc1 Rac8
 Back in the late 1970’s (or was it the early 1980’s?), when your writer was a young, thin
and devilishly handsome lad, rated 2198, I drew with GM Jim Tarjan by copying Tal’s
opening play in this game, but with reversed colours. I still remember the look of pain on
Jim’s face when he agreed to the draw, which indicated the well justified thought: ‘How
could I draw with this clueless fish?’
19 Kh2
Perhaps preparing Bf3.
19 ... f5
Otherwise Black has no useful plan.
20 exf5 Bxf5 21 Ra1
Now what? White owns e4, Black’s queenside initiative attempt looks like it’s going
nowhere, and his h5-knight languishes out of play. In fact, it looks as if Black’s stagnant
position remains within the grips of activity-deprivation, and Tal’s non-initiative reached its
feeble zenith, and has no place to go but downward.
21 ... Nf4!!??
 A natural optimist never chooses the lesser evil. Tal sometimes played as if his guesses
were actual facts. The merchant’s motto: everyone needs something, and it’s our job to
maximize what they will pay for it. If the reward is large enough, there are some of us who
willingly stake their lives on the gamble, and with his last move, Tal did just that.

Question: Didn’t you just say: “and his h5-knight languishes out of play?”

Answer: I freely admit that misstatements may have been made by someone, somewhere.
I will have to check my notes and get back to you on that one. The power of boredom
actually produces a kind of energy in the one bored. You don’t believe me? Then go to the
zoo and watch the neurotically pacing tiger or bear in its enclosure, in its frantic search for
non-existent stimuli. My intuition tells me that Tal was a person who easily got bored and
made such moves for the adrenaline thrill, and also for that delightful look of outraged shock
on his opponent’s face. This is one of the most jarring moves ever played in a world
championship match (only Fischer’s 11 ... Nh5!? against Spassky’s Benoni can match it for
shock value).

Question: You are not going to sit there and

tell me that Tal’s sacrifice is sound, are you?

Answer: Of course it’s unsound, but this is just how Tal played – even with a world
championship on the line. Yet there are multiple hidden benefits within the idea. Tal’s move
accomplishes the following:
1. Tal sweeps Botvinnik away from a strategic game – a position in which he excelled –
into a gnarled mess – the type of game where Tal (who hated dry positions devoid of highs
and lows) was the unquestioned ruler.
 2. The position may be winning for White if Tal was playing a computer, but in this case
it’s not at all easy for a human to navigate the complex lines laid out before him.
3. After such a move, Botvinnik is a virtual certain candidate for time pressure,
increasing Tal’s practical chances even further.
4. Apparently the spectators in the auditorium erupted so loudly after Tal’s move, that
the game actually had to be moved to an inner room to avoid disturbing the players’
concentration. I’m certain this factor flustered Botvinnik psychologically, more than Tal,
since Tal got the audience on his side with his bold sacrifice.
22 gxf4 exf4

23 Bd2?
Tal’s gamble immediately pays off. Correct was the zwischenzug 23 a3! Qb3 24 Bxa7!
Ra8 (24 ... b6 is met with 25 a4!, and if 25 ... Ra8? 26 Ra3 Qb4 27 Na2! Qxb2 28 Qxb2
Bxb2 29 Rxc4 Bxa3 30 Bxb6 when White is up a piece in the ending) 25 Nb5! Rxc1 26
Rxc1 Be5 27 Nd4! Qa4 (this looks great for Black, since he threatens the a7-bishop, d4-
knight and also ... f3 check; White’s next move dispels the illusion) 28 Nf3! (White returns
the piece and gets a winning position) 28 ... Rxa7 29 Nxe5 Qe8 30 Rc7 Ra8 (or 30 ... dxe5
31 d6 Bd7 32 Bd5+ Kf8 33 Rxd7! Qxd7 34 Qxe5 Qg7 35 Qxf4+ Ke8 36 Qe3+ and Black’s
rook falls) 31 Rxb7 Qxe5 32 Qxe5 dxe5 33 Re7 when Black is busted in the ending.
23 ... Qxb2?
The poor tend to waste nothing. Tal blunders right back! Correct is 23 ... Be5! (which
Tal actually wrote down on his score sheet and then inexplicably rejected), threatening ...
f3+ and winning White’s queen. When a crime is too big for a single criminal, what she
needs is an accomplice. 24 f3 Qxb2 25 Nd1 Qxa1 26 Rxa1 Bxa1 leads to a strategically
favourable situation for Black, whose two rooks and three pawns are worth more than
White’s queen and piece.
24 Rab1 f3!?
 24 ... Bxb1 25 Rxb1 Qc2 26 Nb5 Be5 (26 ... Qxb1 27 Nxd6 favours White as well) is
met with the shot 27 Nxd6! Bxd6 28 Qe6+ Kg7 29 Rxb7+ R8c7 30 Rxc7+ Bxc7 31 Qe7+
Kg8 32 d6 Qxd2 33 dxc7 Rxc7 34 Qxc7 Qxf2 35 Qc4+ Kg7 36 Qc3+ when White has all
the winning chances.

Exercise (critical decision): We arrive at a moment of live-or-die

importance. Tal does everything in his power to confuse. Botvinnik can
play 25 Rxb2, or 25 Bxf3. One line leads to a winning position for White,
while the other gives Black a won game. Which one would you play?

25 Rxb2?
An accounting error tends to be fatal in long calculations. The old man loses his footing,
like a once nimble cat (I have a feeling Botvinnik in his prime, circa 1948, would have
survived this assault, and out-calculated and beaten Tal in this exact position), now losing
his footing. Botvinnik settles on a false view. With a defence this intertwined, the corruption
of a single element compromises the integrity of the whole. The correct plan floats in
corporeality and confusion. Botvinnik, who understandably wants to remove queens from
the board, goes astray.
Answer: White can still secure a won game with 25 Bxf3! Bxb1 26 Rxb1 Qc2 27 Rc1
Qf5 (after 27 ... Qb2 28 Bg4! Be5+ 29 Kg2 R8c7 30 Nd1! Qd4 31 Be6+ Kg7 32 Rxc4 Qxc4
33 Qe3! White’s queen threatens to infiltrate h6, and if 33 ... Qh4 34 f4 Bf6 35 Nf2 when
Ne4 is coming; White consolidated and his two pieces are worth a lot more than Black’s
rook and pawn) 28 Bg4 Qe5+ 29 Qxe5 (so queens come off in this line as well) 29 ...
Bxe5+ 30 f4 (defenders, like busy ants in a colony, scuttle about, each with their individual
duties) 30 ... Rxf4 31 Bxc8 Rf2+ 32 Kg1 Rxd2 33 Be6+ Kg7 34 Ne4 and White has good
chances to consolidate his extra piece.
 25 ... fxe2 26 Rb3

Exercise (combination alert): White is hopelessly tied down to the defence of c3. Tal
had envisioned this position and saw a multi-tiered combination. Try and find it:

Answer: Pawn promotion.

26 ... Rd4!
“From now on, all your prayers will be silent ones,” says the rook to the bishop. When it
comes to insurgents, for each one seen, there are 100 more hiding in the shadows of the
populace.
Step 1: Chase White’s bishop to e1.
27 Be1
No choice since 27 Be3?? is met with the crushing 27 ... Rxc3 28 Rbxc3 Rd1.
27 ... Be5+!
Step 2: Manoeuvre the dark-squared bishop to f4, via e5, where it chases White’s c1-
rook to a tactically unfavourable square.
28 Kg1 Bf4! 29 Nxe2!
The only way to play on. If Botvinnik had played 29 Ra1 then Step 3 follows with 29 ...
Rxc3! (deflection) 30 Rxc3 Rd1 and White loses heavy material.
29 ... Rxc1 30 Nxd4 Rxe1+ 31 Bf1 Be4
 In a losing position, our torment is thoughts of how we could have done a hundred things
differently in the past, yet are left without hope or options in the present. In the
combination’s aftermath, Black is up a pawn, with the bishop-pair in an open position.
Conclusion: White is completely busted.
32 Ne2
32 Rxb7?? is met with 32 ... Bd3.
32 ... Be5 33 f4 Bf6 34 Rxb7 Bxd5
Menacing ... Bc4.
35 Rc7
Covering the c4-square.
35 ... Bxa2!
Either Black wins a second pawn, or White allows ... Bc4.
36 Rxa7 Bc4
Tal isn’t going to fall for 36 ... Rxe2?? (intending ... Bd4+ and ... Bxa7), which is met
with the zwischenzug 37 Ra8+ Kg7 38 Bxe2.
37 Ra8+ Kf7 38 Ra7+ Ke6 39 Ra3
Botvinnik saves his piece with the Re3+ trick, but not the game. All Black needs to do
now is to push his d-pawn.
39 ... d5
 40 Kf2 Bh4+
Nyet! The king fails to chase away Black’s annoying rook.
41 Kg2 Kd6
Black’s king gets out of the way of potential knight checks on d4.
42 Ng3 Bxg3 43 Bxc4 dxc4 44 Kxg3 Kd5
It becomes obvious that nothing will stop Black’s passed c-pawn.
45 Ra7 c3 46 Rc7 Kd4 47 Rd7+ 0-1
Botvinnik sealed this move but didn’t resume the adjourned game, since his position was
well past saving.

Game 22
M.Tal-M.Botvinnik
World Championship (Game 7), Moscow 1960
Caro-Kann Defence

1 e4 c6 2 d4 d5 3 Nc3 dxe4 4 Nxe4 Bf5 5 Ng3 Bg6 6 N1e2

 Question: Why did Tal develop his knight to the passive e2-square,
where it clogged his light-squared bishop’s development?

Answer: Tal plays for a knight sacrifice on e6, which actually occurred in another game
in the match.
6 ... Nd7
Alternatives:
a) 6 ... Nf6 7 h4 h6 8 Nf4 Bh7 9 Bc4 e6 10 0-0 Bd6 11 Nxe6!? and here it is! While
reality has its good points, fantasy is so much more fun: 11 ... fxe6 12 Bxe6 Qc7 13 Re1
Nbd7 14 Bg8+ Kf8 15 Bxh7 Rxh7 16 Nf5 g6!! 17 Bxh6+ Kg8 18 Nxd6 Qxd6 19 Bg5 Re7
20 Qd3 Kg7 saw Botvinnik manage to survive the assault and go on to win an ending, where
his extra piece proved to be superior to Tal’s pawns, M.Tal-M.Botvinnik, World
Championship (Game 9), Moscow 1960. This game is annotated in Botvinnik: Move by
Move (of course, this cherished classic should be ordered immediately, read and re-read!).
b) Today, Black normally plays the move order 6 ... e6 7 Nf4 Bd6 8 Bc4 Nf6, which is
considered fine for Black. White has no sound method to sacrifice on e6.
7 h4
Threat: h5.
7 ... h6 8 Nf4
Gaining a tempo on Black’s bishop.
8 ... Bh7 9 Bc4 e5!?
This equalizes, yet as everyone understands: to allow an open game against Tal, is a
dangerous undertaking.

Question: After the continuation 9 ... e6 10 Qe2 Ngf6

 and if 11 Bxe6!? how would Black survive this assault?

Answer: White’s sacrifice is semi-sound at best. Black wrests the advantage the
following way: 11 ... fxe6 12 Nxe6 Qe7 13 Nc7+ Kf7 14 Nxa8 Qxe2+ 15 Nxe2 Nd5 16 c3
Bd6. Black picks up the stray a8-knight and his two minor pieces will be worth more than
White’s rook and two pawns.
Instead, 9 ... Ngf6 10 0-0 Bg8?! was tried by Bronstein, but Black has no need for such
anti-e6-sacrifice contortions.
10 Qe2 Qe7
The queen only temporarily blocks the f8-bishop.
11 dxe5
A forced move, which unravels Black’s development and allows the f8-bishop to move.
11 ... Qxe5 12 Be3 Bc5
Neutralizing White’s activity and more importantly, removing Tal’s queen from the
board, which is an achievement in itself.
13 Bxc5
A move made with deep regret, which relegates White’s would-be attack to oblivion.
Tal recognizes that keeping queens on the board comes at too high a structural cost, after 13
0-0-0?! Bxe3+ 14 fxe3 Ngf6 and I don’t see compensation for White’s weakened structure.
13 ... Qxe2+ 14 Kxe2 Nxc5 15 Rhe1 Nf6
 As was usually the case, Botvinnik came better prepared in the opening phase and
claims full equality. The position is both equal and boring. Such drab, symmetrical
landscapes tended to alarm Tal, who craved complications the way an addict does his drug
of choice.
16 b4!?
Peace is only the outcome when both sides refuse to fight. Sometimes hints of our future
intent escape, without our being aware of it in the present. Tal just couldn’t help himself.

Question: Doesn’t this just weaken White’s structure, with no perceptible gain?

Answer: As usual, the strategic investment in Tal’s pursuit of initiative overtakes all
other considerations. This kick of the knight also leads to a weakening of White’s queenside
structure. We do sense a somethingness in the position, when previously there was the
looming, drawish void of nothingness.
16 Kf1+ Kf8 17 Re2 was the safer route. Black stands not an iota worse after 17 b4!?
Nce4 18 Nxe4 Bxe4 19 Bd3 Bxd3+ 20 Nxd3 g5 21 Re5 gxh4 22 Rae1 Re8 23 Rxe8+ Nxe8
and the players agreed to a draw, M.Tal-G.Ilivitzki, Riga 1955).
16 ... Ncd7
16 ... Nce4? this time fails to 17 Kf3!, winning material.
17 Kf1+ Kf8 18 Bb3 g5!
Botvinnik shoos away White’s knight, while opening activity for his h8-rook and also
giving his king a haven on g7.
19 hxg5 hxg5 20 Nh3 Rg8 21 Red1?!
Tal said he refused 21 Rad1 due to 21 ... Re8 with a near certain draw. However, a
draw is not such a bad result if you stand a shade worse – which I think is the case here with
White.
 21 ... a5! 22 bxa5
White must agree to the structural damage, since 22 a3?? hangs a pawn for nothing after
22 ... a4 23 Ba2 Bxc2.
22 ... Rxa5 23 Rd6
Increasing pressure on both d7- and f6-knights, whose connection to the other may
constitute a tactical disadvantage.
23 ... Ke7 24 Rad1 Re5
24 ... Bg6! is more accurate, after which I think Black stands a shade better in the
ending, due to White’s broken queenside structure.
25 Nh5!
Tal continues to inflict little stabs of distress, despite the lack of counterplay for both
sides. When we set up a trap, we manipulate the opponent’s psychological need to believe
in his own good fortune. If the very act of setting up a cheapo is a morally reprehensible act,
then why is it that we all delight in doing so?
25 ... Bg6?!
Natural positional players, in our hearts, believe that if we evade a tactician’s tricks, we
will either draw or win the game. Botvinnik, lulled by the position’s quiet appearance lets
down his guard, allowing Tal a tactic, although technically, Black’s move isn’t an error,
since he should hold the draw after allowing Tal’s combination. However, the psychological
damage may have cost Botvinnik this game.
25 ... Ra8! prevents Tal’s coming combination, and leaves Black slightly better.

Exercise (combination alert): Most traps are nothing more than elaborately staged
con-jobs. This one is not so elaborate. How did Tal get two minor pieces for the rook?

26 Rxd7+!
 Answer: Attraction/knight fork. Magicians rely on sleight-of-hand, while we chess
players rely on slight-of-mind. Tal was sneaky, even in the most sedate looking positions.
This simple shot may have left Botvinnik crimson-faced and shaken his confidence into one
of those Hamlet-like ‘How-could-I-overlook-something-so-simple?’ internal monologues.
Botvinnik said after the game that he actually saw Tal’s combination and allowed it, since he
hallucinated, forgetting his g8-rook was left hanging.
26 ... Nxd7 27 Rxd7+ Kxd7 28 Nf6+
The knight flies in like a newly released magician’s dove.
28 ... Kd6
“An explanation would be much appreciated,” comments Black’s bewildered king.
29 Nxg8
Tal’s combination isn’t the end of the story, since Botvinnik manages to generate serious
queenside counterplay with his next move. So one careless move pushed Botvinnik into that
scariest of destinations, known as: Who-knows-where?
29 ... Rc5!
Botvinnik correctly sees White’s damaged queenside as a rich source of counterplay,
and Black actually stands no worse here.
30 Nh6
Attacking f7.
30 ... f6!?
30 ... Ke7! is Black’s best path to forcing a draw. For example, 31 c3 b5 32 Ke2 Rxc3
33 Nxg5 c5 34 Bxf7 Bxf7 35 Nhxf7 Ra3 36 Ne5 Kf6 37 Ngf3 (37 f4 Kf5 38 Nd3 c4 39 Nb4
Kxf4 is also drawn) 37 ... Kf5 38 Nd3 Rxa2+ 39 Ke3 Ra3 40 Ke2 Ra2+ 41 Ke3 and a
draw.
31 Ng4 Bxc2
White stands clearly better after the passive 31 ... Ke7?! 32 Ne3, after which c2 is
covered.
32 Nxf6 Bxb3?!
After this inaccuracy, Black struggles, yet still should hold a draw. Instead, with 32 ...
Bf5! 33 g4 Ke5! 34 gxf5 Kxf6 35 Be6 Rc3! 36 Kg2 c5 37 f4 gxf4 38 Nxf4 Ke5 39 Ng6+ Kf6
Black holds the draw.
33 axb3 Rb5 34 Nxg5!
Once again, Tal relies on tactics, this time to eliminate Black’s final kingside pawn.
34 ... Rxb3?
A critical error, after which Black is lost. 34 ... Ke5! 35 Nh5 Rxb3 36 g4 Rb4 37 Nf7+
(37 f3 c5 38 Ke2 b5 39 Kd3 Ra4 40 f4+ Rxf4 41 Nxf4 Kxf4 42 Ne6+ is drawn) 37 ... Ke4
38 Ng3+ Kd5 39 g5 Rg4 40 Nf5 Ke6 41 Ne3 Ra4 42 Nd8+ Ke5 43 Nxb7 Kf4 44 Nc5 (or
44 g6 Ra1+ 45 Kg2 Ra8 and after ... Rg8, Black picks up White’s g-pawn) 44 ... Ra1+ 45
Kg2 Kxg5 and the game is drawn, since in many future scenarios, Black can sacrifice his
rook for White’s final pawn and hold the drawn with lone king versus two knights.
35 f4
 Question: So it becomes a race between white and black
passers. I have read that knights tend to be notoriously clumsy
against a lone rook in such situations. Whose side is faster?

Answer: Observation of data is not a substitute for intuition, which is capable of

discovering a solution, outside the range of the conscious, calculating mind. The comps tell
us that White is winning, with perfect play. However, for us humans, ‘perfect’ is an ideal.
It’s far from easy to coordinate those scattered knights into an effectively unified whole. At
the board, Tal must have just had a sense that he could pull tricks with his two knights to
speed his pawns forward, while slowing Black’s queenside passed pawns.
35 ... Rb1+
I’m not so sure about this move, which activates White’s king. Botvinnik undoubtedly
wanted to get his rook behind Tal’s passed pawns (I was about to write ‘passers’, but just
got a group email from IM Dr. Tony Saidy, where he went on a lengthy lecture about how
writing the word “passer”, rather than “passed pawn”, is just another example of how our
world is degenerating!), while clearing the way for his own passed pawns.
After 35 ... b5 36 Nfe4+ Kd5 37 f5 Ra3 38 f6 Ra7 39 Nf2! b4 40 f7 Ra8 41 Nh7 c5 42
f8Q Rxf8 43 Nxf8 c4 44 Nd1 Kd4 45 Nd7 b3 (or 45 ... c3 46 Nxc3 and wins) 46 Ke2 c3 47
Nc5! b2 48 Nxc3! Kxc3 49 Na4+ White eliminates all of Black’s passed pawns, to emerge
into a won king and pawn ending.
36 Ke2 Rb2+ 37 Kf3 Rb3+
Botvinnik chases White’s king away from his own passed pawns.
38 Kg4
After 38 Ke4?? Rg3 White’s g-pawn hangs and the game is drawn.
38 ... Rb2 39 g3 b5 40 Nfe4+
 When we continually generate threats, forcing defenders to scramble and reshuffle, then
odds are, they fall out of synch at some stage. This dual purpose move clears the path for
White’s f-pawn to roll forward, and also puts the e4-knight into a sacrificing position
against Black’s surging passers ... oops, I mean passed pawns!
40 ... Kd5
This is not the time to go passive with a move like 40 ... Ke7?? 41 f5 b4 42 f6+ Kf8 43
Kh5 Rb1 44 Kg6 Rf1 45 Ne6+ Ke8 46 Nf4! b3 47 f7+ Kf8 48 Kf6 b2 49 Nc5 b1Q 50 Nce6
mate.
41 f5 b4 42 f6 Ra2 43 f7 Ra8

Exercise (combination alert): Tal found a combinational

solution to his promotion problem. What would you play here?

Answer: Interference.
44 Nh7! b3
“Are we there yet?” I would ask repeatedly, on long trips as a kid. “We get there when
we get there,” my father would respond each time. 44 ... Kxe4?? is met with 45 Nf6+ Ke3
46 Ne8!. The final barrier is easily battered aside, and the knight’s interference allows
White promotion.
45 Nd2 b2
Black’s b-pawn nears its goal, vacant-eyed and haggard.
 Exercise (critical decision): White has a choice of two candidate
moves: he can immediately promote, winning Black’s rook with
46 f8Q, or he can play 46 Kf3, sliding his king closer to Black’s surging
passers before he promotes. Only one of the moves wins. Which one?

Answer: 46 Kf3!
Tal sees through Botvinnik’s devious plot, which runs 46 f8Q?? Rxf8 47 Nxf8 Kd4 48
Kf3 Kd3 49 Nb1 Kc2 50 Na3+ Kb3 51 Nb1 Kc2 and Black forces the draw.
46 ... Kd4 47 Ke2 c5 48 f8Q
Now is the correct timing.
48 ... Rxf8 49 Nxf8 c4 50 Ne6+ Kd5 51 Nf4+ Kd4 52 Nb1! 1-0
 Now the queening battle is between a seven foot tall Viking chieftain and Frodo
Baggins. Botvinnik realizes that his odds are not so good when on heads, he loses, and on
tails, he loses as well. After 52 ... c3 53 Kd1 Ke3 54 Nxc3 Kf3 55 Nfe2 White hangs on to
his last pawn, while halting Black’s queening attempt.

Game 23
M.Tal-M.Botvinnik
World Championship (Game 11), Moscow 1960
Grünfeld Defence

1 Nf3!? Nf6 2 g3 g6 3 Bg2 Bg7 4 0-0 0-0 5 c4 c6

Botvinnik wisely attempts to alter the game into a Symmetrical Grünfeld, where Tal
experiences difficulties generating complications.
6 b3!?
Tal avoids the dull line 6 d4 d5 7 cxd5 cxd5. In a chess game, the participants are
notionally at war, yet it certainly doesn’t feel like it just yet. Positions like this are as odious
to the tactician, as the Poisoned Pawn Najdorf is to the quiet strategist.
6 ... Ne4
Forcing White to play d4, which avoids the more complex d3 Reti lines. 6 ... d6 7 d4
leads to a harmless line of the King’s Indian for White. The trouble was Botvinnik was no
fan of the black side of a KID – equal or not.
7 d4 d5 8 Bb2 Be6!?
 Question: Isn’t f5 the more natural square for this bishop?

Answer: True, but Botvinnik develops the bishop to e6 to resolve the central tension in
an attempt to make the game duller still. Tal criticized Botvinnik’s last move, but I like it on
a psychological level, since it attempts to snuff out White’s desire to keep his central
formation fluid.
More natural would be 8 ... Bf5 9 Nbd2 Nd7 10 Nh4 Nxd2 11 Qxd2 Be6 12 e4 dxe4 13
Bxe4. I don’t think that White’s central space means much, since Black’s position remains
weakness free, H.Nakamura-P.Leko, London 2012.
8 ... a5, intending ... a4 at some point, is also thought to be an equalizer here: 9 Nc3 Bf5
10 Rc1 Nxc3 11 Bxc3 Be4! and White hasn’t extracted much from the opening, A.Grischuk-
T.Radjabov, Baku 2014.
9 Nbd2 Nxd2?!
To me, this natural move represents a concession by Black, who took three moves to
swap away a white piece which moved only once. It feels like Botvinnik is trying too hard
to simplify, rather than just play chess. Better was 9 ... f5!, entering a hybrid Dutch
formation, which I think Botvinnik would have played on a lower-rated opponent – and
should have played on Tal. It’s important to remain faithful to your style. After 10 e3 Nd7
Black looks fully equal to me, A.Kharitonov-E.Bareev, Sochi 1987.
10 Qxd2!
 Advantage White. Botvinnik may have expected 10 Nxd2?! c5! after which Black stands
no worse.
10 ... Na6!
The knight is offside on a6 and slightly irrelevant when manoeuvred to c7. Yet this is
Black’s best move. Botvinnik avoided a trio of unfavourable lines:
a) 10 ... Nd7? is unpleasantly met with 11 Ng5.
b) 10 ... c5? fails to 11 dxc5 Bxb2 12 Qxb2 dxc4 13 Nd4! when Black is in serious
trouble.

Question: Why can’t Black win a pawn with 10 ... dxc4?

Answer: White responds with 11 Ng5 Bh6 (after 11 ... Bd5 12 e4 h6 13 exd5 hxg5 14
bxc4 cxd5 15 cxd5 White owns the bishop-pair, g5 is loose, and his d-pawns cramp Black’s
position) 12 Nxe6 Bxd2 13 Nxd8 Rxd8 14 bxc4, regaining the pawn, with the bishop-pair
and a strong pawn centre.
11 Rac1 Qd6
Botvinnik connects his rooks.
12 Ne5
Tal provokes ... f6.
12 ... Rfd8
12 ... f6?! simply weakens Black’s position after 13 Nd3!.
13 Rfd1 Rac8 14 Qa5!
Threat: Ba3. White’s move also opens up tactical possibilities based on Black’s a6-
knight, as shown in the note below. White can also try 14 c5 Qc7 15 h4, gaining space on
both sides of the board.
14 ... dxc4!?
 A slight strategic concession. Instead:
a) 14 ... Qb4?! walks into Tal’s dirty tactical idea: 15 Nxc6! bxc6 16 Qxa6 dxc4 17 Ba3
Qb6 18 Qxb6 axb6 19 bxc4 Bxd4 20 Bxe7 Rd7 21 Bg5 with an extra pawn for White.
b) Possibly Black’s best line may be 14 ... Bxe5 15 dxe5 Qb4 16 Bc3! Qxa5 17 Bxa5 b6
18 cxd5 Bxd5 19 Bh3 Be6 20 Bxe6 fxe6 21 Rxd8+ Rxd8 22 Be1 c5 23 f4 when I must refer
the reader to Tartakower’s quip about the worst bishop being superior to the best knight.
White looks better to me in the ending, despite Black’s healthy queenside structure.
15 Nxc4
The correct recapture. After 15 bxc4?! Bxe5! 16 Ba3 Qc7 17 Qxe5 Qxe5 18 dxe5
Rxd1+ 19 Rxd1 Kf8 20 f4 Ke8 Black’s healthy queenside majority makes up for White’s
bishop-pair.
15 ... Qc7
Alternatively, 15 ... Qb4 16 Ba3! Qxa5 17 Nxa5 Rc7 18 e3 when White owns the greater
share of the centre and also applies mild pressure to Black’s queenside.
16 Qe1!?
A glimpse into Tal’s style. He prefers the favourable middlegame to a favourable ending
after 16 Qxc7 Rxc7 17 Na5.
16 ... Qb8 17 e4

A new plan emerges for White: to enforce a d5 break.

17 ... Bxc4!?

Question: Why did Botvinnik hand Tal the bishop-pair?

Answer: 17 ... Nc7 18 Qe3 Nb5 19 h4 looks unpleasant for Black, who remains without
counterplay and can only await events.
18 Rxc4
 Probably the more accurate recapture, since 18 bxc4 c5 allows Black to puncture
White’s central integrity.
18 ... Nc7
18 ... c5?! is premature: 19 e5! cxd4 20 Rcxd4 Rxd4 21 Bxd4 looks unpleasant for
Black, since the e5-pawn cramps him, and the game opened for White’s bishop-pair.
19 Bh3
Tal forces ... e6, slightly weakening Botvinnik’s dark squares. I prefer 19 h4 first, and
only then Bh3. In this way White can play to loosen the pawn front around Black’s king with
h5 later on.
19 ... e6 20 Bc1!

Question: Why undevelop?

Answer: The bishop wasn’t well placed on b2, since White then must constantly watch
out for ... c5 tricks. Now Black must worry about both Bf4 and Bg5.
20 ... Qa8!
Achieving Reti’s dream.

Question: What?

Answer: Moves like this are why Botvinnik was one of the all-time great strategists:
1. Black’s queen evades the future pin on the c7-knight, after White plays Bf4.
2. Black threatens ... b5, followed by ... Rxd4.
21 Bg5
The routine 21 Bf4? is met with 21 ... b5! 22 Rc2 Rxd4, winning a pawn.
21 ... Re8?!
 The wrong square. Black should continue to apply pressure to d4, with 21 ... Rd7.
22 Qd2 f5!

Our move choices are merely reflections of our internal uniqueness. This is a typical
Botvinnik decision. He refuses to wait and tread lightly. When circumstances push and
knock him down, Botvinnik decides to get up and push back.

Question: Doesn’t Black just weaken with his last move?

Answer: Yes and no. A single action can be simultaneously wholesome and
unwholesome in its parts. The move certainly does weaken Black’s kingside pawns, while
on the plus side, Black seeks a resolution of White’s central dominance, opening d5 for
Black’s knight. In this case I believe Black’s plusses outweigh the minuses of the move.
23 Bh6?!
This move looks detrimental to White’s efforts, since it violates the principle: Don’t
swap one of your bishops away when you own the bishop-pair. White can continue more
aggressively with 23 Re1 Rf8 24 Bg2 fxe4 25 Rxe4 Rce8 26 h4 Nb5 (threat: ... Nd6) 27 h5!
(anyway; ambition is the great inducer, which persuades us into risky action which we
would normally avoid with a cooler head) 27 ... Nd6 28 hxg6 hxg6 29 Rg4 Nxc4 30 bxc4
Qc8 31 Be4 e5 32 Rh4 exd4 33 Bxg6 and despite Black’s extra exchange, White has a
powerful attack in the works.
23 ... Bxh6
Eliminating White’s bishop-pair.
24 Qxh6 Re7 25 Re1 Rf8!?
Possibly more accurate is 25 ... Rce8! so that Black can recapture on f5 with his e-
pawn.
26 Rc5
 Tal refuses to play 26 exf5 gxf5. Botvinnik may well have felt comfortable with the
Stonewall Dutch-like structure. I think Black may be okay here, despite the comp’s
preference for White.
26 ... Qd8
Attacking d4.
27 Re5!? Ref7
With his clock already low, Botvinnik would be better off accepting Tal’s challenge with
27 ... Qxd4! 28 exf5 gxf5 29 Bxf5 Ref7 30 R5e4 Qc5 31 Re5 Qd4 (note that Black can’t
play for a win with 31 ... Qb6?? 32 Bxh7+! Rxh7 33 Rg5+) 32 R5e4 with a draw.
28 Qd2

28 ... Qd6
28 ... f4! generates much needed counterplay: 29 Re2 (not 29 Bxe6?? Nxe6 30 Rxe6
fxg3 31 fxg3 Rf2 32 Qxf2 Rxf2 33 Kxf2 Qxd4+ 34 Ke2 Kf7 35 Rd1 Qxd1+ 36 Kxd1 Kxe6
37 Kd2 Ke5 38 Ke3 c5 and Black wins the king and pawn ending, due to king position and
the queenside pawn majority, which soon produces an outside passed pawn) 29 ... Rd7 30
d5 Rf6 31 gxf4 Rd6 32 Qb4 cxd5 33 Qxb7 d4 34 f5 Rb6 35 Qxa7 exf5 36 Re1 (36 exf5?? is
unplayable, due to 36 ... d3 37 Rd2 Rfc6 38 Rd1 d2 39 Bg2 Rc1 40 Bf3 Qd4 41 Re3 Qa1!
42 Qxc7 Rxd1+ 43 Bxd1 Qxd1+ 44 Kg2 Qg4+ 45 Rg3 Qe4+ 46 f3 Qe2+ 47 Kh3 d1Q and
Black wins) 36 ... Rfc6 37 exf5 Ra6 38 Qb7 Rab6 and the game is drawn by repetition of
moves.
29 Bf1
The bishop did nothing when aimed at f5, so Tal reconfigures it for c4, where it
pressures e6.
29 ... Rd7 30 exf5 Rxf5
After 30 ... exf5 31 Bc4+ Kh8 32 Qc3 Qf6 33 f4 White still boasts a strategic bind.
31 R5e4 Rf6?!
 There was no reason to go passive. Better was 31 ... Rd5! 32 R1e3 Rd8 (32 ... Rxd4??
loses to 33 Rxd4 Qxd4 34 Rd3) 33 a4 Rd7 34 h4 a6 and it isn’t easy for White to make
meaningful progress against Black’s fortress position.
32 h4!?

The lure of attack whispers a thousand endearments into Tal’s ear. This is a pawn offer,
after which Tal is ready to outlay his wealth in defence of his attacking convictions. 32 Qa5
provokes a weakening of the dark squares on Black’s queenside after 32 ... a6, but Tal is
more interested in Botvinnik’s king.
32 ... Kg7?!
Uncharacteristically passive. Botvinnik had to try 32 ... c5! (principle: meet the
opponent’s wing attack with a central counter) 33 h5 cxd4 34 hxg6 Rxg6 35 Bh3. White’s
super active pieces and attacking chances more than compensate for Black’s extra pawn, but
I think Black is better off here than in the game’s continuation, where he remained equal
materially, while keeping his bad position.
33 h5 gxh5!?
Black’s structure continues to get more and more pockmarked by Tal’s sniper fire.
However, it’s no better to allow White’s pawn forward with 33 ... Rdf7 34 h6+ Kg8, after
which Tal can play on Black’s weakened back rank.
34 Rh4 Kg8
Tal suggested 34 ... h6. White still retains excellent chances after 35 Ree4 Rdf7 36
Rxh5.
35 Bd3 Rg7 36 Re5!
Nullifying Black’s attempted sacrifices on g3, while preparing Rexh5, adding pressure
to h7.
36 ... Rff7 37 Qh6
Or 37 Rexh5 Nd5 38 Qe1 Re7 39 Rh6! when White exerts tremendous pressure.
 37 ... Qe7 38 Rexh5 Nd5
Preparing ... Nf6, if necessary.
39 Qd2
Backing off from your original intent is an exclam, if the original intent was misguided.
Tal said he looked at the line 39 Bxh7+?? Rxh7 40 Rg4+ Kh8 41 Qg6 and then saw 41 ...
Nf6! after which White’s attack failed miserably.
39 ... Nf6 40 Rh6 Qd6 41 Rf4

41 ... Qf8?!
The sealed move, played after 40 minutes of thought, which placed Botvinnik in time
pressure when the game resumed. Black’s margin for error is small – so small that the
slightest slip is of fatal proportions. This move strikes us as an aircraft carrier, attempting to
make a U-turn on a narrow part of a river. The ‘safety in numbers’ motto fails to apply in this
instance. Black should hand over a pawn to ease the pressure, with 41 ... Ng4! 42 Rxf7 Kxf7
43 Rxh7 Rxh7 44 Bxh7 Nf6. The position is lost for Black, although White still has a lot of
work to do to convert the technical ending.
42 Qe3! Nd5
42 ... Qe7 43 Bc4 wins, since 43 ... Nd5? is crushed by 44 Rxe6! Nxe3 45 Rxe7 Nxc4
46 Re8+ Rf8 47 Rfxf8 mate. This is the part of the Scooby Doo episode, where those
‘meddling’ kids unmask the villain.
43 Rxf7 Qxf7 44 Qe5!
The intrusive queen peers into Black’s position by pushing her glasses further up her
nose. Tal menaces infiltration to b8, which is stronger than 44 Qxe6 Qxe6 45 Rxe6 Nf4 46
Re3 Nxd3 47 Rxd3 Rd7 when Black may still harbour some hopes for drawing the rook and
pawn ending.
44 ... Nc7 45 Qc5!
There is no way to chase the rats out of Black’s home. Botvinnik’s position continues to
decline in a slow, steady drip. White’s position is a trifecta of ideal conditions for a win:
1. Tal works on Black’s weakened h- and e-pawns.
2. Tal makes stabbing threats to Black’s undefended queenside.
3. Tal goes after Botvinnik’s insecure king.
45 ... Qf3
45 ... a6? fails to 46 Qa7 Nd5 47 Qb8+ Qf8 48 Bxh7+! Kf7 49 Qxb7+ and Black can
resign.

Exercise (combination alert): Sometimes a combination just deserves to come to

life, and we are merely the happy conduit. How did Tal force the win of material?

Answer: Double attack.

46 Bxh7+!
An opponent’s unexpected shot tends to rob us of our innocence. If this isn’t heaven for
the bishop, then it’s a heaven-analogue, where he is granted entry to paradise, while still
alive.
46 ... Rxh7 47 Qg5+ Kh8
Even worse for Black is 47 ... Rg7? 48 Qd8+ Kf7 49 Qxc7+ Kg8 50 Qd8+ Kf7 51 Qd7+
Kg8 52 Qe8+ Qf8 53 Rh8+ which wins Black’s queen.
48 Qd8+ Kg7 49 Rxh7+ Kxh7 50 Qxc7+ Kg6 51 Qxb7
Botvinnik is down two pawns, with no perpetual check.
51 ... Qe4
51 ... Qd1+ 52 Kg2 Qxd4 53 Qxc6 is equally hopeless for Black.
52 Qa6
Covering his a2-pawn, while forcing Black to defend both a7- and c6-pawns.
52 ... Qe1+ 53 Kg2 Qe4+ 54 Kf1 Qb1+ 55 Ke2 Qc2+ 56 Kf3 Qf5+ 57 Ke3 Qg5+ 58
Ke2 Qh5+ 59 Kd2 Kf6 60 Qxc6 Qa5+ 61 Qc3 Qxa2+
The pawn count remains the same: minus two for Black.
62 Ke3 Kf7

Exercise (planning): Tal found a clear winning plan for White. What is it?

Answer: Exchange Black’s e6-pawn, after which White creates two connected passed
pawns (Yes, I alertly remembered IM Tony Saidy’s admonition to avoid the hated term
‘passer’!).
63 d5! exd5 64 Qc7+ Kf6 65 Qc6+ Ke7
Or 65 ... Ke5 66 f4+ Kf5 67 Qxd5+.
66 Qxd5 Qa1 67 Qe4+ Kf7 68 Kf4 Qc1+ 69 Kg4 Qa1 70 Qd5+ Kf8 71 Kf5 Qb1+ 72
Kf6 1-0
The only thing missing is Benny Hill banjo chase music. 72 ... Qb2+ 73 Ke6 and now 73
... Qxf2 is met with 74 Qf5+. The desolation seems to be desolating harder, and with greater
conviction. With the current funding squeeze, Black, who is left with a hopelessly lost king
and pawn ending, lacks resources to continue.

Game 24
M.Tal-M.Botvinnik
World Championship (Game 19), Moscow 1960
Dutch Defence

1 c4 f5
If we itch for a fight, we use any pretext to declare war. A desperate Botvinnik, at this
stage in the match needing to win even with the black pieces, falls back on his old favourite,
the Dutch Defence.
2 Nf3!

Question: Why an exclamation mark for this obvious move?

Answer: Tal realized that Botvinnik planned to use his long time risky favourite, the
Stonewall Dutch. The trouble is the Stonewall doesn’t work well for Black if White holds
back on d4, since then White can play for a d3 and e4 pawn break. So basically Tal forced
Botvinnik out of his favourite line – a clear psychological blow.
2 ... Nf6 3 g3 g6 4 Bg2 Bg7 5 d4
Now d4 is fine, since Botvinnik committed to the Leningrad Dutch.
5 ... d6 6 Nc3 e6!?
Botvinnik opens with a rarely played continuation.

Question: What is Black’s idea?

Answer: The idea may be:

1. To keep White wondering if Black will play for ... e5 or ... d5.
2. The move clears e7 for Black’s queen, which usually plays to the e8-square.
3. Now if White plays for d5, then Black can play ... Ne5 more easily than in the normal
lines.
More often seen is 6 ... 0-0 7 0-0 and now 7 ... Qe8, 7 ... c6 and 7 ... Nc6 are most
commonly played.
7 0-0 0-0 8 Qc2
White’s normal plan in the position is to enforce an e4 break. In this version White
doesn’t worry about future ... Bf5 tempo gains as much, since Black’s ... e6 hinders this
possibility. White’s move also prevents Black from a relieving exchange with ... Ne4. 8 b3
and 8 Re1 are also played here.
8 ... Nc6
Botvinnik plays for an ... e5 break.
9 Rd1
9 d5 Nb4 10 Qb3 Na6 looks fine for Black.
9 ... Qe7 10 Rb1
The idea is to blanket the queenside with white pawns with the plan b4, b5, a4 and Ba3.
10 e4 is also possible: 10 ... fxe4 11 Nxe4 e5 12 dxe5 Nxe4 (the comp prefers 12 ... dxe5
13 Bg5 Bf5) 13 Qxe4 dxe5 14 g4! Kh8 15 h3 Be6 16 Be3 and White secured a strategic
advantage from the opening, M.Podgaets-M.Tal, Alma-Ata 1968.
10 ... a5
Suppressing White’s intended b4. 10 ... e5 11 dxe5 Nxe5 12 Bf4 looks slightly better for
White.
11 a3
Renewing the b4 idea.
11 ... Nd8
The knight removes itself from the coming d5, or b4 and b5, transferring to the kingside.
Botvinnik’s idea is ... Nf7 and eventually ... e5.
12 e4
Since Black retreated the c6-knight, Tal feels the time has come for his e4 break.

12 ... fxe4 13 Nxe4 Nxe4 14 Qxe4 Nf7

After 14 ... e5 15 dxe5 Nf7 (now Black threatens ... Bf5; 15 ... Bf5 isn’t all that
effective, due to 16 Qd5+ Nf7 17 exd6 Qxd6 18 Ra1) 16 Qd5 dxe5 17 Be3 c6 18 Qc5 Bf6
19 Qxe7 Bxe7 White looks a shade better in the ending, since Black must nurse the e5
isolani, watch over the e4-square, and his queenside is also slightly weakened.
 15 Bh3!
Botvinnik finds a way to suppress Black’s ... e5 intent.
15 ... Qf6
Botvinnik sets up the trap ... d5, which wins a pawn, and which Tal deliberately allows
since the resulting position increases complications.
16 Bd2!
Sacrifice is an act which the faithful use to propagate the chess goddess. Tal willingly
allows Botvinnik his idea, handing over a pawn to seize the initiative.
16 ... d5!?

We may see a tempting idea, yet if we lack the necessary resources, our chances of
successful implementation may as well be on the dark side of the moon. Botvinnik attempts
to build an improvised explosive device, made from fertilizer, gasoline, ball bearings and
prayer. Down in the match, he is unable to resist the lure of the free pawn – at the terrible
cost of handing Tal a sustaining initiative. I suppose that when our opponent shames us
through strategic pressure, we get the urge to lash out to save face. Botvinnik said he also
considered abstaining with 16 ... e5! which I believe is Black’s best move. After 17 Bxc8
Raxc8!, a5 is hanging, as is b7, but weirdly enough, White has difficulty proving an edge
here. For example:
a) 18 Qxb7? c6 19 Nh4 g5 20 Ng2 exd4 and Black stands clearly better.
b) 18 Bxa5?! Nd8! 19 Kg2 exd4 and once again, Black’s d4-pawn and pressure down
the f-file give him the edge.
c) 18 dxe5 Nxe5 19 Bf4 Nxf3+ 20 Qxf3 b6 looks fine for Black.
d) 18 d5 Bh6! and White is unable to play 19 Bxa5? due to 19 ... Ng5 20 Nxg5 Qxf2+ 21
Kh1 Bxg5 when Black stands better.
17 Qe2 dxc4!?
Botvinnik reasons that he may as well take the pawn, since Black may be under pressure
without his extra pawn after 17 ... b6 18 Rbc1.
18 Bf4
The comp likes 18 Bg2, and if 18 ... b5 19 Ne5 (19 b3 cxb3 20 Rxb3 Bb7 looks okay for
Black) 19 ... Ra6 20 f4 a4 21 Bb4 Nd6 22 Bc5 with a massive dark-square bind for the
pawn.
18 ... Nd6?!
Clearly, Black’s extra c4-pawn isn’t looking like much of a bargaining chip at this point.
I think Black had to try 18 ... b5 19 a4 Bb7 (threat: ... Bxf3, followed by ... g5, winning a
piece) 20 Bg2 bxa4 21 Bxc7 Bd5 22 Be5 Qd8 and Black looks no worse to me.
19 Ng5!?
A surprise, intending Ne4, challenging Black’s defender of c4, his knight. Botvinnik had
only counted on 19 Ne5 which also looks favourable for White.
19 ... Re8
19 ... Bh6 20 Ne4 Nxe4 21 Bxh6 Qxf2+ 22 Qxf2 Nxf2 23 Bxf8 Nxh3+ 24 Kg2 Kxf8 25
Kxh3 would be a tough ending for Black to hold.
20 Bg2
Preventing ... b5.
20 ... Ra6 21 Ne4!

Tal completes his plan of eliminating the sole defender of c4.

21 ... Nxe4
21 ... Qe7 22 Nxd6 cxd6 23 Qxc4 leaves Black under heavy strategic pressure.
22 Bxe4 b5
Otherwise White plays Qxc4, equalizing the material count, while pressing a clear
strategic advantage.
23 b3!
Tal finally lances the warty c4 protrusion.
 23 ... cxb3
23 ... Bf8 24 bxc4 bxc4 25 Bxc7 and now 25 ... Bxa3 is met with 26 Rb8 Ra7 27 Qxc4
Rf8 28 Bf4 Bd7 29 Rxf8+ Qxf8 30 d5!. Opening the centre seriously exposes Black’s king:
30 ... Qb4 31 Qe2 and Black’s king is in dire trouble.
24 Qxb5
As in all wars, innocent civilians become victims of collateral damage. The double
attack on e8 and b3 regains the pawn, while retaining all his strategic pressure.
24 ... Rf8 25 Qxb3 Rb6 26 Qe3 Rxb1 27 Bxb1 Bb7
27 ... Qf7 28 Rc1 wins Black’s c-pawn.
28 Ba2!

Now e6 and c7 are under simultaneous attack, and one of them falls. Less accurate
would be 28 Bxc7 Qf3 29 Qxf3 Bxf3 30 Rd2 Rc8 31 Rc2 Bxd4 32 Bxa5 Rb8 33 Ba2 Kf7
when Black gets better drawing chances than in the game continuation.
28 ... Bd5
Botvinnik swaps away his last real chance of a successful attack. He probably looked at
28 ... Re8 29 Bxc7 Qf5 30 Re1 and realized he had to swap anyway, with 30 ... Bd5 31
Bxd5 Qxd5 32 Be5 when White had good chances to convert his extra pawn.
29 Bxd5 exd5 30 Bxc7 a4
Covering his a-pawn.
31 Rd3
So Tal is up a pawn and now comes the technical phase of the game. Tal played this
move, rather than the more natural 31 Kg2. Maybe because he wanted to achieve future
swaps via f3, later on.
31 ... Qf5 32 Be5 Bh6! 33 Qe2 Rc8 34 Rf3 Qh3?!
The proximity alarm goes off in White’s camp, sounding the approach of an intruder.
Black threatens a back rank mate. However, Botvinnik may have generated superior chances
to hold the game with 34 ... Rc2! 35 Rxf5 Rxe2 36 Rf6 Ra2 37 Rd6 Rxa3 38 Rxd5 Rb3 39
Ra5 a3 40 d5 Bf8. If Black can somehow swap away his a-pawn for White’s d-pawn, then
the game is a likely draw.

Exercise (combination alert): It looks like Black is the one with the initiative,
right? However, absence of evidence never seemed to stop Tal from finding
hidden combinational ideas. How did Tal seize the initiative?

Answer: Interference/weak back rank.

35 Bc7!
“God looks the other way, when a tiny sin serves a higher truth,” declares the bishop. A
little sand is thrown into Black’s defensive gears, as Tal finds a beautiful
defensive/offensive shot. Chess is an enclosed system, with immutable physical laws. Tal,
like no other player before him, routinely found geometric absurdities which defied such
laws and profited from such violations.
When I was a kid, my younger brother and sister watched this show for pre-schoolers,
called Romper Room. The show’s prime directive was to weed out the rebels and defiers of
authority among us, and turn us all into obedient citizens. Now in the show there was a role
model cartoon kid named Mr. Do-Bee, and also his arch nemesis and anti-social rule-
breaker Mr. Don’t Be. I’m guessing Tal, if he ever watched the show, exclusively rooted for
Mr. Don’t Be! Botvinnik undoubtedly expected 35 Rd3?! Qf5 when Black has some
initiative for the pawn.
35 ... Bf8?!
When we are surprised by an opponent’s shot, we tend to blurt out an instinctual
response, rather than pause to regain composure. This mixed with the confusion of time
pressure – where realizations come slowly, if at all – induces Botvinnik into a slip:
 a) 35 ... Rxc7?? walks into 36 Qe8+ Kg7 37 Qf8 mate. White queen and rook clink
champagne flutes, on hearing the good news about the black king’s sudden death.
b) 35 ... Qd7! is Black’s best defence. After 36 Bf4 White gained a tempo, yet Black can
continue to resist.
36 Qb5!
Another dirty trick. Tal leaves his bishop hanging, attacking the d5- and a4-pawns.
36 ... Qe6
36 ... Rxc7?? 37 Qxd5+ Kg7 38 Qe5+ picks off the stray black rook.
37 Be5
Threat: Rf6, chasing away the only defender of d5. Not only is White up a pawn, but he
menaces Black’s pawns and also Black’s insecure king.
37 ... Qc6 38 Qa5!
Nyet! A minimalist/reductionist, Capa-style player might sigh in dismay over Tal’s
decision, yet it is the correct one. Tal is not about to swap queens – even a pawn up – when
he is the one with attacking chances. After 38 Qxc6? Rxc6 the trouble is White has no way
to defend against the coming ... Rc1+, followed by ... Ra1.
38 ... Ra8 39 Qd2 Rc8 40 Kg2 Qd7 41 h4!

Just as in the previous game in the book, Tal recognizes that the push of his h-pawn
guarantees the further weakening of Black’s king.
41 ... Qg4 1-0
Nothing to see here, folks. Just move along to the next game and don’t obstruct traffic.
Botvinnik, thinking there is no sense in moping for the unattainable, resigned here. “Until
this time, I had very rarely played this type of game. This is why I consider the nineteenth
game my best creative achievement in the match,” wrote Tal.

Question: I understand that Black is down a pawn and also stands

 worse. But isn’t this a premature resignation from Botvinnik?

Answer: I don’t think so. We can only patch and repair a home built on an unstable
foundation so many times, and then it all comes apart, no matter what we do. Also,
Botvinnik, age 48 during this match, was more than double Tal’s age. So by resigning he
essentially gave himself a rest day. We can’t blame him. When we are mentally exhausted
from calculating endless lines, it becomes tiring to follow a hopeless position to its natural
conclusion. White’s strategic plusses ensure further gains. For example, 41 ... Qg4 42 Qd3
Qe6 43 Rf6 Qg4 44 Qb5 Rd8 45 Qxa4. If you are a vampire, you don’t pause to ask your
victims about their blood type. White picks up a second pawn.

Game 25
M.Tal-Y.Averbakh
USSR Team Championship, Riga 1961
Ruy Lopez

1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7 Bb3 d6 8 c3 0-0 9 h3 Na5
10 Bc2 c5 11 d4 Qc7 12 Nbd2 Nc6 13 dxc5
This move was popular at the time. But like all fashions, arose and then faded away into
obscurity. Today, 13 d5 is thought to be White’s best chance for an advantage.
13 ... dxc5 14 Nf1 Rd8 15 Qe2 g6!?
This move is playable, but inadvisable. Averbakh soon comes to regret weakening the
dark squares around his king. Today, 15 ... Be6 is considered an equalizer.
16 Ne3
White extracts an edge after 16 Bh6 Be6 (or 16 ... Bf8 17 Bxf8 Kxf8 18 a4 Rb8 19 axb5
axb5 20 Qe3 Qe7 21 Qh6+ Kg8 22 Ne3 Qf8 23 Qh4 with pressure on the dark squares
around Black’s king) 17 Ng5 Bc4 18 Qf3 and now Ne3 is coming, inducing Black into 18 ...
Bxf1 which hands White the bishop-pair in a semi-open position.
16 ... Rb8 17 Ng5!?
 I wouldn’t describe this move as Zen-like calm. Tal decides not to play a waiting game
and gets straight down to business, hoping to provoke ... h6.
17 ... Bf8
Averbakh decides to lay low and avoid eye contact.

Question: Why can’t Black simply gain a tempo with 17 ... h6?

Answer: The move is very playable, yet psychologically difficult, since ... h6 weakens
the pawn front around Black’s king. White can play 18 Nd5 Nxd5 19 exd5 Bxg5 20 Bxg5
hxg5 21 dxc6 Rb6! 22 Qxe5 Qxe5 23 Rxe5 Rxc6 24 Rxg5 Rd2 25 Be4 Rc7. Black’s
foundation shakes and shivers, yet remains intact. After 26 b3 White won a pawn, while
Black’s d2-rook is annoyingly active, perhaps providing full compensation for the pawn.

Question: This looks perfectly playable for Black.

Why would Averbakh avoid such a line?

Answer: At the board we can’t be 100% certain of anything, since we all hallucinate
and blunder. I can easily see an older GM avoiding such a long calculation against the young
Tal, simply out of paranoia.
18 Qf3!?
Risky, since now Tal’s g5-knight is short of safe squares, should Black play a future ...
h6.
18 ... Be7?
This bishop continues to drift around, working odd jobs. We humans forget that our
cognitive interpretation of data doesn’t necessarily equal external reality. Averbakh, perhaps
seeing a ghost (which was very easy to do when facing Tal!), inexplicably agrees to the loss
of two tempi. I’m not sure why he rejected the thematic move 18 ... Bg7. Black looks fine
after 19 Nd5 Nxd5 20 exd5 Na5 when ... Bb7 is coming and the artificially isolated d-pawn
may become a source of concern for White.
19 Nd5!

White’s feral knights are not the benign farm animals we find in the children’s exhibition
in the zoo.
19 ... Qd6
Of course, Black is unable to play 19 ... Nxd5?? 20 Qxf7+ Kh8 21 Qxh7 mate. Everyone
realizes that the queen’s love can so easily turn to hate.
20 Be3!
A military commander’s most painful choice is to judge which of his or her soldiers is
expendable in battle. In this case Tal self-traps his g5-knight, but keep in mind when Tal
sacrificed material, his motivation was never altruistic.
20 ... h6 21 Nxf6+ Bxf6?
Black had to hand over a pawn to remove queens from the board with 21 ... Qxf6 22
Qxf6 Bxf6 23 Nf3 (the c5- and h6-pawns are under simultaneous attack) 23 ... Be7 24 Bxh6
and White should be winning, with a clean extra pawn.
22 Rad1 Qe7
After 22 ... Nd4 23 cxd4 cxd4 24 Bxd4! Bxg5 (24 ... exd4 25 Nxf7 Kxf7 26 e5 is also
awful for Black) 25 Ba7 White wins material.
 Question: Isn’t White just losing? After all, White’s knight lacks a retreat square.

Answer: Tal foresaw this position and realized that he had a winning combination. Let’s
turn it into an exercise:

Exercise (combination alert): Geometric anomalies connive

against Black’s hopes. White to play and force the win:

Answer: Deflection/annihilation of defensive barrier. This piece sacrifice yields huge

returns.
23 Bxc5!
“If the soul is eternal and indestructible, then where is the sin in killing?” ponders the
bishop. This shot feels like that dream we all experience, where we awaken in shock, a split
second before our falling dream body hits the dream ground.
23 ... Rxd1?!
Life tends to get complicated when we forget the details. This is another mistake in a
lost position. Black puts up greater resistance with 23 ... Qxc5 24 Qxf6 hxg5 25 Rxd8+
Nxd8 26 Qxd8+ Kg7 27 Rd1 Rb6 (27 ... f6?? 28 Rd6 is crushing) 28 Rd5 Qc6 29 Qxg5 f6
30 Qg3 Be6 31 Rd8!? when Black is unable to regain his lost pawn with 31 ... Bxa2? due to
32 Qh4 Bg8 33 Bb3! (attraction/clearance) 33 ... Bxb3 34 Qh8+ Kf7 35 Qg8+ Ke7 36 Qf8+
Ke6 37 Re8+, winning Black’s queen.
24 Rxd1 Qxc5 25 Qxf6 hxg5
25 ... Qf8 is crushed by 26 Nxf7! Qxf7 27 Bb3.
26 Bb3!
Every obstacle has been accounted for, by Black – except one! A neurotoxin has been
introduced into Black’s system, and all Tal needs to do is sit back and wait, while his
opponent collapses on f7 and g6.
26 ... Rb7
26 ... Qf8 27 Qxg6+ Kh8 28 Bxf7 Rb7 29 Qf6+ Kh7 (29 ... Qg7 30 Qxc6 Rxf7 31 Qxc8+
is totally lost for Black) 30 Bg6+ Kg8 31 Qxc6 is hopeless for Black.
27 Qxg6+
Dear Leader instructs the building of yet another grandiose monument in her own image.
27 ... Kf8 28 Qh6+ 1-0

The terms of the agreement are not so generous for Black. The current of variations sway
sinuously, yet always ending up heavily in White’s favour. 28 ... Ke7 29 Rd5! Qb6 30
Qxg5+ Ke8 (try as he may to wriggle free, Black’s king sways back and forth like a tethered
balloon) 31 Qg8+ Ke7 32 Qxc8 leaves Black down three pawns.

Game 26
M.Tal-M.Botvinnik
World Championship (Game 8), Moscow 1961
Caro-Kann Defence

The 1961 world championship rematch, as we all know, was nothing short of a disaster for
Tal, who struggled with health issues and just couldn’t reach his kind of positions. This
game and the following were two where Tal managed to achieve the complication level
necessary to defeat Botvinnik.

1 e4 c6
We shape our opening repertoires in our own stylistic images. Botvinnik realized that the
Caro-Kann’s ultra-solidity was his best bet against Tal’s 1 e4. Compare Tal’s game with this
Nimzowitsch classic, which arose from a similar structure from the French Defence: 1 ... e6
2 d4 d5 3 e5 c5 4 Qg4 cxd4 5 Nf3 Nc6 6 Bd3 f5 (today, 6 ... Qc7 and 6 ... Nge7 are
generally played) 7 Qg3 Nge7 8 0-0 Ng6?! 9 h4 Qc7 10 Re1 Bd7?! 11 a3 0-0-0 12 b4! a6?!
13 h5 Nge7 14 Bd2 h6 15 a4 g5 16 b5 f4 17 Qg4 Nb8 18 c3 Re8 19 cxd4 Kd8 20 Rc1 Qb6
21 a5 Qa7 22 b6 Qa8 (it’s generally a bad sign when your opponent chases your queen into
a corner, without access to a single square) 23 Rc7 Nf5 24 Nc3 Be7 25 Nxd5! Nxd4 26
Nxd4 exd5.

Exercise (combination alert): Herr. Hakansson played the French

with less-than-stellar accuracy. How did Nimzo force mate?

Answer: Queen sacrifice/smothered mate: 27 Qxd7+! 1-0, A.Nimzowitsch-

A.Hakansson, Kristianstad 1922 (27 ... Nxd7 28 Ne6 is mate).
2 d4 d5 3 e5!?
A sign of Tal’s desperation against Botvinnik’s dreaded Caro-Kann. Of course,
attempting to defeat Botvinnik in a closed, strategic position was akin to the attempt to
drown a fish. Against Botvinnik, Tal generally got nothing from the opening when he played
his normal 3 Nc3 dxe4 4 Nxe4.
3 ... c5!?
 At the time, this rare move was considered an odd inversion of what was considered
normal. Today, it’s the height of fashion. I think Botvinnik should have stuck with the 3 ...
Bf5 line.

Question: Why?

Answer: Because in that version, White has a much harder time opening the position.
And as we all know, Tal’s forte was open – not closed! – positions. Later in the match,
Botvinnik returned to the normal 3 ... Bf5 4 h4 (today, this is considered one of White’s most
dangerous lines against the Advanced Caro) 4 ... h6 (Botvinnik isn’t a likely candidate for
the beginner’s trap 4 ... e6?? 5 g4 Bg6 6 h5 Be4 7 f3 and White wins a piece) 5 g4 Bd7 6 c3
c5 7 Bg2 (the bishop ends up hitting a wall on d5) 7 ... e6 8 Ne2 Bb5 9 Na3? (Botvinnik, in
the 8th game of the match, committed the same strategic error in reverse with his 9 ... Nh6?)
9 ... Bxe2 (Botvinnik rids himself of his bad bishop) 10 Qxe2 cxd4 (principle: knights
operate better in locked pawn structures) 11 cxd4 Bxa3! (good strategic judgement;
Botvinnik understands that he can tame Tal’s bishops in this rigid pawn structure) 12 bxa3
Nc6 13 Be3 Qa5+ 14 Kf1 Nge7 15 Rb1 Rb8 16 Bh3 Qa4 17 Rd1 Qxa3 18 Kg2 Qa6 19
Qxa6 bxa6 20 h5 Kd7 (principle: centralize your king in an ending) 21 Rb1 Rb6 22 Kg3
Na5 23 Rxb6 axb6 when White had no visible compensation for his missing pawn, and
Botvinnik efficiently converted to earn the full point, M.Tal-M.Botvinnik, World
Championship (Game 18) Moscow 1961. This game is annotated in Botvinnik: Move by
Move.
4 dxc5
I think this is one of White’s best lines, to enter a line of the Advanced French, but a
move up, since Black took two moves to play ... c5. Instead, 4 c3 Nc6 5 Nf3 cxd4 6 cxd4
Bg4 gives Black exactly what he wants, an Advanced French position, with his traditionally
bad bishop on the outside of his pawn chain.
4 ... e6 5 Qg4!?

Question: Isn’t this a violation of principle?

Answer: A deliberate violation by Tal, who wants to tie Botvinnik’s bishop down to
defence of g7. This is a variant on an old Nimzowitsch theme, which normally ran 1 e4 e6 2
d4 d5 3 e5 c5 4 Qg4. Tal gets a superior version in this one, since he is up a tempo, because
Black took two moves to achieve ... c5.
White can also play to hang on to c5, with 5 Be3 Nh6 6 c3 Nf5 7 Bd4 Bd7 8 Nf3 Nc6 9
a3 a5 10 Be2 a4. Black achieved light-squared compensation for his pawn, J.Nunn-
J.Speelman, British League 2001.
5 ... Nc6 6 Nf3 Qc7 7 Bb5
White can also try 7 Nc3, intending Nb5, and then:
a) 7 ... a6 8 Na4 Nxe5 9 Nxe5 Qxe5+ 10 Qe2 Qxe2+ 11 Bxe2 Bd7 12 Nb6 Rd8 13 b4. I
like the queenside majority in the ending, and also White’s entrenched knight is a pain for
Black.
b) 7 ... Nxe5? 8 Nb5 Qa5+ 9 b4 Nxg4 10 bxa5 Bxc5! 11 Nc7+ Kd8 12 Nxa8 Nxf2 13
Rg1 f6 14 Rb1 Ne4 15 Rh1 and I don’t think that Black can so easily pick off the a8
straggler.
 Question: Can’t Black force a draw with 15 ... Nf2?

Answer: The move is a blunder. White wins with 16 Rb5! Ne4 17 Bd3 Bd6 18 Be3
when his knight will escape its imprisonment from a8.
7 ... Bd7
This move is considered inaccurate by today’s standards. After 7 ... Nge7 8 0-0 Bd7 9
Re1 Ng6 10 Bxc6 Qxc6 11 b4 b6 12 cxb6 axb6 13 c3 Be7 14 Qg3 0-0 Black’s queenside
pressure compensates for his missing pawn, E.Schreiber-J.Barrios Troncoso,
correspondence 2001.
8 Bxc6
Tal hands over the bishop-pair to reduce the pressure on e5.
8 ... Qxc6 9 Be3 Nh6?
An attacker utilizes violence as a method of imprinting a position with his or her
stylistic stamp, while the strategist prefers gentle coercion. Botvinnik sinks all his karma –
both good and bad – into this over-designed contrivance, which lands Black in a
strategically inferior position. He would have been better off playing the more patient 9 ...
h5 10 Qg5 Ne7, intending ... Nf5.
10 Bxh6!
Each asymmetrical swap on the chess board, is in a way, an exchange of hostages.
 10 ... gxh6
Good judgement on Tal’s part. Black’s structure, once without blemish, now sees
impurities creep within. White’s knights adequately deal with Black’s bishops, mainly since
White dominates the d4 – the ideal blockading square.

Question: Is this a correct assessment? Won’t the

game later open up in favour of Black’s bishops?

Answer: I think White’s development lead and control over d4, plus Black’s weakened
h-pawns assure Tal of the advantage. The odd thing is in this, the eighth game, Tal played
with two knights versus two bishops and won; in the 18th game, Botvinnik ended up with
two knights versus Tal’s two bishops, and won. Moral: knights are better than bishops! My
San Diego IM buddies John Watson and Keaton Kiewra tease me all the time about my
knight-over-bishop bias. Well, these two games are clear proof that I was right all along!
11 Nbd2 Qxc5 12 c4!
This move is strong for two reasons:
1. The black king is sure to castle long, since everywhere else on the board is unsafe. So
White threatens to open that sector of the board.
2. Principle: When it’s a battle of development lead versus the opponent’s bishop-pair,
development lead takes precedence, and you should open the game.
12 ... 0-0-0 13 0-0
Tal’s dream has been achieved: opposite wings castling.
13 ... Kb8 14 Rfd1 Qb6
I would think about handing White control over d6 to open the game with 14 ... dxc4!?
15 Nxc4 Be7 16 Qf4 Bc6 17 b3 Bd5 18 Rac1 Qb4. I still like White’s game, but at least
here Black’s bishops achieve some degree of activation.
 15 Qh4
Multipurpose: covering f2 against a future ... Bc5, while keeping Black’s queen tied to
defence of her d8-rook.
15 ... a5?!
Botvinnik suppresses b4 at a high cost of allowing White a future b4 prying-mechanism.
He may have been better off playing 15 ... Rg8.
16 Rac1 Rg8 17 Nb3! a4?!
I think Botvinnik should open the game (not an easy decision against Tal!) with 17 ...
dxc4 18 Rxc4 a4 19 Nbd4 Bc5 20 Rc2 Bb5 21 Rcd2 (threat: Nc6+) 21 ... Rd5 22 Nxb5
Qxb5 23 Rxd5 exd5 24 Rd2 Bb4 25 Nd4 Rc8 26 Rd1 Qd3 27 Qg4 Qe4 28 Qxe4 dxe4, with
an inferior but playable ending for Black.
18 c5!

Terrific strategic judgement from Tal (yes, I realize this is a strange statement to make).
The move acts as a binding agent which solidifies White’s strategic advantage:
1. White dominates d4.
2. White’s c5-pawn cramps Black’s dark-squared bishop.
3. Now Black must be on constant alert for c6 line-opening ideas.
18 ... Qc7
The b6-square was only a temporary way station for Black’s queen. This is the only
move, since Black’s queen must remain on the d8-rook.
19 Nbd4 Rc8
Contemplating ... Bxc5, rather than 19 ... Bxc5!? 20 b4 axb3 21 Nxb3 b6 22 Qf4 Qa7 23
Qxf7 Rdf8 24 Qxh7 Ba4 25 Qd3 Bxb3 26 axb3. Although I don’t see full compensation for
Black’s missing pawn, I would still take this path over the game continuation.
20 b4! axb3 21 axb3
Now the a-file opens, an ominous sign for Black’s king. A key goal of any war effort is
to manufacture new copies to replace the old war machines which are destroyed in battle.
White now follows with b4, which not only covers c5, but also threatens to press forward
later on, with b5 and either c6 or b6 pushes. This position is an example of two knights
choking their counterpart bishops, who have no scope.
21 ... Qd8
Botvinnik, rightfully fearing Tal’s coming attack, forces queens off the board.
Unfortunately for him, White’s strategic domination remains in the ending.
22 Qxd8 Rxd8 23 b4 Rg4?!
This move, attempts to colonize barren territory. Botvinnik plans ... Bg7, ... Re4 and ...
Bxe5, which is too slow. He should play 23 ... Rc8 to suppress b5.
24 b5!

When we commit to such a move, there is no override command which halts the process,
once it begins.
24 ... Rc8
Pursuit of c5 turns out to be a fleeting enthusiasm for Black, since provoking c6 only
benefits White.
25 c6!
Forced, but strong. The game opens, yet disorientingly, Black’s bishops seem to grow
more and more pathetic.
25 ... Be8
Black can’t open lines with 25 ... bxc6? 26 bxc6 Be8 27 Rc2 h5 28 Rb1+ Kc7 29 h3
Re4 30 Nb5+ Kd8 31 Nfd4. Black is crushed, since 31 ... Rxe5 32 Na7 Rc7 33 Ndb5 is
decisive.
26 Rc2 Bg7
If Black attempts to set up a dark-square blockade with his king on b6, with 26 ...
Ka7??, White meets it harshly with the line opening 27 b6+! Kxb6 28 Rb1+ Kc7 (or 28 ...
Ka7 29 Rxb7+ Ka6 30 Ra2+ Ba3 31 Rxa3 mate) 29 cxb7+ forcing mate.
27 Ra1!
Tal gathers energy for his coming attack, ignoring Black’s attack on his e5-pawn.
27 ... Bxe5?
Black reaches for the honey, deep inside the hive. Botvinnik continues to fixate on the
fiction that he will pick off e5 and get away with the crime. Such surges of adrenaline tend
to wipe away subtle distinctions. Thoughts are the fastest thing in the universe, except that
no one knows where they come from, or where they go. And with the help of such irrational
thoughts, we humans are masters of rationalizing – and therefore justifying – our own self-
destructive actions. Botvinnik inexplicably decides to call Tal’s ‘bluff’, forgetting that such
last minute rescues only happen in movies and novels.
Black offers greater resistance with the bleak yet superior options:
a) 27 ... Rxd4 28 Nxd4 Bxe5 29 Rca2! Bxd4 30 Ra8+ Kc7 31 Rxc8+ Kxc8 32 Ra8+
Kc7 33 cxb7 Kxb7 34 Rxe8 Kb6 35 Re7 Kxb5 36 Rxf7 and White converts the ending.
b) 27 ... Re4 28 Kf1 Bxe5 29 Nxe5 Rxe5 30 Rca2 Kc7 31 b6+! Kd6 (31 ... Kxb6 32
Rb1+ Kc7 33 cxb7 Rb8 34 Rc1+! Kd6 35 Ra6+ Kd7 36 Ra7 Re4 37 Nc6 wins) 32 cxb7
Rb8 33 Ra8 Rxb7 34 Rxe8 Rxb6 35 Nf3 when White should convert.
28 Nxe5 Rxd4

Exercise (combination alert): Both sides had foreseen this position,

but only Tal calculated it with accuracy. White wins by force. How?

Answer: Attraction/deflection.
29 Nd7+! 1-0
As it turns out, the relationship of Black’s king behind his defensive wall, is that of
content within a sealed container. 29 ... Bxd7 (29 ... Kc7 30 b6+ Kd8 31 cxb7 is
catastrophic for Black) 30 cxd7 Rd8 31 Rc8+! (attraction) 31 ... Rxc8 32 Ra8+! (deflection;
as you may have guessed, the c- and a-files are popular hiking trails) 32 ... Kxa8 33 dxc8Q+
and White promotes with check, cruelly denying Black a back-rank mate.

Game 27
M.Botvinnik-M.Tal
World Championship (Game 17), Moscow 1961
King’s Indian Defence

1 d4 g6!?
Tal issues a challenge with a Modern Defence move order.
2 e4
Wow. Is Botvinnik really going to play a king’s pawn opening?
2 ... Bg7 3 c4
No. Just kidding. Botvinnik refuses to be lured into the 3 Nc3 d6 main lines of the
Modern Defence.
3 ... d6 4 Nc3 Nf6 5 f3 Nbd7 6 Be3 0-0 7 Bd3 e5 8 Nge2
Botvinnik allows the central tension to remain. Normal is 8 d5 c6 9 Nge2 cxd5 10 cxd5
Nh5 11 0-0 f5 12 exf5 gxf5. Theory considers this position in White’s favour, L.Portisch-
S.Gligoric, Milan 1975.
8 ... Nh5 9 dxe5!?

Question: Didn’t White just throw away his first-move advantage with his last move?

Answer: I think he did just that. I’m not even sure what motivated Botvinnik, other than
to lure Tal out of his pre-match opening preparation. Another reason may be that the
exchange on e5 tends to drain the position of dynamism, which forces Tal into a slow
manoeuvring game, not to his liking. Objectively, White’s only path to an edge lies in 9 d5 f5
10 exf5 gxf5 11 Qd2 Nc5 12 Bc2 a5 13 0-0-0, A.Miles-M.Hebden, Biel 1983.
9 ... dxe5
Black equalized, since he controls d4, while White’s control over d5 can be challenged
with ... c6. 9 ... Nxe5 is also probably okay for Black, since he picks up the bishop-pair.
10 0-0
Denying Tal opposite wings castling.
10 ... c6 11 Qd2 Qe7 12 Rad1 Nc5 13 Bb1 Ne6
The knight is optimally placed on e6, where it eyes both d4- and f4-squares.
14 Qe1
Botvinnik removes his queen from the d-file and prepares Qf2, taking aim at Black’s
queenside.
14 ... Bf6!?
An odd move, perhaps contemplating ... Bg5. Black looks slightly better after the more
natural 14 ... Nhf4 15 g3 Nxe2+ 16 Nxe2 b6 17 Rd2 Ba6 18 b3 Rad8.
15 Kh1
Leaving g1 open for his bishop, in case Black plays ... Bg5.
15 ... Nhf4 16 g3 Nxe2 17 Nxe2 h5?!

Question: Why criticize this move? Tal obviously hopes to generate a kingside attack.

Answer: This move presumes an unearned privilege. Our individual styles make it so
that one player may consider a point – like generating an attack – as a desperate need, while
other players may consider it a trivial matter. Botvinnik’s opening choice paid off, since Tal
impatiently plays for an attack, underestimating the danger to his own king, by launching his
would-be assault, without preamble. 17 ... b6 is about even, although I would take Black if
given a choice.
18 Qf2 b6 19 f4!
The most low-down, dirty trick an opponent can play on us, is to introduce a note of
realism into our blissful attacking fantasy. Botvinnik’s superior strategic understanding
recognizes that it is Black’s king who is in greater danger.
19 ... exf4!?
It’s never easy to say goodbye to our own misconceptions. Tal, still believing himself to
be the one attacking, impulsively allows White an open g-file. It’s understandable that he
rejected passive defence after 19 ... Bb7 20 f5 with an unpleasant position for Black.
20 gxf4 Bb7 21 e5?!
The wrong push, which allows Black back into the game. White achieves a clear
advantage if he plays 21 f5! Ng7 (21 ... Ng5 22 Nf4! Nxe4 23 Qg2 Ng5 24 Rde1 gives
White a winning attack) 22 fxg6 fxg6 23 Nf4 c5 24 Qg2 Bxb2 25 Nd5! Bxd5 26 exd5 Rxf1+
27 Rxf1 Rf8 (27 ... Qxe3?? walks into mate after 28 Qxg6) 28 Rxf8+ Qxf8 29 Bxg6 when
White’s bishops rule the board.
21 ... c5+ 22 Rd5!!

Powerful positional play. Sometimes we are drawn to an idea’s sheer audacity. In this
case, the idea is completely sound. One of the most difficult decisions in a game is resource
allocation. In this case Botvinnik correctly offers an exchange to strengthen his centre. The
roles of the players are curiously reversed from normal, where it is Botvinnik who portrays
the dreamer, and Tal, oddly, is the pragmatist. Instead, 22 Kg1?! Bh4 23 Ng3 f5 is slightly in
Black’s favour.
22 ... Bg7 23 Kg1 Nc7!
Tal agrees to take the exchange but only with the knight, rather than his influential light-
squared bishop.
 24 Nc3!?
Botvinnik refuses to back down with 24 Rd6 Rfd8 25 Rxd8+ Rxd8 26 Nc3 Qe6 27 b3
g5!? when Black begins the process of undermining White’s centre. Chances look about
even.
24 ... Nxd5
Tal finally accepts Botvinnik’s challenge.
25 cxd5
White’s monster centre obviously defrays the spiralling cost of his war.
25 ... Rad8 26 Be4
26 d6?? Qe6 only endangers White’s king.
26 ... Ba8
Just in case White tries to pull off a d6 trick.
27 Qg3 b5?

Exercise (planning): Tal’s old weakness again: impatience. He refuses to

do nothing and wait (which happens to be his best plan). Instead he
misguidedly activates his queenside pawn majority. In doing so, he
self-undermines a key square. How did Botvinnik exploit Tal’s last move?

Answer: Simply pile on to c5, which falls.

28 Qf2! Qd7
28 ... Rc8? allows 29 Nxb5 with a winning advantage.
29 Bxc5 Rfe8 30 Qg3
More accurate is 30 a3.
30 ... Rc8
Tal missed an idea to activate his position with 30 ... a5! 31 a3 b4 32 axb4 axb4 33
Bxb4 Rb8 34 Ba3 Rb3. White’s advantage has been reduced and his pieces placed on the
defensive.
31 b4!
White’s piece and pawn are worth more than Black’s inactive rook.
31 ... Kh8?

Exercise (combination alert): Tal’s last move was a blunder.

White has access to a decisive continuation. What is it?

32 Qf3?
Botvinnik falls prey to the disease of all positional players: we wait and manoeuvre,
when we should strike.
Answer: White achieves a winning position with 32 e6!, annihilating the defensive
barrier:
a) 32 ... Rxe6 33 dxe6 Qxe6 34 Bxa8 Rxa8 35 Qf3 Rd8 36 f5 when White is up a piece
and attacking, while Black is left with only fishing chances. However, the alternatives are
even more hopeless.
b) 32 ... fxe6? 33 Qxg6! (this queen is the kind of woman our mothers warned us about)
33 ... Bxc3 34 Kh1!! (Black has no defence to this deadly quiet move) 34 ... Qg7 35 Qxh5+
Kg8 36 Rg1 and Black can resign.
c) 32 ... Qb7? 33 f5! Qb8 34 Qg5 Qe5 35 exf7 Red8 36 Qxg6 (threat: f6) 36 ... Qf6 37
Qxh5+ Qh6 38 Qxh6+ Bxh6 39 Bd4+ Bg7 40 f6 Bh6 41 Rf4!! (attraction; threat: Rh4) 41 ...
Bxf4 42 f8Q+! (clearance) 42 ... Rxf8 43 f7+ (clearance again!) 43 ... Be5 44 Bxe5 mate.
32 ... a6 33 Kh1?!
Botvinnik’s inaction continues to erode his hard won advantage. 33 e6! still favours
White after 33 ... fxe6 34 Bxg6 Bxc3 35 Qxc3+ Qg7 36 Qxg7+ Kxg7 37 Bxe8 Rxe8 38 d6.
White owns a deeply entrenched and extra passed d-pawn. I don’t think the presence of
opposite-coloured bishops will be enough to save Black here.
33 ... f5!

This shot breaks up White’s imposing centre, at the cost of a pawn – a bargain for Black.
34 exf6 Bxf6 35 Bxg6 Qg4?!
Black may well save himself with 35 ... Bxc3! (the bishop begins Operation Fresh Start,
with a song on his lips and a light heart) 36 Qxc3+ Qg7 37 Qxg7+ Kxg7 38 Rg1 Kf6 39
Bd4+ Ke7 40 Re1+ Kd7 41 Bxe8+ Rxe8 42 Rxe8 Bxd5+ 43 Kg1 Kxe8 44 a3 Kf7 and Black
easily draws due to the opposite-coloured bishops.
36 Qd3! Rg8 37 Be4
After 37 Rg1! Qxf4 38 Qd1! Black must return the exchange, since 38 ... Qh4?? is met
with 39 Bf2! Qh3 40 Rg3 Qh4 41 Qd2 (threatening mate on h6) 41 ... Bg7 42 Ne4 when the
dual threats of Bd4 and Ng5 are decisive.
37 ... Rce8 38 Bf3?!
Time pressure addicts like Botvinnik (and unfortunately, also your writer) seem to thrive
on self-imposed hardship. With this move Botvinnik throws away the entirety of his
advantage, on the eve of time control. He still retains an edge after 38 Nd1! Rg7 39 a3 Bb7
40 Ne3 Qh3 41 Rf2 with two pawns for the exchange and an active position.
38 ... Qxf4 39 Ne2
White minimizes Black’s edge with 39 Be4! Qh4 40 Bf2 Qg4 41 Qe3 Qg5 42 Qxg5
Rxg5 43 Be3 Bxc3 44 Bxg5 Bxb4 45 Bg2.
39 ... Qh4
39 ... Qc4! 40 Qf5 Qh4 41 Qd3 Be5 42 Rf2 Rg7 is slightly in Black’s favour.
40 Bf2 Qg5 41 Ng3
Worrying Black about Ne4 ideas, while pressuring h5.
41 ... Rd8! 42 Be3
 42 Nxh5?? is short circuited by 42 ... Rxd5.
42 ... Qe5 43 Rd1 Rg4!

Tal finds a clever method of activating this rook.

44 a3
44 Bxg4?? Rxd5 is an instant game-ender.
44 ... Bb7 45 Bb6 Rd7 46 Be3 Rh4
Threat: ... Qxg3.
47 Nf1?!
White’s passivity level continues to rise. 47 Kg1 puts up greater resistance.
47 ... Rc4 48 Bg2 Rg7 49 Qd2 h4
Tal imperceptibly makes progress, introducing ideas like ... Qh5, followed by ... h3,
while simultaneously preventing Ng3.
50 h3
A concession, weakening g3.
50 ... Qb2
The haze distorts, so that Tal experiences difficulty perceiving a variation’s true
distance. He uncharacteristically opts for a winning technical ending, when he should keep
queens on the board. 50 ... Qh5! 51 Bc5 Rxg2 52 Kxg2 Qg6+ 53 Kh1 Rc2 is immediately
decisive.
51 Qxb2 Bxb2 52 Bc5 Rd7 53 Ne3 Rc1!
Removing a key defender of d5.
54 Rxc1 Bxc1 55 Bd4+ Kg8 56 Ng4 Bg5
Cutting off a cheapo check on f6.
57 Kg1
Likewise, 57 Bf6 Bf4 58 Bxh4 Kf7 59 Nf6 Rc7 60 Bf3 Rc3 61 Kg2 Rxa3 62 Bh5+ Kg7
should be decisive for Black.
 57 ... Bxd5 58 Ne5 Bxg2
Tal refuses to hoard every penny and decides to simplify into a pawn up ending, with the
bishop-pair. However, it begins to get tricky since there are so few pawns remaining on the
board. Slightly more accurate is 58 ... Rd6! 59 Bc5 Bxg2 (59 ... Rd8 60 Bb6 Rd6 61 Bc5
Bxg2 comes to the same) 60 Bxd6 Bxh3 is a similar ending to what happened in the game,
except that White’s knight isn’t as well placed in this version.
59 Nxd7 Bxh3 60 Nc5 Bc8
Now comes the purely technical stage.
61 Kf2 Kh7 62 a4!

Endgame principle: Pawn swaps help the defending side.

62 ... bxa4 63 Nxa4 Bf4 64 Kf3 h3! 65 Bg1 h2
Winning a piece, but the story is not over, since Tal will remain with only one pawn on
the board, which dramatically increases Botvinnik’s drawing chances. There was no rush to
win the piece. I think stronger was 65 ... Bd2!, picking off b4, and remaining two pawns up.
66 Bxh2 Bxh2
If Botvinnik can remove the final pair of pawns from the board, Black’s winning task
increases exponentially.
67 Ke4 Bd7 68 Nc5 Bb5 69 Kd5 Kg6
Black’s biggest issue is his distant king, who is unable to help secure his final pawn.
70 Ne4 Kf5 71 Nc3 Bf1 72 Kc5 Be5 73 Nb1 Ke6 74 Nd2 Bd6+ 75 Kb6 Bg2
Not 75 ... Be2 76 Nb3 Bxb4?? (the bishop reneges on his vow of poverty) 77 Nd4+ and
the game is drawn.
76 Nb3
After 76 Kxa6 Bxb4 77 Nb3 Bf1+! 78 Kb7 Kd7 79 Kb6 Kd6 80 Kb7 Bc3 81 Kb6 Bc4
82 Na5 Bd4+ 83 Kb7 Be6 84 Ka6 Kc5 85 Nb7+ Kb4 86 Nd8 Bc8+ 87 Nb7 Kc4 the knight
falls.
 76 ... Bxb4 77 Kxa6
It isn’t easy to get our bearings in pawnless positions, since we lose our marker which
distinguishes inside from outside. The scenery, like the stage of an old, abandoned theatre,
never seems to alter.

Question: Is this a draw, or a win for Black?

Answer: Such stark landscapes look like a dead alien world, completely devoid of
human meaning. The tablebases tell us it’s a win for the bishops’ side, but in 75 moves! I
once reached a two bishops versus knight ending in the final round of an American Open. I
explained to the director that the comps worked out a win for the bishops in 75 moves. The
director told me the U.S. Chess Federation rules stipulate that you have only 50 moves to
win if there are no captures on the board (I believe FIDE gives you the necessary 75
moves). I tried and actually trapped his knight, but was forced to cede the draw when my
opponent reached the 50-move mark with his knight still alive on the board.
77 ... Bf1+ 78 Kb6 Kd6 79 Na5 Bc5+! 80 Kb7 Be2!
 The key to winning these positions:
1. Force the knight away from its king.
2. Tie up the knight so that it is denied movement.
3. Walk your king over to help trap the knight.
This all sounds very clear, but I assure you, it isn’t easy to pull off the win at the board.
According to the comps, Tal played this difficult ending to near-perfection.
81 Nb3 Be3 82 Na5 Kc5 83 Kc7 Bf4+! 0-1
Butlers are skilled in telling unwanted guests “No! Please go away!”, but in a very
polite way. Botvinnik’s king is gently ushered away from his knight. The details of the
autopsy indicate: 84 Kc8 Kb5 85 Nb3 (now knight and king have been separated, but 85
Nb7 Kb6 86 Nd8 Bg4+ wins the knight next move) 85 ... Be3! (step 2 accomplished; the
bishops continue to smear their nasty influence, and the knight is cornered) 86 Kd7 Bd1 87
Na1 (this is the classic Mammoth-in-the-tar-pit situation) 87 ... Bd4 (“I learned long ago that
prayer doesn’t always achieve your desire,” the bishop instructs the unfortunate knight), and
so on and so forth.
Step 3. The knight falls. This up and down epic is my favourite game of the 1961
rematch with Botvinnik.

Game 28
M.Tal-H.Hecht
Varna Olympiad 1962
Queen’s Indian Defence

Tal’s games are always played over with some degree of incredulity. This is one of the most
mind-bending of the book.
 1 d4 Nf6 2 c4 e6 3 Nf3 b6 4 Nc3
Tal really isn’t a Karpovian Queen’s Indian type of player, so he switches back to a
sharper line of the Nimzo-Indian.
4 ... Bb4 5 Bg5 Bb7 6 e3 h6 7 Bh4 Bxc3+ 8 bxc3
Black gives up the bishop-pair to damage White’s structure. Theory considers Black’s
chances to be fully adequate at this stage.
8 ... d6 9 Nd2 e5
More recently, after 9 ... Nbd7 10 f3 Qe7 11 Qa4 e5 (Black tends to play this move
sooner or later) 12 Bf2 0-0 13 Be2 Nh7 14 Bd3 f5 15 Qc2 Qf7 16 e4 fxe4?! (Black should
keep the position blocked with 16 ... Qg6 17 0-0 f4) 17 Nxe4 Nhf6 18 Ng3 the position
opened in favour of White’s bishops, Wang Yue-E.Tomashevsky, Sochi 2009.
10 f3 Qe7
Threat: ... exd4 and ... Qxe3+.
11 e4 Nbd7 12 Bd3 Nf8

Question: What is the point of this retreat?

Answer: The move is actually the main line, and also Komodo’s top choice in the
position. The idea is to break the f6 pin with ... Ng6, or play ... g5, followed by ... Ng6. The
slight trouble with the move is that it weakens c5, a fact which Tal immediately attempts to
exploit.
13 c5!?
For Tal to play safely is almost a contradiction in terms. This disruptive move was a
new idea at the time.
 Question: Doesn’t it drop a pawn?

Answer: In such structures, the c5 break is a standard issue theme for White. It’s a
deliberate sacrifice, based on the principle: Open the position when you control the
bishop-pair.

Question: Is it sound?

Answer: I’m not so quick to vouch for this gambit. As we all understand by now, a
line’s credibility never worried Tal much, if in exchange he seized an initiative and created
confusion. Objectively, the move is dubious. Safer (and duller!) is 13 Nf1 Ng6 14 Bf2 Nf4
15 Ne3 Nxd3+ 16 Qxd3 Nh5 17 g3 0-0 18 0-0 Qe6 19 Rfd1, E.Bacrot-Z.Almasi,
Torrelavega 2007. Black only looks a shade worse after 19 ... Nf6.
13 ... dxc5
Black must accept, since:
a) 13 ... Ng6?! 14 cxd6 cxd6 15 Bb5+ Kf8 16 Bf2 allows White to undouble his pawns
and seize the initiative, without the cost of a pawn.
b) 13 ... exd4 14 cxd6 Qxd6 15 Qa4+ c6 16 0-0! with a dangerous attack for White.
Now if 16 ... dxc3?? (best is the line 16 ... Ne6 17 Nc4 Qc5 18 cxd4 Qxd4+ 19 Bf2 Qxd3
20 Rfd1 b5 21 Rxd3 bxa4 22 Nd6+ Ke7 23 Nxb7 with advantage to White) 17 Nc4! Qxd3
18 Rad1 b5 19 Rxd3 bxa4 20 Nd6+ and wins.
14 dxe5
In this way Tal gets a mobile kingside pawn majority, while opening the game further for
his bishops.
14 ... Qxe5 15 Qa4+
 This disruptive check also discourages Black from castling long.
15 ... c6
I think this move is slightly inaccurate. After 15 ... N6d7! White is hard pressed to prove
full compensation for the pawn:
a) 16 Nc4? is way too aggressive: 16 ... Qxc3+ 17 Ke2 Ne6 (threatening a big check on
f4, as well as d4) 18 Bg3 Nd4+ 19 Ke3 b5 20 Qa5 Nc2+ 21 Ke2, B.Zuckerman-E.Mednis,
Cleveland 1975. Now, for some inexplicable reason, Mednis took a perpetual check by
moving his knight back and forth, from c2 to d4. Instead, Black wins easily with 21 ... Nb4!
22 Qa3 Qxd3+ 23 Qxd3 Nxd3 24 Na5 Bxe4 25 fxe4 N7e5 leaving White down three pawns
in the ending.
b) 16 Rc1 Qe6? (16 ... Ng6! 17 Bg3 Qg5 18 Rd1 0-0-0 and Black stands clearly better,
since White can’t play 19 Qxa7?? Qe3+ 20 Be2 Nde5 with a winning attack) 17 0-0 Bc6 18
Qa6 Ne5 19 Bc2 g5 20 Bg3 Nfg6 and I don’t believe in White’s compensation for the pawn,
G.Garcia-A.O’Kelly de Galway, Havana 1963.
16 0-0 Ng6
Comps may grab c3, but humans can just ‘feel’ the danger, and most of us wouldn’t even
consider 16 ... Qxc3!? 17 Nc4! Ng6 (not 17 ... Qxd3?? 18 Rfd1 Qc3 19 Nd6+ Ke7 20 Nf5+
Ke8 21 Nxg7+ Ke7 22 Nf5+ Ke8 23 Rac1 when Black, faced with multiple threats, can
resign) 18 Nd6+ Kf8 19 Bxf6 Qxd3 20 Nxb7 gxf6 21 Rad1 Qe3+ 22 Kh1. Black remains
alive, but I still wouldn’t take on this position.
17 Nc4!?
Tal again offers c3.
17 ... Qe6
Now what? Black threatens ... b5, as well as ... Nxh4. Black can also accept the
challenge with 17 ... Qxc3 18 Nd6+ Kf8 19 Bxf6 Qxd3 20 Nxb7 gxf6 21 Qxc6 Kg7 22 Nd6
Ne7 23 Qc7 Qd4+ 24 Kh1 Rhd8 25 Rad1 Qe5 26 f4 Qe6 27 Qb7 Rab8 28 Qxa7 Nc6 29
Qc7 Nd4 30 Nf5+ Kh7 31 Nxd4 cxd4 and I don’t believe Black stands worse.
18 e5!
Tal ignores both threats and generates a few of his own.
18 ... b5!
18 ... Nxh4?! is met with 19 Nd6+ Kf8 20 Rae1 g5 21 Nxb7 Nd5 22 g3 Nxc3 23 Qc2
(both black knights hang) 23 ... Qxa2 24 Qxa2 Nxa2 25 gxh4 Nb4 26 Rd1 a5 27 Be4 Ke7 to
prevent Rd7. 28 Nd6. I prefer White’s extra piece to Black’s four connected queenside
passers, mainly since White isn’t done attacking.
19 exf6!?

Huh?

Question: Doesn’t a sentence require a subject and an object?

Answer: Your normally verbose writer – the envy of politicians worldwide – is

inexplicably stunned into incoherent mumbling. It isn’t easy to ‘tell’ the white queen what
she may and may not do. The desire to complicate at all costs may be clouding Tal’s
judgement, yet we, the spectators, are very glad he took this route, which offered his
imagination free reign. If we commit a crime, the judge isn’t going to buy our ‘I-promise-
not-to-do-it-again’ defence. Tal rejects the strategic route and instead, gives up his queen!
It’s as if Tal is on the computer and a message pops up, and he presses the ‘Remind me later’
key, to reality itself.
White secures an edge with the objectively stronger line 19 Qb3! (of course, Tal nearly
always set aside practicality in favour of a more imaginative idea) 19 ... Nxh4 20 Nd6+ Kf8
21 Qxe6 fxe6 22 Nxb7 Nd5 23 Nxc5 Ke7 24 g3 Nf5 25 Rfc1 when c4 is coming and I prefer
White’s chances in the ending, mainly due to his powerfully posted c5-knight.
19 ... bxa4
 Now Black’s pawn structure warps and loses its original shape, like summer’s spoiling
fruit in the bowl. After 19 ... 0-0 20 Qc2 Nxh4 21 Na5 Rab8 22 fxg7 Kxg7 23 Qf2 c4 24
Bxc4 bxc4 25 Nxb7 Rxb7 26 Qxh4 Black’s ragged structure offers White some winning
chances.
20 fxg7 Rg8

Exercise (combination alert): A glance at the position hints to us that

White’s compensation for the queen feels hollow and insubstantial. It
appears as if the bleating lamb is tied to her e6 outpost, while the hunter
hides in ambush for the hungry tiger. This view fails to correspond with reality,
and impossibility is just a move away. How would you continue Tal’s attack?

Answer: Deflection/removal of the guard/knight fork.

21 Bf5!!
“If my heart is pure, then no harm can come to my body,” reasons the bishop. The once
derelict attack revives. If chess is a purely mental game, then why do we all feel that awful
jolt which runs through our body when we get hit with an unexpected shot? “Here my
composure finally left me. It took minutes before I was in any fit state to calculate the
following variations,” wrote the exasperated Hecht.
The rote 21 Rfe1?? fails miserably to 21 ... Nxh4 22 Rxe6+ fxe6 23 Nd6+ Ke7 24 Nxb7
Rxg7 25 g3 Nxf3+ 26 Kg2 Rb8 27 Ba6 Rf7 when Black is up an exchange and converts
easily. However, 21 Bxg6!? is possible: 21 ... fxg6 22 Rfe1 Bc8 23 Nd6+ Kd7 24 Ne4!!
(better than 24 Rxe6 Kxe6 25 Bg3 when White does get some compensation for the
exchange) 24 ... Qf7 25 Bg3! (principle: when hunting the enemy king, don’t chase it;
instead, cut off potential escape routes) 25 ... Kd8 26 Rad1+ Bd7 27 Be5 g5 28 Bf6+ Kc7
29 Be5+ and Black has nothing better than to take the draw with 29 ... Kd8 30 Bf6+.
 21 ... Nxh4
Hecht musters an adequate reply, finding the second best defence, which is actually
something of an achievement in his flustered state:
a) 21 ... Qxc4?? works out disastrously for Black after 22 Rfe1+ Ne5 23 Rxe5+ Qe6 24
Rxe6+! fxe6 25 Bg6+ Kd7 26 Rd1+ Kc8 27 Bf6 and there is no good defence to the coming
Bf7.
b) 21 ... Ba6! is Black’s best chance: 22 Bxe6 fxe6 23 Nd6+ Kd7 24 Ne4 Bxf1 25 Nf6+
Kc7 26 Nxg8 Rxg8 27 Bg3+ (this way White saves his own hanging bishop, while picking
up Black’s on f1) 27 ... e5 28 Kxf1 Rxg7 29 Re1 Kd6 30 Rd1+ Ke6 31 Bf2 c4 32 Bc5 Rb7
33 Rd6+ Kf7 (33 ... Kf5?? 34 g4+ Kg5 35 h4+! wins a piece, since Black is unable to play
35 ... Nxh4?? 36 Be3 mate) 34 g3. The complicated ending is dead even according to
Komodo.
c) 21 ... Qxf5?! 22 Nd6+ Kd7 23 Nxf5 Nxh4 24 Nxh4 Rxg7 25 Rfe1 is a clearly
superior ending for White.
22 Bxe6 Ba6!
Avoiding 22 ... fxe6?! 23 Nd6+ Ke7 24 Nxb7 Rab8 25 Nxc5 Rxg7 26 g3 Rb5 27 Nd3
Rd5 28 Rad1 with a clear structural advantage for White.

Exercise (combination alert): How did Tal manage to extract his

hanging pieces safely? White to move and not lose material:

Answer: Zwischenzug/defensive move.

23 Nd6+! Ke7
The black king gazes at the floor, like a scolded yet still defiant six-year-old. Three of
Tal’s pieces hang simultaneously, yet he foresaw a way to save all of them.
24 Bc4!
 This bishop moves about as nimbly as if there were three of him. This is just magic, the
way the variations keep working in Tal’s favour, despite his impossible-looking difficulties.
24 ... Rxg7 25 g3 Kxd6 26 Bxa6
If impermanence runs through an entity, then it is automatically subject to decay:
1. Tal is only down a single pawn, and his vastly superior structure offers him the edge,
since Black will be hard pressed to hang on to all his pawns.
2. Tal’s bishop is likely to outgun Black’s knight.
3. Black’s active king still offers him chances to hold the game.
26 ... Nf5?!
Inaccurate. Black’s best chance to save the game is to secure the b-file with 26 ... Rb8!
27 Rad1+ Kc7 28 Rde1 f5 29 f4 Ng6 30 Bd3 Ne7 31 Re5 Rd8 32 Bxf5 Rd5 33 Bh3 Rxe5
34 fxe5 Rg5 35 Re1 Nd5 36 e6 Rg8. Black’s active king should be enough for him to draw.
27 Rab1!
Now Rb7 is a serious threat.
27 ... f6?!
With time pressure approaching, our priority is speed, at the expense of care. 27 ... Ne3!
may yet be enough to hold the position after 28 Rf2 Nd5 29 Rd2 Ke7.
28 Rfd1+ Ke7 29 Re1+
29 Bb7! Rb8 30 Bxc6 Rxb1 31 Rxb1 a3 32 Kf2 would be very tough for Black to hold.
29 ... Kd6 30 Kf2 c4?!
An overreaction. Black puts up greater resistance with 30 ... Rd7 31 Re4 Ne7 32 Rxa4
Nd5 33 Ra3 Re7 34 Bd3, which offers White winning chances.
31 g4?!
This hasty move places White’s gains at risk, allowing Black’s knight back into the fight.
31 Re4! looks more promising for White.
31 ... Ne7 32 Rb7 Rag8!
 Tal said he missed this defensive idea.
33 Bxc4 Nd5 34 Bxd5 cxd5 35 Rb4?!
Another slip, which may allow Black enough counterplay to hold the game. After 35
Reb1 f5 36 h3 Rxb7 37 Rxb7 Rc8 38 Rxa7 Rc4 39 Ra6+ Ke5 40 Rxh6 fxg4 41 hxg4 Rxc3
42 Kg3 a3 43 Rh5+ Kd4 44 Rh8 Kd3 45 g5 White is faster in the race. If necessary, he can
sacrifice his rook for Black’s d-pawn and then his two connected and advanced passers
should beat Black’s rook and distant king.
35 ... Rc8?!
An opportunity, once missed, rarely repeats itself. 35 ... f5! 36 h3 h5! may still save
Black: 37 Kg3 hxg4 38 fxg4 fxg4 39 Rxg4 Rxg4+ 40 hxg4 a3 41 Rb1 Ke5 42 Rb3 Rc8 43
Rxa3 a5 and I don’t see how White makes progress.
36 Rxa4 Rxc3 37 Ra6+ Kc5 38 Rxf6 h5 39 h3 hxg4 40 hxg4
White’s two connected passed pawns are decisive.
40 ... Rh7
I would have followed the endgame principle: Passed pawns must be pushed, with 40
... d4.
41 g5 Rh5 42 Rf5 Rc2+ 43 Kg3 Kc4 44 Ree5!
Intending g6.
44 ... d4 45 g6 Rh1 46 Rc5+ Kd3 47 Rxc2 Kxc2 48 Kf4
Avoiding the moronic blunder 48 g7?? Rg1+ 49 Kf2 Rxg7 when it is Black who wins.
48 ... Rg1

Exercise (calculation): A forlorn pocket of black

resistance remains. How did Tal sweep it aside?

Answer: Interference. White wins the promotion race.

 49 Rg5! 1-0
If 49 ... Rxg5 50 Kxg5 d3 51 g7 d2 52 g8Q d1Q 53 Qb3+ Kd2 54 Qxd1+ Kxd1 and the
f-pawn promotes.

Game 29
M.Tal-R.Letelier Martner
Havana 1963
Ruy Lopez

1 e4 e5 2 Nf3
2 Bc4 Bc5 3 Qh5 (my father’s move represents a literalist interpretation of the position)
3 ... Nc6 (I suspect that some improvement can be found here, for Black’s side) 4 Qxf7 mate
was F.Lakdawala-C.Lakdawala, Montreal 1968. This was my very first chess game (which
should have been a hint that I had no natural ability and should have taken up some other
activity, like basket weaving!). Tal tries a variant on this theme against an IM!
2 ... Nc6 3 Bb5 a6 4 Ba4 d6 5 c3 Bd7 6 d4 Nge7
The idea can be to follow up with two potential developmental plans:
Plan a) Play ... g6, ... Bg7, ... 0-0 and maybe later ... f5.
Plan b) Play ... Ng6, ... Be7 and ... 0-0.
7 Bb3
Threat: Ng5.
7 ... h6 8 Nh4!?

Tal dreams an uninterpretable dream, letting go of all calculations and measurements,

and just trusting his attacking instincts. The orderly sequence goes askew, and a new idea is
born from the fusion of a Ruy Lopez and a Scholar’s mate.
 Question: What is the point of White’s amateurish-looking move?

Answer: Tal’s last move – as awkward as a horse-drawn carriage attempting a U-turn

on a busy city intersection – clearly feels like a derivative element, missing the position’s
full truth. The idea is to prevent ... Ng6, while threatening Qf3, with a Scholar’s Mate threat
on f7, which is surprisingly difficult to deal with. Now saying this, I don’t really think Tal’s
8 Nh4 is such a great move. But as we all understand about Tal’s games: when an idea hangs
upon imagination, rather than reality, this doesn’t mean it won’t succeed over the board, due
to the confusion/fluster factor. Just what this position will become is uncategorized, for the
moment.
8 ... g5?
Disagreement flares into open warfare. Letelier incorrectly decides this is a tempo-
gaining opportunity just too tempting to pass up. This weakening thrust is an overreaction
and he never does gain the expected tempo. Alternatives:
a) 8 ... g6 has never been played. It is met with the amateurish yet strong 9 Qf3!, forcing
Black into an awkward stance with 9 ... Rh7 10 d5 Na5 11 Bc2. I prefer White’s position.
b) I think 8 ... exd4! may be Black’s best line, utilizing the principle: Counter in the
centre when attacked on the wing: 9 cxd4 (the Goring Gambit-like move 9 0-0!? fails to
achieve its end after 9 ... dxc3 10 Nxc3 g5!; I prefer Black’s chances after 11 Qh5 Ne5) 9 ...
Nxd4! (small details are a gauge for a position’s big picture; this trick maximizes Black’s
otherwise scarce resources) 10 Qxd4 Nc6 (the double attack regains the piece) 11 Qd5
Qxh4 12 Qxf7+ Kd8 was W.Browne-B.Ivkov, Nice Olympiad 1974. Chances look about
even after 13 Nc3 Ne5 14 Qd5 (14 Qf4?? walks into a knight fork after 14 ... Nd3+) 14 ...
Bc6 15 Qd1! (15 Qd4?! Qg4! is a problem for White, who must submit to 16 Kf1 with an
inferior position) 15 ... Bxe4 16 0-0. White looks like he gets adequate attacking chances for
the pawn, since Black’s king doesn’t look all that secure on d8. Still, a pawn is a pawn, and
if given a choice, I would probably take Black’s side.
c) 8 ... Na5 9 Bc2 c5 10 dxc5 (10 d5? is met with the discovery shot 10 ... Nxd5 11 exd5
Qxh4; Black picked up a pawn and White’s development lead fails to fully compensate) 10
... dxc5 11 Qf3 Ng6 12 Nf5 Qf6 was M.Tal-P.Keres, Moscow 1973. Tal stood a shade
better, due to the d5 hole.
9 Qh5!
My father must have been influenced by this Tal game. Let’s all drop the pretence of
sophistication and admit that we love to threaten vulgar one move mates! White’s attack for
now looks more like a band of brigands than a formal army.
9 ... Rh7
The only viable move to cover the threatened mate.
 Exercise (combination alert): Tal found a mini combination here. What did he play?

Answer: Pin.
10 Bxg5!
Forward moving pieces represent a rising demographic on the kingside. This move
doesn’t win a pawn, but instead, strips Black’s king of protection, while bringing in another
attacker.
10 ... exd4 11 f4
Tal cuts out defences based on ... Ne5. Even stronger was 11 Nf5!.
11 ... Qc8
Threatening to trap White’s queen with ... Bg4, as well as ease the pressure with a
soothing exchange, with ... Be6.
12 f5!
Tal gives something (control over e5) to get something (a clogging of Black’s light-
squared bishop and queen, which cuts out ... Be6 ideas).
12 ... dxc3
The comp suggests the inhuman idea 12 ... Ne5!? 13 cxd4 Nd3+ 14 Kd2 Nxb2 15 Nc3
c5 16 Rac1 when White clearly has the more promising attack.
13 Nxc3 Ne5 14 Bf6!?
Intending to eliminate Black’s best piece. 14 Be3 is also quite promising.
14 ... Ng8?!
The right idea, but one move early. Black should try 14 ... Nd3+ 15 Kd2 and only now
play 15 ... Ng8! 16 Bg5 Nc5 17 Rae1 Rg7 18 Be3 Qd8 19 Kc2 Nxb3 20 axb3 Qe7. Black
castles long and the worst is behind him.
 15 Bxe5!

Question: I understand that the motivation is to eliminate

Black’s powerful e5-knight. But in doing so, doesn’t White
hand over both the bishop-pair, and also the dark squares?

Answer: All true, yet strategic considerations are swept aside, since Tal’s move is in
preparation for a coming combination.
15 ... dxe5 16 Ng6!
Tal could never resist such moves. White can also play the simpler 16 Nf3! Nf6 (16 ...
Bd6?? is instantly crushed by 17 Ng5!) 17 Qh4 Qd8 18 Nxe5 Qe7 19 Nf3 Bxf5 20 0-0-0
when Black’s king is in serious danger, stuck in the centre.
16 ... Bd6
 Exercise (combination alert): We can evolve and refine an idea only so much.
At one point we must implement it. How did Tal prosecute his attack?

Answer: Attraction/discovered check/pin.

17 Bxf7+!
Danger, which once lurked on the periphery, now steps out into the open and stares
directly at Black’s king. It wasn’t in Tal’s nature to suppress any idea (no matter how
risky!), even when implementation meant the sacrifice of material.
17 ... Kxf7
Forced, since 17 ... Rxf7?? loses to the Nimzowitschian 18 Nh8!.
18 Nd5!
Tal spurs his horses headlong into the battle zone. Lasker’s principle: The threat (of
discovered check) is stronger than the execution. The d5-knight can’t easily be kicked
away, since ... c6?? walks into a fork on b6. Also when Black plays ... Nf6, White then
exchanges, which saves his queen from a tempo loss. Tal’s move is more accurate than 18
Nxe5+ Kf8 19 Ng6+ Kg7 20 0-0 Qd8 when f6+ which is prevented, while Black gets help
with ... Qg5 next.
18 ... Kg7
Forced, since Nxe5+ was huge, with White’s other knight on d5.
19 0-0
Threat: f6+.
19 ... Nf6 20 Nxf6 Kxf6
 Exercise (combination alert): Black’s position is not a good place
to be if you are playing Mikhail Tal. Find White’s next shot:

Answer: Clearance/attraction.
21 Nxe5! Qe8!
The only move. Both versions of acceptance are disastrous for Black:
a) 21 ... Bxe5?? 22 Qg6+ Ke7 23 Qxh7+ is hopeless for Black’s harried king.
b) 21 ... Kxe5?? 22 f6+ Bf5 (22 ... Ke6?? 23 Qd5 mate) 23 Rxf5+ Ke6 24 Rf3! (threat:
Qf5+ and Qxh7+) 24 ... Rh8 25 Rd1! when Qd5+ and then e5 follows, and Black’s king has
no hope of survival. 25 ... c6 is met with 26 Qg4+ Kf7 27 Qg7+ Ke6 28 Rxd6+! Kxd6 29
Qe7 mate.
22 Nxd7+ Rxd7 23 e5+!
Tal finds one blow after another, this time opening central lines to hunt down Black’s
king.
23 ... Bxe5
Likewise, 23 ... Qxe5 24 Qg6+ Ke7 25 Rfe1! (Larsen once wrote that whichever rook
you move, will be the wrong one; it becomes apparent in a few moves this is the right rook)
25 ... Bc5+ 26 Kh1 Be3 27 Qh7+ Kd8 (27 ... Kd6?? 28 Rad1+ and now we see why it was
correct to give check on e1 with the f1-rook, rather than the a1-rook; White wins) 28 Qg8+
Qe8 29 Qb3! is hopeless for Black, since 29 ... Re7?? gets clobbered by 30 Qxb7 Rc8 31
Rad1+ Rd7 32 Qb3 Rxd1 33 Rxd1+ Ke7 34 Qxe3+. His sister’s coarse, jeering laughter
assaults the black king’s delicate ears.
24 Qxh6+ Kf7 25 Rae1
The final inactive attacker enters play.
25 ... Rd5?
 When we linger in an inferior position, we reach a place which is neither life nor death.
Necessary was 25 ... Rad8 26 Kh1 Rd5 27 Qh7+ Kf6 28 g4 Qg8 29 Qh6+ Ke7 30 f6+ Kd7
31 f7 Qg7 32 Rxe5! Qxe5 33 f8Q Rxf8 34 Qxf8 Qxb2 35 Qf7+ Kc6 36 Qe6+ Kc5! 37 g5!
when White should win, but must still be careful, due to his own exposed king.
26 Qh7+!
“Come with me if you want to live,” recites the queen, who is a big fan of the
Terminator movies.
26 ... Kf6

Exercise (combination alert): Find one problem-like idea and Black collapses:

Answer: Rook lift/queen trap.

27 Re4!!
Threat: Rg4 and Rg6 mate.

Question: Wasn’t Tal’s last move a blunder?

Black has the discovery combination 27 ... Bd4+!.

Answer: We must borrow from the wisdom of Howard Hughes: Just because you can’t
see a germ, doesn’t mean it won’t get you. Please see the game continuation!
27 ... Bd4+ 28 Kh1! 1-0
Sometimes the worst-case scenario is when our wishes are answered. In hopeless
circumstances, we tend to cling to the comforting fiction that a miraculous rescue is still
within the realm of possibility. Black’s ‘saving’ move 28 ... Qxe4 is met with 29 Qg6+ Ke7
30 f6+! and Black’s unfortunate queen reminds us of figure skater Nancy Kerrigan, just
moments before she received the Tonya-ordered metal pipe to the knee.
Game 30
M.Tal-E.Vasiukov
USSR Championship, Kiev 1964
Caro-Kann Defence

1 e4 c6
A surprise choice from Vasiukov. Of course today, the Caro-Kann is a mainstream
opening. Tal in his notes called it a “half-forgotten” line, which regained popularity by the
advocacy of Botvinnik, Smyslov and Petrosian.
2 Nc3 d5 3 d4 dxe4 4 Nxe4 Nd7
The ultra-solid Smyslov line of the Caro-Kann.
5 Nf3
More common today are 5 Bc4 and 5 Ng5.
5 ... Ngf6 6 Ng3
A line like this would appeal to Tal, since all the pieces remain on the board. Instead,
after 6 Nxf6+ Nxf6 7 c3 Bg4 8 h3 Bxf3 9 Qxf3 e6 10 Bc4 Be7 11 0-0 Nd5 Kasparov’s
bishop-pair didn’t lead anywhere and Karpov held a draw with ease, G.Kasparov-
A.Karpov, Seville 1987.
6 ... e6 7 Bd3 c5

Black’s thematic freeing break in this line.

Question: Doesn’t it violate the principles: Don’t open the

game or create confrontation when lagging in development?

Answer: It does violate both, yet the move is absolutely playable, since it follows the
principle: If you are cramped, free yourself! We often deal with conflicting principles, and
must then be guided by our intuition. In this instance, Black’s weakness-free position is solid
enough to absorb the opening of the game and the mild confrontation level. Remember that
principles can be deliberately violated when we slightly veer. It’s only when we deviate
widely from the norm that we are begging for trouble.
8 0-0 cxd4
Black’s main move. 8 ... Be7 is the chief alternative.
9 Nxd4 Bc5
Also playable is 9 ... Be7, preserving c5 for the knight.
10 Nf3

Question: By moving to f3 rather than to b3, doesn’t White basically lose a tempo?

Answer: Correct. Tal’s last move constitutes a tempo loss. The reason he rejected the
b3-square is that he had his mind on attack, and wanted his knight on the kingside, rather
than the queenside. 10 Nb3 Be7 11 Re1 0-0 12 Qf3 a5 13 a4 Nb6 was S.Tiviakov-
M.Adams, 1st matchgame, New York 1994. White holds only a sliver of an edge after 14
Nh5 Nbd5.
10 ... 0-0 11 Qe2
11 b3, preparing to aim the dark-squared bishop at Black’s king, is also logical. After 11
... b6 12 Bb2 Bb7 13 Qe2 Qc7 despite White’s ominous looking attackers, I think his edge is
more optical than real, since Black’s position is well constructed to absorb a coming attack,
B.Spassky-M.Filip, Nice Olympiad 1974.
11 ... b6?!
Black creates unnecessary difficulties for himself after this inaccuracy. 11 ... Qc7!
prevents Tal’s next move.
 12 Bf4!
Tal alertly cuts off the c7-square for Black’s queen.
12 ... Bb7 13 Rad1
Now matters are slightly awkward for Black’s queen, who has no safe refuge. The e7-
square is risky, since White may have future Nf5 tricks.
13 ... Nd5
A defender is lured away from Black’s king. 13 ... Qc8 is awkward, but still very
playable:
a) 14 Kh1!? (I don’t understand this move; surely White has more dynamic options than
a waiting move?) 14 ... Rd8 15 Ng5 Be7 16 Nf5 Bf8 17 Nd4 Re8 18 Nb5 e5 19 Bg3 was
M.Chandler-P.Wells, London 1985. I actually prefer Black’s position after 19 ... a6.
b) White may have been worried about 14 Rfe1 Ng4?, but Black’s move is a blunder,
since White has 15 Bxh7+! Kh8 16 Nd4! and White won a pawn. If 16 ... Ngf6 17 Be4 Bxe4
18 Nxe4 Bxd4 19 Rxd4, and now if Black tries 19 ... e5?? 20 Bxe5! Nxe5 21 Nxf6 gxf6 22
Qh5+ Kg7 23 Rxe5! wins.
14 Bg5 Qc7
Black’s queen gets to her preferred square at the cost of allowing Tal to seize control
over f6, with his next move.
15 Nh5!
When a fox (Tal’s knight) attempts to engage in a conversation with a hen (the g7-pawn),
the hen, would be well advised to disbelieve anything the fox tells her. Suddenly, Black’s
king is in need of defenders. The knight eyes potential g7 sacrifices.
15 ... Kh8!

Question: What is the point of Black’s last move?

 Answer: It serves two purposes:
1. Now Bxh7 sacrifices don’t come with check, which gives Black zwischenzug options.
2. Black plans either knight to f6, and if White chops twice, then Black is ready for ...
Rg8, taking aim at g2, while covering his own king.
16 Be4
16 c4 is met with 16 ... N5f6, and if 17 Nxf6 Nxf6 18 Bxf6?! gxf6. White has no
available inroad to Black’s king and can only play to neutralize, with 19 Be4 Rad8, after
which White has less than nothing.
16 ... f6
Slightly more accurate is to prepare ... f6 with 16 ... Rae8 17 Rfe1 (Tal said he planned
17 c4. which is well met by 17 ... f6! 18 Bc1 f5! 19 Bc2 N5f6 20 Ba4 Re7 21 Nxf6 gxf6!; ...
Rg8 follows, and I prefer Black), and only now play 17 ... f6!. White’s bishop must retreat
to c1, since 18 Bh4?! is met with the clever shot 18 ... Ne3! 19 Rxd7 Qxd7 20 Bxb7 Nxg2
21 Kxg2 Qxb7. White is technically up material, with two knights for a rook and pawn. Yet
White stands worse, since Black soon pushes his e- and f-pawns, endangering White’s king.
17 Bh4
17 Bc1 Rae8 18 c4 f5! transposes to the above note.
17 ... Bd6?!
Intending ... Nc5, picking off the bishop-pair. However, the move allows Tal some
unexpectedly dirty tricks in a few moves. Black equalizes completely with 17 ... f5!.
18 c4! Ba6

Exercise (combination alert): It appears as if Black seized the initiative.

After all, c4 is under attack, while every method of protecting c4 either
loses material or loses the initiative for White. What did Tal play?
 Answer: Annihilation of defensive barrier. Tal completely ignores the c4 threat and
sacrifices a piece to expose Black’s king. Computer analysis shows that this sacrifice is
completely sound.
19 Nxg7!!
The nanosecond after I make any kind of complex sacrifice, I am filled with misgivings.
Not so with Tal, who lived for such moves, which startle our eyes, as well as our
imaginations. Every imaginative idea is always bound by a position’s resources to actually
pull it off over the board. In this case, the comps back Tal’s sacrifice up 100%. Tal reverts
to his not-so-hidden bestial tendencies, striking before the enemy is allowed time to
reorganize countermeasures. We all knew this move was coming. It was a matter of ‘when’,
rather than ‘if’. Instead:
a) 19 b3?? walks into a fork after 19 ... Nc3.
b) 19 Rc1?! loses the initiative to 19 ... Nc5! 20 Rfd1 Rad8 and Black stands better.
c) 19 Bd3 is playable, but not nearly as much fun as Tal’s g7 sacrifice. Black stands fine
after 19 ... Nf4 20 Nxf4 Bxf4 21 b4 (the greedy 21 Qxe6? walks into 21 ... Nc5, after which
Black chops on d3 and then on c4, with possession of the bishop-pair) 21 ... Ne5 22 Nxe5
Qxe5 23 Bg3 Qxe2 24 Bxe2 Bxg3 25 hxg3 Rad8, with even chances in the ending.

19 ... Kxg7!
Vasiukov finds the only move.

Question: Why doesn’t Black toss in the zwischenzug 19 ... Bxc4?

Answer: It works out in White’s favour after 20 Nxe6! Bxe2 21 Nxc7 (suddenly, Black’s
entire army seems to be hanging) 21 ... Bxd1 22 Nxa8 Be2 23 Bxd5 Bxf1 24 Kxf1. White got
two pieces and a pawn for a rook, and is winning.
 Question: What about the fact that White’s a8-knight is trapped?

Answer: The a8-knight may not have a safe square, but then I ask the question: how does
Black win the knight? The answer is he can’t. If White has all day, then eventually he will
free the knight.
20 Nd4!
Dual purpose: White threatens Nxe6+ and Nxc7, and also Qg4+, followed by cxd5.
20 ... Nc5?!
Black is better off entering the wild complications of 20 ... Rae8, which is met with 21
Qg4+ Kh8 22 Nxe6 Rxe6 23 Qxe6 Bxh2+ 24 Kh1 Nc5 25 Qxd5 Bb7 26 Qf5 Bxe4 27 Bxf6+
Kg8 28 Qg5+ Bg6 29 Qd5+ Bf7 30 Qd4. White’s rook and two pawns are superior to
Black’s two minor pieces, mainly since his king remains insecure.
21 Qg4+?!
Tal goes astray himself. He gains a clear advantage with 21 Nxe6+! (undermining) 21 ...
Nxe6 22 Bxd5 Ng5 23 Bxg5 fxg5 24 Bxa8 Rxa8 25 Qg4 and Black’s trouble is 25 ... h6 is
met with 26 h4 Bc8 27 Qe4 Bb7 28 Qe6 Qc6 29 Qg4. Black’s king is in far more danger
than White’s.
21 ... Kh8 22 Nxe6 Nxe6 23 Qxe6 Rae8!?
23 ... Bxh2+ leads to a draw after 24 Kh1 Rae8! 25 Qxd5 f5 26 Bxf5 Bxc4 27 Qd7
Qxd7 28 Bxd7 Re4 29 g3 Bxf1 30 Rxf1 Rxh4 31 gxh4 Bg3 32 Kg2 Bxh4, with a drawn
ending.
24 Qxd5 Bxh2+ 25 Kh1

25 ... Qf4?!
Alternatives are:
a) 25 ... Qxc4?! 26 Qxc4 Bxc4 27 Rfe1 Be5 28 b3 Be6 29 Bc6 Rc8 30 Bd7 Bxd7 31
Rxd7 a6 (not 31 ... Rc7? 32 Rxc7 Bxc7 33 Re7 Be5 34 Rxa7 and Black is losing) 32 Bg3!
Bxg3 33 fxg3 Rfe8 34 Rxe8+ Rxe8 35 Rd6 Re2 36 a4 b5 37 Rxa6 Rb2 38 axb5 Rxb3 39
Rb6 Kg7 when White gets some practical chances, but the ending should be drawn with
correct play.
b) 25 ... f5! and Black has no problems after 26 Bxf5 Bxc4 27 Qd7 Qxd7 28 Bxd7 Re4,
transposing to the above note.
26 Qh5! Qxe4 27 Rfe1! Qg6 28 Qxg6
Okay fine. No more attack for Tal. When an attack dies, we feel like a widow, whose
financial provider passed away without life insurance, and left her with many children to
feed. We can only marvel at the skilled thrust and parry, as attack and defence flow in
perfect balance – until Black’s next move!
28 ... hxg6?
Black continues with his plan, yet we sense inapplicability in its timing. The correct
path to the draw lies in 28 ... Rxe1+! 29 Rxe1 hxg6 30 Kxh2 Bxc4 31 b3 Bg8 32 Re7 g5 33
Bg3 Rf7 and Black holds the draw.

Exercise (combination alert): When we flub the end of a sequence, it’s as if we write a
well-polished speech, and then go blank with silence, just when we deliver it. We can plan
with meticulous detail, and inevitably, we miss a simple element which we failed to factor
into our calculations. Very few of us are able to out run our clocks in a chess game. Black’s
last is a tactical error, which allowed Tal a combination. How?

Answer: Overloaded defender.

29 Bxf6+!
Now Black’s previous good fortune begins a slow and steady reversal.
29 ... Kg8 30 Rxe8 Rxe8 31 Kxh2 Bxc4 32 Rd7!
 Question: White is only up one pawn in a bishop
of opposite colours ending. Why isn’t this drawn?

Answer: The future is that mirage-like place, where all our misfortunes of past and
present can be magically fixed – but not in this instance. Three factors against Black holding
the draw:
1. White’s rook dominates the seventh rank.
2. This, in conjunction with the opposite-coloured bishops, allows White’s king to walk
into Black’s position, via the dark squares.
3. Which in turn means that Black’s king will not be safe, despite the reduced material.
32 ... Re6 33 Bc3
If 33 Rg7+ Kf8 and White should avoid 34 Rxg6?? Kf7 when Black wins a piece.
33 ... Bxa2 34 Rxa7 Bc4 35 Kg3
Time for White’s king to join the fight.
35 ... Bd5 36 f3 Kf8 37 Bd4 b5 38 Kf4 Bc4 39 Kg5!
 The white king’s rising status is obviously of deep concern for Black’s now endangered
king.
39 ... Ke8 40 Ra8+ Kf7
Likewise, 40 ... Kd7 41 g4 Bd3 42 b4 Kc6 43 f4 Re2 44 f5 gxf5 45 gxf5 and Black’s
king isn’t nearby to halt White’s passed f-pawn.
41 Ra7+
The sealed move. Vasiukov was unable to hold the game, despite laborious adjournment
analysis.
41 ... Ke8 42 b4 Bd5
Or 42 ... Bf1 43 g4 Be2 44 f4 with f5 to follow at the proper moment.
43 Ra3!
Enabling g4.
43 ... Kf7 44 g4 Re2 45 Bc5 Re5+ 46 Kh6! Re6
46 ... g5?? is met with 47 Bd4 Re8 48 Kxg5.
47 Rd3 Bc6 48 Rd8 Re8 49 Rd4 Re6
Strong (or weak) GMs don’t fall for 49 ... Bxf3?? 50 Rf4+, winning Black’s bishop.
50 f4
At last, Tal mobilizes his kingside pawn majority.
50 ... Ke8
Following 50 ... Re2 51 Rd6 Be4 52 Rd8 Rh2+ 53 Kg5 Rg2 54 Rd7+ Kg8 (54 ... Ke6??
55 Re7+ Kd5 56 Re5+ wins) 55 Kf6! Rxg4 56 Rg7+ Kh8 57 Bd6 Rg2 58 Re7 Bf5 59 Rb7,
and if 59 ... Bd3 60 Be5 Bc4 61 Ke7+ Kh7 62 Rb8 g5 (62 ... Bg8 63 Rxb5 is also
completely lost for Black, who is too tangled to halt White’s passer) 63 f5 Bg8 64 Rxb5
Black is again unable to save the game.
51 Kg7!
Tal cuts off f8 and f7 from Black’s king.
 51 ... Be4
If 51 ... Bd7 52 g5! Rc6 53 Rd5 Re6 (or 53 ... Ra6 54 Rxd7! Kxd7 55 Kf7 Rc6 56 f5
gxf5 57 g6 and White wins) 54 f5! gxf5 55 g6 Bc6 56 Rd4 Be4 57 Kh7 Bb1 58 Rf4 Re1 59
g7 Rh1+ 60 Kg6 Ba2 61 Rxf5 when there is no defence to the coming Rf8+ and g8Q.

Exercise (combination alert): In this ending, as in our world, the

dead far outnumber the living. Resistance is diminished, but not
quelled completely. Despite the simplification, Tal found an elegant
finish, involving a potential mating attack. What would you play here?

Answer: Overloaded defender/mating net.

52 Bb6! Bf3
Not 52 ... Rxb6 53 Rxe4+ Kd7 54 f5 gxf5 55 gxf5 with an easy win, since Black’s king
is unable to participate in halting the f-pawn.
53 Rd8+ Ke7 54 Rd3! Be2
After 54 ... Bxg4 55 Bd8+ Ke8 56 Bg5! the threat of Rd8 mate forces Black to play 56 ...
Re7+ 57 Bxe7 Kxe7 58 Kxg6 with an easily won ending for White.
55 Bd8+!
The picture of Tal’s mating net begins to become clear: White transfers his bishop to g5,
after which Rd8 mate becomes a unsolvable issue for Black.
55 ... Ke8
“Good help is so hard to find,” thinks the king, as he contemplates the colossal
incompetence of his defenders.
56 Rd2 Re3
Or 56 ... Bxg4 57 Bg5 and once again, Black must hand over the exchange to avoid the
Rd8 mating net.
 57 Bg5
Black’s king crossed the gates of the prison and is unlikely to ever leave. The white
king/rook/bishop team visits endless torments upon him. Even in simplified endings,
thoughts of delivering checkmate dominated Tal’s daydreams. There it is. White threatens
mate on the move.
57 ... Bd3 58 f5 1-0
Black’s defenders are relegated to dusty corners, all three unable to halt White’s passed
pawn. If 58 ... Rf3 59 fxg6 Rh3 60 Rxd3! Rxd3 61 Kh7 Rh3+ 62 Bh6 and there is no remedy
to the coming g7 and g8Q.

Game 31
F.Thorbergsson-M.Tal
Reykjavik 1964
King’s Indian Defence

1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4 e4 0-0 5 f4

The King’s Indian Four Pawns Attack. Its prime directive: to gain everything, we must in
turn risk everything (which just happens to be Tal’s prime directive as well!). It’s an
opening designed for those players with a low boredom threshold. He or she is the eternal
teenager, looking for thrills, who never grows up, and never wants to. Still, I logically ask:
if you are the 50:1 underdog against Tal, is it really wise to pick the sharpest possible line,
which ensures tactics and crazy complications – Tal’s specialty?

Question: Why can’t White gain a tempo with 5 e5?

Answer: He can, at serious risk of overextension. It feels to me like this is a lousy

version of the Alekhine’s Defence Four Pawns Attack for White, after 5 ... Ne8 6 f4 d6 7
Be3 c5! (a completely sound pawn sacrifice, which exploits White’s development lag, by
forcing the centre open) 8 dxc5 Nc6!. Fischer went on to exploit his development lead with
a brilliantly famous queen sacrifice, R.Letelier Martner-R.Fischer, Leipzig Olympiad 1960.
I didn’t put this game in Fischer: Move by Move, since Everyman continues its
unreasonable and arbitrary policy of limiting my books to under 1,000 pages.
5 ... d6 6 Nf3 c5 7 d5 e6 8 Be2 exd5 9 exd5!?
Why bother to play an ultra-sharp line, and then chicken out with a safe move? Sharper
and probably stronger is 9 cxd5 Bg4 (this move follows the Benoni rule of swapping away
the light-squared bishop – Black’s worst piece) 10 0-0 Nbd7 11 h3 Bxf3 12 Bxf3 Re8 13
g4!? h6 14 h4 when all hell soon breaks loose, Ding Liren-R.Rapport, Biel 2013.
9 ... b5!?
 When we follow theory, we are supported by past knowledge. When we bang out a new
idea, we leap joyfully into the dark abyss of the unknown. By making a sacrifice like this
one, we basically declare: “I am willing to base a very important decision on a hunch.” This
isn’t quite a novelty, since a game was played with this move the previous year. I don’t
know if Tal was familiar with the game, or he came up with this idea independently. As
usual, he shakes up the divinely established order of things with this sacrifice. I will pre-
empt your coming question by telling you this is a rare instance of a sound sacrifice from
Tal. Normally, ... b5 sacrificial ideas are only played when White is forced to take b5 with a
pawn, rather than the c3-knight.

Question: What does Tal get for the pawn?

Answer: The following: 10 Nxb5 will be followed by 10 ... Ne4!. This creates several
issues for White:
1. The c3-square is covered, so his b5-knight must then retreat to the awkward a3-
square, where it costs heavy time to get the knight back into the game.
2. Black’s dark-squared bishop has been unleashed, which in conjunction with the
coming ... Rb8 means that White may experience difficulty in developing his queenside,
since b2 needs protection.
3. Black’s unchallenged e4-knight is a formidable piece.
4. As a result of the sacrifice Black gains open lines, active pieces and a development
lead. Conclusion: We don’t have to make every pawn offer based on exact measurement.
Tal’s pawn sacrifice looks sound to me.
10 Nxb5
10 cxb5 gives Black a nice looking Benko Gambit after 10 ... a6 11 0-0 axb5 12 Bxb5
Na6, G.Bagaturov-A.Shirov, Frunze 1989.
 10 ... Ne4

The position is familiar and strange in equal parts. It looks 50% Benoni and 50% Benko
Gambit.
11 0-0 a6 12 Na3
White decides to press the matter and hang on to his extra material. It may have been
wiser to return the pawn and complete development with 12 Nc3! Nxc3 13 bxc3 Bxc3 14
Rb1 when White certainly stood no worse, and might even be a shade better, due to his extra
space, J.Pinter-J.Banas, Stary Smokovec 1974.
12 ... Ra7
This move was new. The rook may swing over to b7 or, more likely, e7. 12 ... Re8 was
played in the only game which preceded Tal’s game: 13 Bd3 Ra7 14 Qc2 Rae7 15 Bd2
(rather than hand over the important dark-squared bishop, perhaps White should consider
the cumbersome 15 Nb1!?, intending to challenge the e4-knight next move, with either Nc3,
or Nbd2) 15 ... Nxd2 16 Qxd2 Bg4 17 Nc2?! (White is just worse if he returns the pawn; he
should try 17 h3 Bxf3 18 Rxf3 Nd7 19 Raf1) 17 ... Bxb2 18 Rab1 Bg7 and Black’s
domination of the dark squares gave him the edge and he went on to win, A.Chaikovskaia-
V.Rozlapa, Moscow 1963.
13 Bd3 Re7 14 Nc2
White brings his knight somewhat back into play. It covers d4 and e3, yet is unable to
challenge Black’s e4-knight.
14 ... Rfe8 15 Re1
White is intent on retaining his extra pawn. Once again, I would think about returning the
pawn in order to disperse Black’s pressure with 15 Nd2 Nxd2 16 Bxd2! Bxb2 17 Rb1 Rb7
18 Re1, when White stands no worse.
15 ... Nd7 16 Ne3
 Question: Why didn’t White challenge the e4-knight with 16 Nd2?

Answer: It loses on the spot. Let’s pause for an exercise:

Exercise (combination alert): White’s last move was a blunder. Black to play and win:

Answer: Clearance: 16 ... Nf2!! 17 Kxf2 Rxe1 18 Nxe1 Qh4+ 19 Kf1 Bd4. We have
seen this theme before in Tal’s Benonis. Mate is threatened on f2 and White loses material,
no matter what he plays.
16 ... Ndf6 17 Qc2 Nh5!
Inducing a weakening of his opponent’s light squares, which become a harbinger of
White’s future decline.
18 g3
We all hate that feeling where we are certain that our opponent is up to something, but
what that ‘something’ is may be beyond our ability to prognosticate.
18 ... Bd4!!
 And here is Tal’s ‘something’. His grand adventure is, of course, accompanied by risk of
loss, and the cost of Black’s war is underwritten by the sacrifice of the defender of his dark
squares. When rational thought is disengaged, the result will either be confused chaos, or
flawless intuition, steering us infallibly in the right direction. I believe Tal’s move is a case
of the latter.

Question: Isn’t Black’s dark-squared bishop the holy grail of the Benoni?

Answer: It is, and Tal’s nonconformist last move shows just what lengths he was willing
to take to seize the initiative. It’s an attempt to distil an abstract notion into solid form. The
move is so exceptionally deep, that the comps initially prefer White’s position, and only
later apologize and prefer Black.
19 Nxd4 cxd4 20 Ng2
It may have been better to try 20 Nf1 Nhf6 21 b4 Qb6 22 Bb2 Bf5 23 a3 h5, and if 24
Rac1?! Nc3! 25 Rxe7 Rxe7 26 Bxf5 Re2! 27 Nd2 gxf5 28 Qd3 Nfe4 with huge
complications and approximately even chances.
20 ... Ng5!
Tal finds a clever zwischenzug shot which seizes control over White’s kingside light
squares.
21 Rxe7 Nh3+
Zwischenzug.
22 Kf1
22 Kh1 is scary since White’s back rank is eternally weak, as long as the h3-knight
remains in place.
22 ... Rxe7
Tal didn’t like 22 ... Qxe7 23 Bd2 when Re1 follows.
 23 Bd2?!
23 f5! prevents Black’s coming manoeuvre.
23 ... Nf6!
Tal immediately seizes upon the slip, transferring his knight to g4.
24 Nh4?
24 f5 Ng4 25 Nf4 Nhf2! 26 h3 Nxd3 27 hxg4 Ne5 28 c5 Nxg4 29 cxd6 Ne3+ 30 Bxe3
Rxe3 31 Qc7 Qe8 leaves White somewhat overextended. Yet this line is still better than the
game’s continuation for White.
24 ... Ng4
“Two riders were approaching and the wind began to howl.” Black’s knights menace the
h2-, f2- and e3-squares. There are insects that mimic the attributes of twigs and leaves to
camouflage their location. Black’s knights qualify for this status, as they seem to ‘belong’ in
White’s territory, when in reality, they don’t.
25 Nf3
25 b4 may be slightly more accurate, yet still favours Black. The comp found the line 25
... Qe8 26 Nf3 Re3! 27 Nxd4 Qe7 28 Qc3 Bf5! 29 Nxf5 gxf5 30 Be1 Rxg3!! 31 hxg3 Ne3+
32 Ke2 Nxd5+ which wins White’s queen.
25 ... Re3!
Another uninvited attacker slips into White’s territory.
26 Kg2
26 Nxd4? is met with the crushing shot 26 ... Nxf4! 27 gxf4 Qh4 which leaves White
helpless.
26 ... Qe7 27 Re1
27 Kxh3?? is unplayable due to 27 ... Rxd3!, winning instantly.

Exercise (combination alert): It appears as if White is rallying.

 This would be true, if not for Black’s coming combination ...

Answer: Annihilation of defensive barrier.

27 ... Nxf4+! 28 gxf4 Rxe1!
This forces a retreat of a key defender.
29 Nxe1
Obviously the rook is taboo, due to 29 Bxe1?? Ne3+, forking king and queen.
29 ... Qh4
The Czarina is displeased. Now the queen enters, threatening h2, f2 and even e1, which
becomes apparent next move.
30 Bc1
Unfortunately for White, this is the only move, since f2 must be covered, as well as the
e3 fork square.
30 ... Qxe1
A predator looks for one in the herd with a weakened gait. So Tal regained his
sacrificed piece, while retaining his winning attack.
31 h3 Nh6
Seizing control over f5.
32 f5
It’s uncharitable to give a player a question mark when he or she is busted anyway. This
is complete desperation, which actually speeds White’s defeat. 32 Qd2 Qh4 33 f5 Nxf5 34
Bxf5 Bxf5 is also hopeless for White, since h3 falls, and his king remains exposed.
32 ... Nxf5 33 Bf4 Nh4+! 34 Kh2 Nf3+ 35 Kg2

Exercise (combination alert): White’s king acutely feels the presence of malevolent,
watchful eyes all around. Black has two clear wins in the position. Find one of them:
 Answer: Attraction.
35 ... Bxh3+!
The bishop strives to exact obedience from White’s king. Tal just had a way of
mesmerizing his prey with a hypnotic stream of sacrifices.
Answer no.2: Also winning is 35 ... Qg1+! 36 Kxf3 Bxh3 37 Bh6 f5! (threat: ... Bg4+
and ... Qe3 mate, which leaves White’s king curiously helpless) 38 Bc1 Bg4+ 39 Kf4 Bd1!,
and if White attempts to save his queen with 40 Qd2, Black plays 40 ... Qg4 mate.
36 Kxf3
36 Kxh3 Qh4+ 37 Kg2 Ne1+ wins the house.
36 ... Qg1!
“Even as a child, you were always slow to understand. So I will patiently explain your
new circumstances,” lectures the queen, to her brother, who has heard all this before. This
deadly finishing touch threatens ... Bg4+ and ... f5 mate.
37 Bxg6
If 37 Ke2 Bg4+ 38 Kd2 Qf2+ 39 Kc1 Qe1+ 40 Qd1 Qxd1 mate.

Exercise (combination alert): I assure you that coming combination

won’t heavily tax you mentally. Black is winning if he simply
recaptures on g6. However, he has a much stronger move. What is it?

Answer: Zwischenzug. Black picks up both white bishops.

37 ... Qg4+! 38 Kf2 Qxf4+ 39 Kg1 hxg6 0-1

Game 32
V.Smyslov-M.Tal
USSR Team Championship, Moscow 1964
English Opening

1 c4 g6 2 Nc3 Bg7 3 g3 c5
The Symmetrical English.
4 Bg2 Nc6 5 b3
Smyslov experimented with this idea, with rotten results. He scored only one and a half
points from five tries with it, losing to Tal and Fischer, and drawing with Gligoric, Averkin
and Ivkov.
5 ... e6
Intending to build a big centre with ... Nge7 and ... d5. Alternatively, 5 ... Nf6 6 Bb2 0-0
7 Nf3 d5! (also okay for Black is the more modest 7 ... d6 8 0-0 Bg4 9 h3 Bd7 10 e3 Ne4 11
d4 Nxc3 12 Bxc3; the exchange on c3 freed Black’s position and he looks fine, V.Smyslov-
S.Gligoric, Hastings 1969/70) 8 Nxd5 Nxd5 9 Bxg7 Kxg7 10 cxd5 Qxd5 11 0-0 Bd7 12
Rc1 Rad8 13 Qc2 b6 14 Qb2+ e5 and I already slightly prefer Black, due to his extra
central space, V.Smyslov-O.Averkin, Moscow 1969.
6 Bb2 Nge7 7 Na4?!

Well, I never claimed that Smyslov’s opening play represented the Age of Reason. This
is one of Smyslov’s more unfortunate ideas, where he tried it on Tal and lost, and then
unwisely repeated it on Fischer and got strategically crushed. Smyslov calls into question
the old order of things and seeks radical change. However, in some cases – as in this one –
the old order of things is correct.

Question: Why give this move a dubious mark? Smyslov

seeks to eliminate Black’s best piece, his g7-bishop.
 Answer: Yes, the g7-bishop is Black’s best piece, but this is easily outweighed by
White’s time-wasting Nc3, Na4 and Nxb2 expedition. Reti frowned upon fianchettoed
knights! Later Smyslov tried 7 Qc1!? (the idea is to dominate the central dark squares with
Ne4, but the idea is much too clunky to give White an edge) 7 ... d6 8 h4 (8 Ne4 is met with
8 ... e5) 8 ... h6 9 e3 0-0 10 Nge2 e5 11 0-0 Bf5 12 Ne4 Rc8 13 d3 b6 14 N2c3 Be6 when ...
d5 or ... f5 may follow and Black already holds the edge, V.Smyslov-B.Ivkov, Palma de
Mallorca 1967.
7 ... Bxb2 8 Nxb2 0-0 9 e3
This is awfully slow. There is something to be said about rapid development with 9 Nf3.
Even there Black stands at least even after 9 ... d5, W.Lombardy-A.Adorjan, Olot 1974.
9 ... d5 10 Nf3
Smyslov’s move looks better than 10 cxd5 which helps Black develop: 10 ... Nxd5 11
Ne2 b6 12 d4?! Ba6! 13 dxc5 Qf6! 14 Nc4 Nc3! 15 Nxc3 Qxc3+ 16 Kf1 Rfd8 17 Qc1
Bxc4+ 18 bxc4 Qd3+ 19 Kg1 Rac8 20 cxb6 axb6 and White’s counterplayless position
gasps like a goldfish in a long-uncleaned aquarium. Black’s monster development lead
easily outweighs White’s extra pawn, and Fischer goes on to clobber Smyslov, V.Smyslov-
R.Fischer, Palma de Mallorca 1970. This game is annotated in Fischer: Move by Move.
10 ... Nf5
In order to suppress White’s future d4 break. I also slightly prefer Black’s position after
the simple 10 ... e5.
11 0-0 b6

Question: Why doesn’t Black seize more space with 11 ... d4?

Answer: Your suggestion is met with 12 Nd3! which is awkward for Black to meet. If
12 ... Qe7 (not 12 ... dxe3? 13 dxe3 Qe7 14 Nd2! Bd7 15 Ne4 b6 and now White has the
combination 16 Ndxc5!, intending 16 ... bxc5? 17 Qxd7! winning) 13 e4 Ng7 14 e5! When I
suddenly prefer White, who may be able to make use of his e5 point.
12 Na4!?
Smyslov acts like he has all day. In the game he spent four moves to get his knight to the
out-of-play a4-square.
12 ... Bb7
A positional player would probably go for 12 ... dxc4 13 bxc4 Bb7 with a tiny edge for
Black.
13 cxd5
13 Nc3?! is met with 13 ... dxc4 14 bxc4 Qd3 15 Qe2 Qxe2 16 Nxe2 Rad8 17 Rfd1
Rd7, with an unpleasant ending for White, due to the pressure down the d-file.
13 ... exd5!
Tal was always tempted into unbalanced pawn structures. In this instance, I also think
it’s Black’s best move, even if 13 ... Qxd5 is at least even for Black.
14 d3
A move of sly caution, where White refuses to divulge his intent of a future e4, or a d4
pawn break.
14 ... Qf6
Tal holds back on opposite wing majorities with 14 ... d4 15 e4 Nd6 which also looks
promising for him, since White must watch out for ... Nb5, aiming for the c3-square.
15 Qd2 Rad8 16 Rfd1 Rfe8 17 Rab1
Thinking about a3 and b4 later on.
17 ... Nd6
Strong is the immediate 17 ... d4! 18 e4 Nd6, intending ... Nb5 and ... Nc3.

18 Ne1!?
 Question: It feels to me like Smyslov degenerates into full passivity mode
in this game. Isn’t it high time to challenge Black’s space with 18 d4?

Answer: Black gets the advantage in that line as well. For example, 18 ... Ne4 19 Qb2
Bc8 (threat: ... Bg4) 20 Re1 cxd4 (20 ... Bg4 is met with 21 dxc5! which saves the piece) 21
Nxd4 Ne5 (threat: ... Nd3) 22 Rbd1 Bg4 23 f4 Bxd1 24 fxe5 Qg5 25 Bxe4 dxe4 26 Rxd1
Qxe3+ 27 Qf2 Qxf2+ 28 Kxf2 Rxe5. The ending is clearly in Black’s favour, with a rook
and two pawns for White’s ineffective two knights.
18 ... d4!

Well timed. The air of manifest dullness and routine is shattered in the space of a single
move. Black’s superior majority and extra space give him a clear advantage.
19 e4 Qe7
Even stronger is to discourage White’s f4 with 19 ... g5!, and if 20 f4 gxf4 21 gxf4 (21
Qxf4 Qe5 22 Nf3 Qxf4 23 gxf4 Nb4 24 Rb2 Nb5 looks like a tough ending for White) 21 ...
Qh6 looks highly unpleasant for White.
20 Nc2!?
Smyslov refuses to activate his pieces and continues to defend. I would try 20 f4 to
move forward the kingside pawn majority.
20 ... f5!
Tal shaves away the undesired, as easily as my gardener with my rose bushes. If his
opponent won’t act, then Tal will. Also promising is 20 ... g5!.
21 exf5 Ne5!
A cloud of disquiet gradually forms over White’s king. Tal offers a pawn to seize control
over the light squares.
22 f4
Not much choice, since 22 Ne1 Bxg2 23 Kxg2 Nxf5 just looks awful for White.
 22 ... Nf3+ 23 Bxf3 Bxf3 24 Re1 Qe2!?

“You may have right on your side, but I have power on mine,” says the queen, who
postures in impersonation of the statue she just commissioned of herself in a heroic pose.
When we have a choice of flight or execution, most of us will pick the former. Not Tal
though! Of course the correct annotation for this move is ‘??!!’. Kasparov gave Tal’s move
two exclams, which may be more a reward for Tal’s imagination, rather than an indicator of
the move’s effectiveness.
Tal’s construct lacks objective support and the sacrifice is yet another bluff, but one
which ensures a draw and forces White to find only moves to secure it – virtually
impossible in Smyslov’s time pressure. Generally when we give up our queen, it is in
feverish expectation of an immediate reward, which isn’t to be in this case. Everyone else in
the world would have played the mundane, but objectively stronger line 24 ... Qd7! 25 fxg6
Bb7! (threat: ... Qc6) 26 h3! (this is no time to get greedy and open lines for Black by
chopping on h7) 26 ... Qxh3 27 Qh2 Qxh2+ 28 Kxh2 hxg6 29 g4 Kf7 with a clear advantage
for Black in the ending.
25 Rxe2
White’s rook feigns surprise, when he expected just such treachery from Black’s queen
all along.
25 ... Rxe2 26 Qxe2?
An entire nation can perish if its single, powerful leader dies. A move made with the
philosophy: we can believe in some admirable principle (like grab material and then return
nothing!), yet it may not be applicable if we don’t have the time on our clocks to work out
the precise details. When defending, it’s difficult to allocate the correct quota of resources
to the correct component, at the correct time. This move fits none of the three! It’s the
practical – yet incorrect! – decision. Smyslov, in a desperate bid to rearrange his fate,
mistakenly returns the queen and hopes he can save the inferior ending.
 Question: Why did Smyslov return the queen? Can’t he survive 26 Qc1?

Answer: Your (greedy!) suggestion revives White’s fortunes, if he can find a series of
only-moves. With Smyslov low on time, I’m not so certain he would have found the drawing
line: 26 ... Rg2+ 27 Kf1 Rxh2 28 Ne1 Bd5, and now he must find 29 g4!, avoiding the lines:
a) 29 Rb2?? Rh1+ 30 Kf2 Nxf5 when, believe it or not, White is helpless. If 31 Rc2
Ne3 32 Rd2 Ng4+ 33 Ke2 Re8+ and White gets mated.
b) 29 Kg1?! leaves Black clearly better after 29 ... Rh1+ 30 Kf2 Nxf5 31 Nxc5 bxc5 32
Qxc5 Ne3 33 Qe7 Ng4+ 34 Ke2 Rf8 35 Rc1 Bf7 (intending ... Re8) 36 Kf3 h5 37 Rc7 Rh2
38 Qd6 Rf2+ 39 Ke4 Re2+ 40 Kxd4 Rxe1 Black has good winning chances with a rook and
two pieces for White’s queen and two pawns.
After 29 g4 gxf5 30 g5 Re8 31 Rb2 Rh1+ 32 Kf2 Re3 33 Nxc5 bxc5 34 Qxc5 Black
must take a draw: 34 ... Rh2+ 35 Kf1 Rh1+ 36 Kf2 with perpetual check.
26 ... Bxe2 27 Nb2 gxf5!
This move, in conjunction with a future ... Bh5, freezes White’s kingside pawn majority
and in essence puts Black up a pawn.
28 Re1 Bh5 29 Nc4
29 Re7 is met with 29 ... Re8! gaining a tempo over the game continuation.
29 ... Nxc4 30 bxc4 Re8!
Tal foresees that best chance is a battle between White’s clunky knight and his superior
bishop.
31 Kf2 Rxe1 32 Kxe1
 White’s position, like that of a terminal patient, feels itself growing weaker with the
passage of time, due to the following issues:
1. White’s d3- and c4-pawns are fixed on light squares, and therefore perpetual targets
of Black’s bishop.
2. White’s knight lacks scope or targets and is clearly inferior to Black’s remaining
bishop.
3. Black’s queenside pawn majority is healthy and may be activated by transferring the
king to c6 and then playing ... a6 and ... b5. If White plays a4 to discourage this plan, then
Black fixes yet another pawn target for his bishop, with ... a5!.
32 ... Kf8 33 Kd2 Ke7 34 Ne1 a6 35 a4!?

White’s structure is flecked with a patina of corrosion, which only degenerates with
time. Of course, this allows Tal to fix yet another target, dooming White’s king or his knight
to eternal servitude to defence of a4.

Question: Can White save the game if he avoids a4?

Answer: Some items can be mended if you break them, and some can’t. Avoiding a4
also loses: for example, 35 a3 b5 36 cxb5 axb5 37 h3 Kd6 38 Ng2 h6 39 Nh4 Ke6 40 Ng2
Bf3 41 Nh4 Bd5 42 g4? fxg4 43 hxg4 Kf6 (White’s knight is frozen, since Nf5 will be met
with ... h5!) 44 Ke2 Be6 45 Kf3 Bd7 46 f5 Bc6+ 47 Ke2 (47 Kf4?? b4 promotes to a new
queen) 47 ... Bd5 48 Ng6 b4 49 axb4 cxb4 50 Kd2 Bf3 51 Kc2 Bxg4 52 Kb3 Bxf5 53 Nf4
Kg5 54 Ng2 Bxd3 and Black wins.
35 ... a5!
Threat: ... Be8 and ... Bxa4. This forces White’s king to baby-sit a4.
36 Kc2 Be8 37 Kb3 Bc6 38 Ka3
38 Nc2 allows the bishop to infiltrate with 38 ... Bf3 39 Ne1 Bd1+ 40 Ka3 Ke6 41 Ng2
Kf6 42 Ne1 Kg6 43 h3 Kh5 44 Ng2 and now Black wins with 44 ... Bg4! 45 hxg4+ (45 h4
Bh3 46 Ne1 Kg4 is a trivial win for Black) 45 ... Kxg4 46 Kb2 Kxg3 47 Ne1 h5 when
White is helpless against the coming push of the h-pawn.
38 ... Kf6 39 Kb3
“In diesem Leichtfigurenendspiel ist der langschrittige schwarze Läufer nützlicher als
der weiße Springer,” says my ChessBase annotations, which is all well and good, except I
don’t speak German. But you don’t need to speak German to understand that it roughly
translates to: “Black’s bishop is clearly superior to White’s knight”.
39 ... Kg6
The sealed move. There were only a few hours until resumption, so Tal didn’t have the
time to fully work out the win in the adjourned session. Black’s biggest problem is how to
infiltrate White’s fortress with his king.
40 Ka3 Kh5 41 h3
Of course, Black’s king can’t be allowed in. The move does create a third pawn
weakness on h3. If Black’s bishop attacks it and provokes h4, this allows Black’s king entry
into White’s position, via g4.
41 ... Kg6 42 Kb3 Kg7 43 Ka3 Kf6 44 Kb3
 Exercise (planning): How does Black make progress?

Answer:
Step 1: Transfer the bishop to d1, which zugzwangs White’s king.
44 ... Be8! 45 Ng2
Likewise, after 45 Nf3 Bh5 46 Ne5 Bd1+ (this post on d1 effectively imprisons White’s
king) 47 Ka3 Ke6 (to prevent Nd7+ and Nxb6) 48 Nc6 Bc2 (tying the knight down to
defence of d3, which leaves White perilously close to zugzwang) 49 Ne5 Ke7 50 Nc6+ Kd6
51 Ne5 Ke6 52 g4 h6 (zugzwang!), White can resign.
45 ... Bh5 46 Kc2!
Cutting off ... Bd1+, which would leave White’s king on a3, without a single move.
Step 2: Transfer the bishop to f1, which either wins the h3-pawn, or provokes h4,
allowing Black’s king infiltration.
46 ... Be2!
Threat: ... Bf1 and ... Bxh3.
47 Ne1 Bf1!
Mission accomplished. Tal forces a gap in Smyslov’s fortress.
48 Nf3!
This is White’s best path to resistance. If 48 h4? Be2! (and not 48 ... Kg6?? 49 Nf3 when
suddenly White holds the game, since Black’s king is denied entry to g4 when White posts
his knight on e5) 49 Kd2 Bh5 50 Kc2 Be8 51 Kb3 Bc6 52 Ka3 Kg6 and there is no defence
to Black’s king weaving into the kingside light squares and picking off all of White’s
kingside pawns.
48 ... Bxh3 49 Ng5
This retains material equality for now, but fails to save the game.
 49 ... Bg2 50 Nxh7+ Kg7 51 Ng5 Kg6 52 Kd2
This move gives up a4, which is the equivalent of resignation. Other defences fail too:
for example, 52 Ne6 Kh5 53 Nc7 Bc6 54 Kb3 Kg4 55 Nd5 Kxg3 56 Nxb6 Kxf4 and the
passed f-pawn will win the game for Black.
52 ... Bc6 53 Kc1

53 ... Bg2!?
Kasparov gives this move an exclam, without explanation. The move wins, but I don’t
see why Tal didn’t play the simple 53 ... Bxa4, which converts rather easily after 54 Nf3
Kf6 55 Ne5 Be8 56 Kb2 b5 57 cxb5 Bxb5 58 Kb3 Ke6 59 Kc2 (59 Nc4?? Bxc4+ 60 Kxc4
a4 is zugzwang; Black promotes) 59 ... Kd5 60 Kb2 c4 61 dxc4+ Bxc4 62 g4 fxg4 63 Nxg4
Ke4 and Black wins.
54 Kd2 Kh5 55 Ne6
Also hopeless are the lines:
a) 55 Ke2 Kg4 56 Kf2 Bc6 57 Nf7 (intending Nh6+ and Nxf5, which Black ignores) 57
... Bxa4! 58 Nh6+ Kh5 59 Nxf5 Bd7 60 Nd6 a4 61 Ne4 a3 62 Nd2 Ba4! and White is unable
to prevent promotion.
b) 55 Nf7 Kg4 56 Nh6+ Kxg3 57 Nxf5+ Kxf4 58 Ne7 Bb7 (cutting off Nc8) 59 Ke2 Kg5
60 Kd2 Kf6 61 Ng8+ Ke6 62 Nh6 Bc6 when a4 falls and the win is easy.
55 ... Kg4 56 Nc7 Bc6!
More accurate than 56 ... Kxg3 57 Nd5 when Black is unable to chop the knight.
57 Nd5 Kxg3 58 Ne7
Alternatively, 58 Nxb6 Kxf4 59 Ke2 Kg3 and the f-pawn moves forward.
58 ... Bd7 59 Nd5 Bxa4 60 Nxb6 Be8
This move loses a tempo. When we are winning, yet make a second or third best move,
the good news is we have a large margin for error. 60 ... Bc6 is more accurate.
61 Nd5 Kf3 62 Nc7
 Smyslov gains a tempo, but it doesn’t help him.
62 ... Bc6 63 Ne6 a4 64 Nxc5 a3 65 Nb3
Instead, 65 Ne6 a2 66 Nxd4+ Kxf4 67 Nb3 Ba4 (the bishop advertises his superiority
and is deeply gratified any time he is able to contribute to his counterpart’s distress) 68 Na1
(the springer – yes, I am now fully fluent in German, just from reading the ChessBase notes!
– is cornered) 68 ... Kf3 69 c5 f4 leaves White in zugzwang.

65 ... a2! 66 Kc1

66 Nxd4+ Kxf4 67 Nc2 Ba4 68 Na1 transposes to the above note.
66 ... Kxf4 67 Kb2 Ke3 68 Na5 Be8 69 c5 f4 70 c6 Bxc6
I knew Tal, with his flair for the dramatic, would pick this line. Also easy is the
pedestrian 70 ... f3 71 c7 Bd7 and Black wins.
71 Nxc6 f3 72 Ne5 f2 0-1
73 Ng4+ Ke2 74 Nxf2 Kxf2 75 Kxa2 Ke2 is a won könig and pawn ending for Tal.

Game 33
M.Tal-L.Portisch
4th matchgame, Bled 1965
French Defence

1 e4 e6 2 d4 d5 3 Nc3 Nf6 4 Bg5 dxe4

This line, today fully mainstream, was rarely played at the time of the game, since it was
considered an attempt to grovel a draw with Black. Today, players like Morozevich turned it
into a viable winning weapon for Black, based on unbalancing ... gxf6 ideas.
5 Nxe4 Nbd7
In my Classical French book, a cherished classic ... oh, wait, I already said that about
one of my other books. Anyway, I advocate the aggressive Morozevich variation 5 ... Be7 6
Bxf6 gxf6!? where in the database we see an endless row of games played by Morozevich
from Black’s side.
6 Nxf6+ Nxf6 7 Nf3 c5

Question: Isn’t Black in violation of the principle: Avoid opening

the game or creating confrontation when lagging in development?

Answer: He does just that. His ambitious move is in anticipation of a reward: Portisch
wants a ... Nd7 Smyslov Caro-Kann-like position, but up a full tempo, since he didn’t waste
time on ... c6, as he plays ... c5 in a single jump.
8 Bc4
Also played here are the lines 8 Bd3, 8 Bb5+, 8 dxc5, and 8 c3.
Just to let you know that Tal wasn’t an all-seeing-all-knowing player. Here is an
example of one of his colossal goof-ups: 8 Qd3!? Be7 9 Bxf6!? Bxf6 10 Qb5+?! Bd7 11
Qxb7 Rb8 12 Qxa7 Rxb2 13 Bd3 cxd4 (when we defeat an opponent strategically, without
the aid of tactics, we achieve a bloodless revolution; Petrosian already owns the bishop-
pair in an open game, and also the superior structure) 14 0-0 Bc6 15 Qa3 Qb6 16 Bc4? Rb4
17 Qd3 0-0 18 a3 Ra4 19 Rfd1? Qa7 (19 ... e5 is also awful for White) 20 Ra2??
 Exercise (combination alert): White confronts bankruptcy, both structurally and
materially. Boy, you talk about having an off day! Tal’s last move ends a god-awfully
played game with a god-awful blunder. How did Petrosian finish the game?

Answer: Attraction/skewer: 20 ... Rxc4! ( ... Bd5 follows the recapture on c4) and 0-1,
M.Tal-T.Petrosian, Curacao 1962. “ ... I have fairly successfully endeavoured to forget all
the games which I played at Curacao,” quipped Tal, who was seriously ill during the event.
8 ... cxd4
Alternatively, 8 ... Be7 9 Qe2 Qa5+ 10 Bd2 Qb6 11 0-0-0 cxd4 12 g4!? 0-0 13 g5 Ne8
14 h4 Nd6 15 Bd3 Bd7 with approximately level chances, V.Ivanchuk-L.Ljubojevic,
Monaco (rapid) 1997.
9 0-0!?
 Tal, of course, isn’t worried and offers his d4-pawn to increase his development lead.
9 ... Be7

Question: Can Black get away with 9 ... Bc5?

Answer: It depends on who you ask. An attacker will say “Not a chance!” while a
materialist may tell you “Why not? A pawn is worth some discomfort!”. After 10 Qe2 Qb6
11 Rad1 to my mind, White’s development lead and attacking chances against Black’s
insecure king offer full compensation for the pawn, I.Nataf-H.Gretarsson, Bermuda 1999.
10 Qe2 h6 11 Bf4
The idea is to keep h6 open as a potential sacrificial target.
11 ... 0-0 12 Rad1
So Tal will regain his sacrificed pawn.
12 ... Bd7 13 Rxd4!?
Love of attack is a kind of malignant influence, when we consider it in a position which
may not warrant it. Rooks tend to be clumsy pieces when lifted early, in a crowded
middlegame. A more cautious-minded player would have played 13 Nxd4 to reserve the
third rank for operations with his rook.
13 ... Qb6 14 Qd2!?
 A sneaky new move.

Question: What is White’s idea?

Answer: Dangerous thoughts begin to form in Tal’s mind. He threatens Rxd7, as well as
Bxh6 sacrifices. Portisch undoubtedly expected 14 Rd3. Tal, after the game, told Portisch
that he had no memory of playing the following, earlier game, which is good news for the
rest of us, who constantly forget our theory at the board! 14 ... Bb5 15 Bxb5 Qxb5 16 Ne5!?
(a rational man, like your writer, would have played the chickenhearted 16 Be5) 16 ... Qxb2
17 Rg3 Rfc8!? (17 ... Kh7 18 Rb3 Qxa2 19 Rxb7 Nd5 is okay for Black) 18 Bxh6 Bf8 19
Qe3 Qxc2 20 Re1?! (objectively, Tal should play 20 Bxg7 Bxg7 21 Rxg7+ Kxg7 22 Qg5+
Kf8 23 Qxf6 Qh7 and White has full compensation for the sacrificed material) 20 ... Rc7 21
h3 Rac8 22 Bxg7 Bxg7 23 Rxg7+ Kxg7 24 Qg5+ Kf8 25 Qxf6 Qh7 26 Re3 Qf5 27 Qh8+
Ke7 28 Qh4+ Qf6 29 Qb4+ Ke8? (29 ... Rc5! 30 Nd3 Qc3 31 Qh4+ Qf6 32 Qb4 Qc3 is
drawn) 30 Qb5+ Kf8 31 Rf3! Qd8?? (Black’s only chance lies in 31 ... Qg7! 32 Qb4+ Ke8
33 Qa4+ Kf8 34 Qd4) 32 Qb4+! Ke8 33 Rg3! and apparently Tal didn’t lose every game at
Curacao, 1-0, M.Tal-P.Benko, Curacao 1962.
14 ... Bc6 15 Bxh6!?
 Sometimes when we launch into a critical decision, we outwardly pretend to understand
more than we actually do. This attack, once begun, is a sentient creature which demands to
be fed. Well, we all saw it coming. This move is in a sense the only move, since if White
didn’t sacrifice, then his earlier Qd2 made no sense.
15 ... Ne4!
Black’s game shudders violently, yet the foundation remains fully intact. Only with this
move does hope’s surge rise within Black’s position. The players engage in a central melee,
whose outcome is not so easy for either party to foresee. This zwischenzug is necessary,
since Black must avoid:
a) 15 ... gxh6?? 16 Qxh6 Be4 17 Bd3 when Black is busted.
b) 15 ... Bxf3? 16 Bxg7 Kxg7 17 Qg5+ Kh8 18 Rh4+ Nh7 19 Qxe7 and Black is forced
into 19 ... Qd8 20 Qxd8 Rfxd8 21 gxf3 with two extra (although doubled) pawns for White
in the ending.
16 Qf4 gxh6
16 ... Bf6 17 Rxe4 Bxe4 18 Qxe4 gxh6 is a near-transposition to the game continuation.
17 Rxe4 Bxe4 18 Qxe4
 18 ... Rad8
Instead, 18 ... Bf6 transposes to the above note.

Question: Why can’t Black play 18 ... Qxb2?

Answer: This is rather greedy! White’s attack flares after 19 Ne5 Qb4 20 c3 Qd6 21
Qg4+ Kh8 22 Qh5 Kg7 23 Rd1 Qc5 24 Rd4 Rad8 25 Rf4 when Black must return material
with 25 ... Rd1+ 26 Qxd1 Qxe5 27 Rg4+ Kh8 28 Qf3. White has the better chances, since he
regained his invested material, while retaining the initiative.
19 b3

Question: Who stands better?

Answer: Both lines of argument state their case, with neither side on top. In the end, Tal
failed to derive profit from his transaction. His attack is a lot like eating a chocolate bar
with the wrapper still on. For the exchange Tal got a pawn and an insecure black king, a fair
deal for both sides – but to my mind, slightly fairer for Black, whose side I prefer.
19 ... Bc5 20 Qf4 Kg7 21 Qe5+ f6
In the animal world, a good hider is a skilled survivor. This gives Black’s king some
lateral defensive coverage with his rooks, at the cost of weakening g6.
22 Qg3+ Kh7 23 Re1 Rg8 24 Qh4
White’s queen must retain coverage of f2. The once thriving industry of White’s attack,
grows slack, with would be white attackers struggling to find jobs.
24 ... Rd6 25 Kf1
Now White eyes f6.
25 ... f5!?
 This move is multipurpose:
1. If White plays Bd3, it doesn’t arrive with check.
2. It enables ... Rg4, which knocks White’s queen off her coverage of f2.
3. It cuts off both Re4 and Qe4+.

Question: Portisch’s move accomplishes three useful functions. Why then did you not
give it an exclamation mark? Is there a down side to the move?

Answer: A couple of points:

1 The move weakens e6, since Black no longer has an ... e5 option.
2. The move allows White a prying mechanism, with h3 and g4.
26 h3!
Tal cuts off ... Rg4, while preparing g4.
26 ... Rg6 27 g4!?
Tal wasn’t the wait-around-and-do-nothing type. This move weakens both f2 and f3, but
it also increases the pressure on Black’s king. For now, White’s kingside initiative is only
one of modest proportions. This is all about to change, with Black’s next move.
27 ... Rd7??
Portisch stands slightly better after 27 ... Qc6!.
 Exercise (combination alert): Paranoid imagination is so much worse than the
actual event we so greatly fear. A dire sin on the chessboard is to allow any alteration in
your position to go unnoticed. Portisch looks at the panoramic
picture, missing a key detail. Black just blundered on his last move, ignoring a square
which was too crucial a point to avoid consideration. How did Tal exploit it?

Answer: Attraction/pin/knight fork.

28 Rxe6! Rd1+!
The only move to continue playing:
a) 28 ... Rxe6?? 29 Ng5+ Kg7 30 Nxe6+ Kh7 31 Qf6 Qd6 32 Qxf5+ Kh8 33 Qf6+ Kh7
34 Bd3+ Kg8 35 Qg6+ Kh8 36 Qe8+ forces mate.
b) 28 ... Bxf2?? 29 Rxb6 Bxh4 30 Rxg6 Kxg6 31 Ne5+ Kg5 32 Nxd7 leaves White up a
piece in the ending.
29 Kg2 Rxe6 30 Bxe6
Black’s king finds himself surrounded by attackers.
30 ... fxg4
Unfortunately for Black, the bishop is immune, since 30 ... Qxe6?? walks into the fork 31
Ng5+. Instead, 30 ... Qa6! threatening ... Qf1+ looks like Black’s best practical chance: 31
c4 Qxa2 32 Bxf5+ Kg7 33 b4! (an attempt to deflect pressure away from f2) 33 ... Bxb4 (33
... Qxf2+ 34 Qxf2 Bxf2 35 Kxf2 is a hopeless ending for Black) 34 g5! Qa6 35 Ne5 Rd6 36
Qh5! with decisive infiltration to f7, since a disproportionate number of defenders
wandered away from the defence of their king.
31 Qxg4 Rd8
 Exercise (combination alert): How did Tal finish his opponent off?

Answer: Mating net. There is no defence to the coming queen check on g6.
32 Ne5! 1-0
32 ... Qc7 33 Qg6+ Kh8 (the king waves away his sister with a petulant gesture of
impatience; she, however, soon get his full attention) 34 Qxh6+ Qh7 35 Ng6 is mate.

Game 34
M.Tal-B.Larsen
6th matchgame, Bled 1965
Alekhine’s Defence

1 e4 Nf6 2 e5 Nd5 3 d4 d6 4 Nf3 dxe5!?

Larsen, the great provoker, attempts to steer the game to non-theoretical regions. This
early exchange on e5 is in fashion today. For Tal it must have been a surprise, since in 1965
everyone played 4 ... Bg4, 4 ... g6, or 4 ... Nc6.
5 Nxe5 e6?!
When you choose to provoke a fight, then pick your enemy with great care. This isn’t
such a great line for Black, since his c8-bishop gets hemmed in. Today, Black’s main line is
5 ... c6.
6 Qf3!
 When we were kids, our main strategy was to go for cheap mates on f7. Tal lures
Larsen’s queen to f6, after which she gets chased mercilessly by White’s minor pieces.
6 ... Qf6

Question: Why must Black comply and bring the

queen out, when he can retreat his knight to f6?

Answer: 6 ... Nf6 is possible, but no improvement: 7 Be3 and advantage White, who
defends d4, while simultaneously suppressing ... c5, F.Nepustil-M.Nicholls, correspondence
2010.
7 Qg3
Threat: Bg5, followed by Bd3. White reinforces his e5-knight, while dodging Black’s
intended swap.
7 ... h6 8 Nc3!
Tal wants to eliminate Black’s only developed piece (I don’t count Black’s queen as a
developed piece), intending Ne4.
8 ... Nb4
Larsen goes after c2. 8 ... Nxc3 strengthens White’s centre: 9 bxc3 c5? (a violation of
the principle: don’t allow the position to open when lagging in development) 10 Bb5+ Bd7
11 Nxd7 Nxd7 12 Qc7 (double attack) 12 ... Rd8 13 Qxb7 cxd4 14 0-0 with an
overwhelming development lead for White, T.Kawagoe-M.Bertel, correspondence 2010.
9 Bb5+
Tal plans Ba4 next, covering c2.
9 ... c6 10 Ba4

Question: Isn’t White’s bishop out of play on a4?

 Answer: Only temporarily. Later he can play for Ne4, followed by c3 and Bc2, re-
routing the bishop to its most menacing diagonal.
10 ... Nd7 11 0-0 Nxe5
White’s e5-knight was too powerful to be left unchallenged forever.
12 dxe5 Qg6!?
Larsen, by now desperate to remove queens from the board, is willing to enter an
inferior ending.
13 Qf3!?

Question: Why would Tal reject the queen swap,

after which he gets a clearly favourable ending?

Answer: From a purist standpoint, this move is incorrect. From a practical perspective,
Tal’s decision to retain queens on the board hands Larsen exceedingly difficult over-the-
board survival decisions. Amazingly, Tal chooses theory and conjecture, over the here-and-
now clear advantage of the simple 13 Qxg6! (objectively, White’s best move, which hopes
to win a battle without suffering casualties) 13 ... fxg6 14 Ne4. White is clearly better in this
ending, which Tal, strangely enough, described as “a small achievement”. Komodo assesses
at a healthy ‘+0.91’ advantage for White, who enjoys a territorial advantage, superior
development, the preferable structure and a weak point on d6.
13 ... Qf5
13 ... Nxc2?? walks into the cheapo 14 Bxc6+! when Black gets crushed after 14 ... Ke7
15 Qd1! (threatening mate on the move, while dodging a queen swap) 15 ... f6 16 Be4.
14 Qe2
Oh, nyet you don’t. Tal sidesteps the queen swap, while covering c2.
 14 ... Be7 15 a3
Also tempting is 15 f4 0-0 16 a3 Nd5 17 Ne4 (threat: c4, followed by Bc2) 17 ... b5 18
Bb3 h5 (to suppress g4) 19 c4 bxc4 20 Bxc4 Rb8 21 h3 Qg6 22 Bd3 f5, which is
strategically awful for Black.

Question: If so, then what was Tal’s justification

for rejecting such a clear strategic advantage.

Answer: I think Tal regarded “strategic” as a dirty word! He just didn’t interpret chess
in such a logical, incremental manner, and preferred a tactical route – even one which
objectively isn’t as strong as this line.
15 ... Nd5 16 Nb5!

The knight races about with the carefree purpose of a puppy who slips out of her leash,
and prances about as she pleases. Long accumulated pent-up emotion explodes. Tal,
deciding that half-measures won’t do, is willing to pay a staggering tax for the privilege of
attacking Black’s king. Larsen said he completely overlooked this startling shot.

Question: Is this sacrifice sound?

Answer: We may be sceptical of some of Tal’s sacrifices, yet we would simply be

negating reality if we denied their effect. This isn’t a clear sacrifice of goods and bads, or
ups and downs. Sound or unsound is not the point. For Tal, not sacrificing here is akin to
parents telling their six-year-old: “We will give you your presents a week after Christmas to
take advantage of all the 50%-off sales.” For the piece, White gets:
1. Only one pawn.
2. Black’s king gets stranded in the middle of the board, with no refuge in sight.
 3. White gets a virtually endless, sustained initiative based on number 2 on the list.
Conclusion: Despite Komodo’s near-even evaluation, Black’s position strains even the
most skilled human defender. So from a practical human standpoint, White holds a clear
advantage.
16 ... cxb5!
Larsen correctly accepts the offer. When our opponent proceeds with almost insulting
assurance, we become desperate to punish the presumption. Now the players engage in a
more focused discussion on the fate of Black’s king, versus White’s material deficit.

Question: If the position is so tough for Black after

the acceptance, why not simply decline with 16 ... 0-0?

Answer: Overly extreme vigilance can be as self-destructive as outright carelessness. In

this version White gets a powerful initiative, without having to pay anything for it. For
example, 17 Nd4 Qg6 18 c4 (ejecting Black’s best posted piece, while clearing the way for
Bc2) 18 ... Nb6 19 Bc2 f5 20 exf6 Qxf6 21 Qe4 when Black is busted, without even the
solace of being up material.
17 Qxb5+ Kd8
Resigning himself to the fact that Black’s king faces a hoard of enemies in the middle.
This is forced, since 17 ... Kf8?? 18 Qe8 is mate. “Even a king lacks the power to command
death,” the queen tells her brother, as she retrieves her blade from his belly.
18 c4 Qxe5?
After this move, Black’s unfortunate king and queen experience the tribulations of an
unreflective life. Larsen returns the piece in the wrong way. Correct is 18 ... Nb6! 19 Qa5
(threat: c5) 19 ... Kc7! 20 c5 Kb8! 21 cxb6 axb6 22 Qb5 Qe4 23 Bd1 Bc5! (threatening e5)
24 Bf3!? (I can’t see Tal ever agreeing to play a move like 24 Qe2) 24 ... Qxe5 (threat: ...
Bxf2+ and ... Qxb5) 25 Qd3 when White retains compensation for the pawn, but no
advantage.
19 cxd5 Bd6 20 g3 Qxd5 21 Qe2
 For the pawn, White receives:
1. A massive development lead.
2. Black’s king remains huddled in the centre, miles away from safety.
3. Open central lines enhance White’s advantages in numbers 1 and 2 on the list.
Conclusion: Black is busted.
21 ... Ke7
Larsen hopes to have time for ... Rd8 and ... Kf8. He never gets the time to complete his
intent.
22 Rd1 Qa5 23 Qg4
Protecting his bishop, while pressuring g7.
23 ... Qf5!
This is a loveless marriage between black queen and king. In the end she abandons him.
Larsen finds Black’s best defensive option:
a) 23 ... h5?? 24 Qxg7 Qxa4 25 Bg5+ Kd7 26 Qxh8 and Black gets slaughtered.
b) 23 ... e5 24 Qh4+ f6 25 Bd2 g5 26 Qc4 Qc7 27 Qd3 Be6 28 Rac1 Qb8 29 Ba5!
leaves Black in zugzwang. If 29 ... f5 30 Bd7!! e4 31 Qd2 Bxd7 32 Rc7! and Black’s
position crumbles after this pretty interference shot.
c) 23 ... g5?? 24 Rxd6! Kxd6 25 Qd4+ Ke7 and now White has the killing zwischenzug
26 Bd2!, winning on the spot, since 26 ... Qd8 walks into mate after 27 Bb4+.
24 Qc4
Tal avoids Larsen’s trap 24 Qxg7?? Be5 25 g4 Bxg7 26 gxf5 Rg8 27 Kf1 exf5 when
suddenly Black is the only one with winning chances.
24 ... Qc5
If 24 ... Rd8 25 Be3 and Black is unable to complete development, since 25 ... Bd7?? is
met with the crushing 26 Rxd6! Kxd6 27 Qb4+ Kc7 28 Bf4+ e5 29 Qc5+ Bc6 30 Bxe5+
Kd7 31 Rd1+ Ke8 32 Bxc6+ bxc6 33 Re1, with a winning attack.
 25 Qd3!

One of the ideas behind this move is that a future ... Rd8 can be met with a Qh7
infiltration.
25 ... Qd5
25 ... Rd8 puts up greater resistance. In the end, Black also loses after 26 Bf4 Qb6 27
Bxd6+ Rxd6 28 Qc3 Rxd1+ 29 Rxd1 Kf8 30 Qe5 (Black is unable to develop his entire
queenside sector) 30 ... Kg8 (30 ... f6?? is met with 31 Qh5!, threatening mate on e8; after
31 ... Kg8 32 Qe8+ Kh7 33 h4 Qc5 34 h5 there is no good defence to the coming Qg6+ and
Rd8+) 31 h4 Kf8 32 Rd4! (intending to meet ... Kg8 with Rg4!; note how in all these lines,
Black’s queenside pieces remain still as wood carved images on your mother’s shelf) 32 ...
a6 33 Kg2 a5 34 Kg1! and zugzwang.
26 Qc3
There wasn’t a chance in a trillion Tal would agree to 26 Qxd5? exd5 27 Bb3 when
White regained the lost pawn, at the cost of his decisive advantage.
26 ... Be5 27 Qe1 Qc5 28 Bd2
Threat: Bb4.
28 ... Kf6 29 Rac1
White’s developmental resources continue to multiply, while Black’s remain in endless
stagnation.
29 ... Qb6 30 Be3 Qa6
The black queen’s fluttering squawks remind us of a dozing parrot, startled in her cage.
White’s position is taut with expectation. After Tal’s next move, Black’s position collapses,
but if 30 ... Qxb2 31 f4! when Black’s bishop is unable to retreat, since then White has
Bd4+, winning Black’s queen.
31 Qb4!
It goes without saying that White’s queen is a person of a rather touchy disposition.
Threat: Bb5, trapping Black’s queen.
31 ... b5
Likewise, 31 ... Qe2 32 Qe4! (threat: Bg5+) 32 ... Qh5 (32 ... Qxb2 loses instantly to 33
f4) 33 f4 Bxb2 34 Rc7 Qa5 (34 ... Qf5 35 Qb4 forces resignation) 35 Bd4+ Bxd4+ 36
Qxd4+ Kg6 37 Bc2+ leaves Black without a viable response.
32 Bxb5 Qb7
Do you sense an MC Escher endless staircase feel to the position? In a way, this game
looks like it was played by two beginners, since this is Black’s 13th queen move of the
game, while Tal moved his queen 12 times – yet each queen move was perfectly justified!
33 f4!
Chasing away Black’s only defender.
33 ... Bb8
33 ... Bc7 34 Rd7!, and if 34 ... Bxd7 35 Bd4+ e5 (35 ... Kg6 hangs the queen to 36
Bd3+) 36 fxe5+ Kg5 37 h4+ when every king move allows a check from White’s light-
squared bishop, winning Black’s queen.

Exercise (combination alert): All of Black’s army remains on the

first rank, except for his unrepentant king and queen. White
has several paths to the win. Find the simplest solution:

Answer: Trapped piece. Black loses a rook.

34 Bc6! 1-0

Game 35
A.Kapengut-M.Tal
Latvian Championship, Riga 1965
Sicilian Defence

1 e4 c5 2 Nf3 e6 3 d4 cxd4 4 Nxd4 Nc6 5 Nb5

5 Nc3 Nf6 6 Ndb5 d6 7 Bf4 e5 8 Bg5 gets us into a true Pelikan Sicilian.
5 ... d6 6 Bf4
6 c4 is White’s most common line here.
6 ... e5 7 Be3 Nf6 8 Bg5
Instead, 8 N1c3 a6 9 Na3 b5 10 Nd5 Rb8 is about even.
8 ... a6 9 N5c3
After 9 Bxf6 gxf6 10 N5c3 f5 11 Bc4 Qg5 12 g3 Be6 13 Nd5?! (White should try 13
Nd2) 13 ... 0-0-0 I already prefer Black, whose bishop-pair and development lead outweigh
White’s control over d5, P.Leko-S.Movsesian, Moscow (blitz) 2008.
9 ... Be7 10 Bxf6 Bxf6 11 Nd5

Question: Are we in a Pelikan Sicilian?

Answer: Not quite. This is a hybrid Pelikan-like line. In a Pelikan proper, Black has ...
b5, while White’s b1-knight would be on the more awkward a3-square.
Instead, after 11 Nd2 0-0 12 Nc4 b5!? 13 Qxd6 bxc4 14 Qxc6 Be6 15 Be2 Rb8 Black’s
development lead and bishop-pair give him full compensation for the pawn, G.Kasparov-
V.Anand, Frankfurt (rapid) 1998.
11 ... Bg5 12 Bc4 0-0 13 0-0 Be6 14 Bb3
A couple of years later Kapengut tried 14 Nbc3 b5 15 Bb3 Nd4 16 Ne2 Nxb3 17 axb3
Qb8 18 b4 Qb7 19 Nec3 Rac8 20 Qh5 f6 21 Rfd1 Rc4 22 Rd3 Bxd5 23 Rxd5 Rxb4 24 h4
Bh6 25 Rxd6 Qc7 26 Qd1 Rxb2, A.Kapengut-S.Furman, Leningrad 1967. At this point
White has a strategically won game if he finds 27 Rdxa6!, since 27 ... Qxc3?? is impossible,
due to 28 Qd5+ Kh8 29 Ra8 g6 30 Rxf8+ Bxf8 31 Qf7 Ba3 32 Rd1 Qc8 33 Qxf6+ Kg8 34
Rd8+ winning.
14 ... Bh6

Question: I realize Black has the bishop-pair, but I don’t like

the fact that White dominates d5, and Black’s d6-pawn is backward
and a potential future target. What is Black’s plan in such a position?

Answer: Admittedly, we can’t accurately describe Black’s structure as a naturally

occurring geological formation. In such structures, Black’s main sources of counterplay are:
1. Play on the queenside, in conjunction with the open c-file and possible a ... b4 pawn
break.
2. An ... f5 break, usually prepared with ... Kh8, ... g6 and only then ... f5. If White
captures on f5, then Black normally recaptures with the g-pawn, opening the g-file for a
potential attack.
15 Nbc3 Kh8
15 ... Nd4 doesn’t bother White, since his knights hold their own against Black’s bishop-
pair after 16 Ne2 Nxb3 17 axb3.
16 Qd3 Rc8 17 Rad1
White hopes to apply pressure to d6 down the open file.
17 ... Qh4
Tal begins to fish for kingside counterplay, based on a future ... f5 break, as discussed
above.
18 f3!?
I’m not so sure about this move, which reinforces e4 at the cost of weakening his
kingside dark squares. I would try 18 Nb6! Rcd8 19 Ne2, which covers against ... Nd4, and
temporarily prevents ... f5 while preparing to possibly swing the knight to g3. I slightly
favour White’s chances here.
18 ... Nd4 19 Ne2?!
Now the advantage swings to Black. White should try 19 g3 Qh5 20 Ne7 Rcd8 21 Ncd5
intending c3 next.
19 ... Bxd5!
When there is no way to conciliate opposing positions, conflict is inevitable.
20 Nxd4!
White alertly avoids Tal’s devious trap 20 Bxd5? Rxc2!, and now if 21 Nxd4?? the
killer zwischenzug 21 ... Bf4! forces mate in three moves, after 22 h3 Rxg2+! 23 Kxg2 Qg3+
24 Kh1 Qh2. Instead, after 20 exd5?! Nxb3 21 axb3 f5 Black is the one with the more
dangerous pawn majority.
20 ... exd4 21 g3
Also advantageous for Black is the line 21 Bxd5 Be3+ 22 Kh1 Bf4 23 g3 Bxg3 24 Rd2
Be5 25 Bxb7 Rb8 26 Qxa6 Qh3 27 Bd5 Rxb2 28 Bb3 Rc8 29 Rdf2 g5. The presence of
opposite coloured bishops favours Black in this position. Still, this may have been White’s
best line, given his circumstances.
21 ... Be3+! 22 Kg2
22 Qxe3?? loses to 22 ... dxe3 23 gxh4 Bxb3 and White’s problem is that he has no time
to recapture the bishop, since ... e2 is threatened.

Exercise (combination alert): White looks okay here. After all, when Black moves
his queen, Bxd5 follows, with an acceptable position. What did Tal play instead?

Answer: Queen sacrifice.

22 ... Bxb3!!
 The reverential attitude is shattered with a blast of disrespect.
23 gxh4 Bxc2 24 Qa3
24 Qe2 is met with 24 ... Ba4! with threats of both ... Rc2 and ... Bb5, with a clear
advantage for Black.
24 ... Bxd1 25 Rxd1 f5

Question: If we play games of intrigue, then don’t we first

have to accumulate the capital to finance them? Isn’t Tal’s
sacrifice more a guess, than an actual statement of fact?

Answer: Let’s look at the data. For the queen, Tal got:
1. A rook and bishop.
2. White must watch out for infiltration on both c2 and f2.
3. White’s king remains in serious danger.
Conclusion: On an objective level, Tal’s compensation is tentative, rather than assured.
Yet I think Black stands better on a practical level, since White’s moves become very diffi​-
cult to find.
26 exf5 Rc5!
If a single powerful leader fails to arise, multiple minor leaders fill the power vacuum
by assuming command. The rook is transferred into the attack, via f5.
27 Qb4 Rcxf5
Tal decides to sever his relationship with his queenside pawns. Now White faces vexing
questions:
1. Should he capture b7?
2. Should he capture d6?
3. Should he defend f3?
 28 Qxd6?!
How infuriating, when we agree to a negotiation with our opponent, and then a few
moves later, realize that we have been cheated. This is the jump ball moment, where White’s
defensive chances can go either way. The first serious inaccuracy is induced, after which
White’s position threatens to submerge further into misery. White may hold the draw with:
a) 28 Qxb7! Rb5 29 Qxa6 Rxb2+ 30 Kh1 h6 31 Qxd6 Rxf3 32 h5 Rff2 33 Qe5 Rxa2 34
Qe8+ Kh7 35 Qe4+ with perpetual check.
b) 28 Rf1 is admittedly passive, but probably still okay for White after 28 ... Rc5 29 f4
b5 30 Qa5 Rc6 31 f5 d5 32 Qb4 Rff6 33 Kh3 h5 34 Qe7 Kh7 35 Qe8 Rc2 36 Qxh5+ Rh6 37
Qg4. Now Black must take a repetition draw with 37 ... Rf6, since the greedy 37 ... Rxb2??
fails to 38 f6! gxf6 39 h5 with a winning position for White after 39 ... d3 40 Rf5 d2 41
Rxd5.
28 ... Rxf3 29 Rxd4??
This move is a bit like watching Jaws and then going for a relaxing swim in the ocean.
Now Tal gets the happy Disney version. It isn’t a great idea to steal material earlier in the
game, at the cost of handing the opponent the initiative, and then later be conflicted with a
sudden attack of conscience. This attempt to buy Black off fails, since it hinders
communications in the defensive supply line.
White’s only chance to hold the game lay in the line 29 h5! (seizing control over the
crucial g6-square is necessary for White’s continued survival) 29 ... R3f6 30 Qe7 b6 and
it’s not clear if Black can force the win.
29 ... Rf2+!
This move is not a request, but instead a demand, spoken in an unanswerable tone.
30 Kg3
30 Kh1?? walks into 30 ... Rf1+ 31 Kg2 Rg1+ 32 Kh3 Rf3+, winning the house.
 Exercise (combination alert): The crown fits aslant on the white king’s
head, and there are many who feel he is about to be deposed. Black has
multiple paths to the win. How would you close the game out?

Answer: Double attack. White’s queen is attacked, while ... Rg6+ can’t be prevented.
30 ... R2f6! 0-1
Black attackers are bound in a singular interest. 31 Qd5 Bxd4 32 Qxd4 Rg6+ 33 Kh3
Rf3 is mate.
Answer no.2: Also winning is 30 ... Bxd4, and if 31 Qxd4 R8f3+ 32 Kg4 h5+! 33 Kxh5
Rg2! 34 Qe4 Rff2 35 a4 Rxb2 with a winning position for Black.

Game 36
M.Tal-B.Larsen
10th matchgame, Bled 1965
Sicilian Defence

I love playing through Tal-Larsen battles, since they, along with Korchnoi, were the most
uncompromising risk-takers of their day.

1 e4 c5
It was impossible to prepare for Larsen, since he played virtually every known opening,
with White and Black.
2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 e6 5 Nc3 d6 6 Be3 Nf6 7 f4 Be7 8 Qf3!?
Unsurprisingly, Tal picks the most cut-throat continuation, planning to castle long and
toss in a quick g4. 8 Be2 is the more positional route.
8 ... 0-0
Larsen is up for a fight. Black can also apply the principle: Counter a coming wing
attack with the immediate central counter: 8 ... e5 9 Nxc6 bxc6 10 f5 d5 11 exd5 cxd5 12
0-0-0 0-0 13 Nxd5 (Komodo suggests the unplayed 13 g4!?) 13 ... Nxd5 14 Bc4 Bxf5 15
Bxd5 Qc8 16 Qf2 Rb8 17 Bxa7!? Rb5 18 Be3? (White may still be okay after 18 Kb1) 18 ...
Be6 19 Bxe6 Qxe6 20 Kb1 Ra8 and Anand found himself under a heavy attack, V.Anand-
A.Khalifman, Zurich (rapid) 2009.
9 0-0-0 Qc7 10 Ndb5
This move doesn’t cost Black time, since he later regains the lost tempo with ... a6.
Kasparov suggests the immediate 10 g4. If 10 ... Nxd4 11 Bxd4 e5 12 fxe5 dxe5 (12 ...
Bxg4? 13 Qg3 is clearly in White’s favour) 13 Qg3 with a slight edge for White.
10 ... Qb8 11 g4
 When the players voluntarily enter such a line, they operate on violently disparate
assumptions, in that they both believe that the other will be mated. They can’t both be
correct, since on the chess board, the rules specify that only one side can be checkmated in a
given game.
11 ... a6 12 Nd4 Nxd4 13 Bxd4 b5!?
Larsen pursues his own queenside agenda, rather than insert the thematic central counter
with an immediate ... e5. The trouble is White’s attack in Larsen’s continuation just feels
faster. More thematic is 13 ... e5! threatening both White’s d4-bishop and also ... Bg4.
According to the comps, this is Black’s best move, yet White still enjoys a rather substantial
plus score, which may tell us that Black’s game is not so easy to navigate – even if his or her
position is inherently playable. 14 g5 Bg4 15 Qg3 exd4 16 Rxd4 Be6 17 f5 was Z.Hracek-
V.Babula, Ostrava 2010. Black should be okay after 17 ... Nh5 18 Qh4 Qa7! 19 Rd1 Bxa2
20 Nxa2 (or 20 Qxh5 Qe3+ 21 Rd2 Bxg5 22 Qe2 Be6 23 fxe6 fxe6 with the better chances
for Black) 20 ... Qe3+ 21 Kb1 Qxg5 and Black stands at least even, since his control over
the dark squares looks slightly more important than his opponent’s control over the light
squares.
14 g5 Nd7 15 Bd3 b4 16 Nd5!?
 I dream of owning a dog who faithfully trots at my heel, and obediently does what he or
she is told. Tal’s knight is quite a bit like the dogs I own now, and in the past: they do
whatever the hell they please! Tal’s move is played with the philosophy: nothing is broken,
but I’m going to fix it anyway. Such moves fill us with both the thrill of the hunt, and also
apprehension at our own recklessness. This attack, once begun, is much like a nest of new-
born birds, all screaming to be fed. Tal decides he won’t allow himself to be preoccupied
with trivial matters, and refuses to retreat, preferring payment on an endless debt. Most of us
expend so much of our energy and resources attempting to rescue ourselves from our own
past unwise decisions. Tal is the exception, since he so often profited from his own unwise
move choices!

Question: You seem to be completely sympathetic to Tal’s shady

sacrifices in the book, when objectively they should get either a
dubious, or a full question mark. Why doesn’t this one get a dubious mark?

Answer: Tal was the carnival huckster, whose fortunes always arose by pandering his
tactical wares to the gullible, which in his case was the rest of the world! I fully admit this
one is just plain unsound. Kasparov called it “a losing move”. For the piece, Tal got one
pawn and slows Black’s attack, while speeding up his own – just not enough.
But let’s not forget the psychological factor. Tal’s outrageous sacrificial decision so
shocked Larsen, that it induced back-to-back blunders. In chess, I believe the intuitive
represents a more potent force than the scientific, so I’m not going to criticize a player’s
unsound (yet effective!) style when it works, over and over again.
Certainly more reasonable is 16 Ne2. Most of us prefer to meet provocation with polite
deference. In battle, evasion is often the wiser strategy, when compared to head-on
confrontation. So then why does it always feel like the less honourable course? After 16 ...
e5 17 Be3 exf4 18 Qxf4 Ne5 19 Nd4 Qc7 20 Be2 Be6 Black stood no worse, C.Koch-
M.Smet, correspondence 1999.
16 ... exd5 17 exd5
White’s formidable bishops reassemble the conclave. Both take aim at Black’s king.

Exercise: (critical decision): White threatens Qe4, with a decisive

double attack on h7 and e7. This means Black has a choice of two
defensive pawn moves: 17 ... f5 or 17 ... g6. One of them allows White’s
attack to bloom, while the other snuffs it out. Which one would you play?

17 ... f5?
“The wrong pawn,” writes Kasparov. Black suffocates White’s attack with the
counterintuitive refutation:
Answer: 17 ... g6!! is the master equation, which solves Black’s defensive needs.
 Question: Won’t Black get mated if White goes for the
simple plan of h4 and h5, followed by Rh2 and Qh1?

Answer: Believe it or not, Black’s fears of getting mated are groundless. For example,
18 h4 Nc5 (this move either eliminates White’s light-squared bishop, or chases it away) 19
Bc4 Bf5! 20 h5 Qc7 21 b3 (21 Rh2 is met with 21 ... Ne6! 22 dxe6 Qxc4 23 hxg6 fxg6 with
a winning position for Black) 21 ... Rac8 22 Rde1 (or 22 Rh2 Bd8! 23 Rdh1 Be4 24 hxg6!,
which is tricky, but it doesn’t bother Black, just as long as he doesn’t fall for the mate in
one; 24 ... fxg6! 25 Qh3 Bxh1 26 Bxc5 Rf7! is a key defensive move, and after 27 Bxb4 Be4
White’s attack is at an end, since 28 Qe3 is met with 28 ... Qb6! 29 Qxe4 Re7! winning) 22
... Qd7 23 Rh2 Bd8 24 Qh1 Bb6 25 Bf6 Rfe8. When we escape a grave danger, a deep
feeling of serenity washes over us. White’s attack hit a dead end and Black converted,
David-C.Coco, correspondence 1997.
18 Rde1!?
Unbelievably, Tal shows no interest in regaining his invested piece with the line 18
gxf6! Nxf6 19 Rhg1 Rf7 20 Rde1 Qb7 21 Bxf6! Rxf6 (not 21 ... Bxf6?? 22 Re8+ Rf8 23
Bxh7+ Kxh7 24 Rxf8 and Black can resign) 22 Qe4 g6 23 Qxe7 Qxe7 24 Rxe7 Rxf4.
White’s chances are at least even in this ending.
18 ... Rf7?
Our ego always demands that we declare that our losses are either the result of an act of
God, or a natural disaster, rather than through our opponent’s skill, or our own
incompetence. Now White is virtually winning! Larsen, who was a notorious sore loser (he
once wrote an article about how his infamous 0-6 Candidates’ match loss to Fischer was a
fluke), would still be okay after 18 ... Bd8! when Black’s position demonstrates its
durability against every conceivable line of attack. Correct is 19 Qh5! Nc5 and only now 20
Bxg7! Nxd3+ (20 ... Kxg7?? walks into mate after 21 Qh6+ Kg8 22 g6) 21 Kb1! Qc7 22
Bxf8 Nxe1 23 Rxe1 Qf7 24 Qxf7+ Kxf7 25 Bxd6 a5 26 Re5 with only an edge for Black in
the ending.
19 h4!

Intending h5 and g6.

19 ... Bb7 20 Bxf5!?
When we attack and win, we can go for heavy reparations, as in this case. Or we can
play for complete annihilation of the enemy. Sometimes fancy isn’t always best. White
should ignore his mini-combination and ruthlessly open lines with:
a) 20 h5! and Black is helpless. For example, 20 ... Bxg5 (or 20 ... Nf8 21 g6 hxg6 22
hxg6 Nxg6 23 Qh5 Nh4 24 Rxe7! Rxe7 25 Rxh4 Qf8 26 Bxf5 and Black is mated) 21 fxg5
Ne5 22 Bxe5 dxe5 23 Bxf5 Kh8 24 g6 Bxd5 25 Qxd5 Rxf5 26 h6! Ra7 27 Rh4 Rf6 28 Rd1
gxh6 29 Rxh6 Rg7 30 Qxe5! when White’s exploitation of Black’s weak back rank is
decisive.
b) Tal also considered 20 g6! hxg6 21 h5 g5 22 Bxf5! with a strong attack for White, but
this still isn’t as decisive as line ‘a’. Black has no choice but to continue with 22 ... Rxf5 23
Rxe7 Qf8! (23 ... Rf7?? 24 Rxf7 Kxf7 25 Qe4! is deadly) 24 Rxg7+ Qxg7 25 Bxg7 Kxg7 26
Qe4 Raf8 27 Rg1 Bxd5 28 Qe7+ R8f7 29 Qxd6 Nf8 30 fxg5 when White’s queen and three
extra pawns will beat Black’s rook and two minor pieces.
20 ... Rxf5 21 Rxe7 Ne5
The defenders groan in protest, barely able to hold off the attackers, and the seemingly
eternal cascade of involuntary concessions continues. Larsen returns the piece in
desperation, seeing the line 21 ... Rf7? 22 Rxf7 Kxf7 23 g6+! hxg6 24 h5. Black’s king
won’t survive.
22 Qe4 Qf8 23 fxe5 Rf4 24 Qe3
 24 ... Rf3?
A time pressure mistake. Black offers better resistance with 24 ... Bxd5! 25 exd6 Rxd4!
(25 ... Bxh1? 26 Qe6+ Kh8 27 Bxg7+ Qxg7 28 Rxg7 Rf1+ 29 Kd2 Kxg7 30 d7 Rf7 31 Qe5+
Kg8 32 Qe8+ Rf8 33 Qe6+ Rf7 34 g6 hxg6 35 Qxg6+ wins, since 35 ... Rg7 is met with 36
Qe8+) 26 Qxd4 Bxh1 27 b3 (White has the deadly threat of h5 and h6) 27 ... Bf3! (this loses
the bishop, but weirdly enough, it is Black’s best move: 27 ... Kh8? is met with the crushing
28 h5 intending h6, and 27 ... a5? 28 h5 a4 29 bxa4, and if 29 ... Rxa4 30 h6 forces mate) 28
Qc4+ Kh8 29 Rf7 Qe8 30 Rxf3 Qe1+ 31 Kb2 Qe5+ 32 c3 Qh2+ 33 Ka1 Re8 34 Rf1 bxc3
35 Qxc3 Qxd6. Conversion won’t be all that easy in the major piece ending.
25 Qe2 Qxe7
25 ... Bxd5 26 exd6 Qf4+ 27 Qd2 is also lost for Black.
26 Qxf3 dxe5 27 Re1 Rd8 28 Rxe5
As all corporate raiders understand: when you control 51%, you seize control over the
entirety. White is up two pawns, with zero compensation for Black.
28 ... Qd6
Also hopeless for Black is 28 ... Bxd5 29 Rxe7 Bxf3 30 c3! bxc3 31 Rxg7+ Kf8 32
Bxc3.
29 Qf4! Rf8
29 ... Bxd5?? 30 Re8+ wins Black’s queen.
30 Qe4 b3!?
The idea is to clear the b4-square for Black’s queen. Of course, such a protest is more a
formality than an actual danger to White.
31 axb3 Rf1+ 32 Kd2 Qb4+ 33 c3 Qd6
 Exercise (combination alert): White has two
methods of forcing the win. Find one of them:

Answer: Deflection/weak back rank.

34 Bc5!
Answer no.2: Also effective is 34 Re8+! Kf7 35 Rd8!! (deflection) 35 ... Qxd8 (or 35
... Qh2+ 36 Kd3 Rd1+ 37 Kc4 Qc7+ 38 Bc5 and Black’s checks run out) 36 Qe6+ Kf8 37
Bc5+ forcing mate. “My single shortcoming of character is my inability to forgive,” laments
the bishop.
34 ... Qxc5 35 Re8+ Rf8 36 Qe6+ Kh8
 Exercise (combination alert): Finish Tal’s combination:

Answer: Weak back rank.

37 Qf7! 1-0
37 ... Qxd5+ 38 Qxd5 Bxd5 39 Rxf8+ Bg8 is game over for Black.

Game 37
V.Tseshkovsky-M.Tal
Sochi 1970
Ruy Lopez

1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7 Bb3 0-0 8 d3
 Question: Why do so many titled players take this quiet route? Now if and
when White achieves d4, he expends two tempi to accomplish his goal.

Answer: For decades, penicillin saved countless lives. Today’s bacteria grows smarter,
developing resistive defences against the compounds within antibiotics. Moral: nothing lasts
forever. In the same way, the Ruy Lopez Marshall Gambit terrorized many a white player
over the last century. Today, many top GMs opt for the quieter d3 line, which is designed to
avoid the Marshall Gambit, after 8 c3 d5.
8 ... d6 9 c3 Na5 10 Bc2 c5 11 Nbd2 Re8
11 ... Nc6 is played slightly more often.
12 Nf1 Bf8
 Question: What is Black’s plan in such positions?

Answer: In this case it would be to play for a thematic ... d5 break. Black must first
reinforce e5 before he can achieve his goal.
13 Ne3 Bb7
13 ... Nc6 prevents White’s next move.
14 b4!?
Also played is 14 Nf5 d5 15 Bg5 h6 16 Bh4 g6 17 Ne3, L.Vajda-B.Gyurkovics,
Hungarian League 2009. I don’t believe Black stands worse after 17 ... d4.
14 ... cxb4 15 cxb4 Nc6 16 a3 d5
Tal achieved his thematic central pawn break and equalized comfortably. 16 ... a5 is
another viable model: 17 bxa5 Nxa5 18 Rb1 Qd7 19 Bb2 Bc6 20 Qd2 g6 21 d4 exd4 22
Qxd4 Bg7 23 e5 dxe5 24 Nxe5 Qxd4 25 Bxd4 Nd7 26 Nxc6 Nxc6 27 Bxg7 Kxg7 and the
players agreed to a draw, V.Tseshkovsky-A.Karpov, Leningrad 1971.
17 Nxd5 Nxd5 18 exd5 Qxd5
Now White must contend with a backward d-pawn.
19 Bb3 Qd7!
Tal correctly keeps close watch over his f7-square. His move looks more accurate than
19 ... Qd6 20 Ng5 Re7, G.Serper-Y.Yakovich, Pinsk 1986. At this point White looks better
after 21 Be3! intending Bc5.
20 Ng5 Nd8 21 Qh5
 Tseshkovsky opts for a not-so-surreptitious deployment around Tal’s king.

Question: It feels as if Tal is White, not

Black in this position! Is Black’s king in danger?

Answer: Sometimes the weakest point in the enemy camp – in this case, f7 – is not
always the best target, since that is the one which the defender most heavily guards.
Everything is under control. Black can defend every attacked point and stands no worse.
Also, if White’s initiative should fade, then he may get stuck nursing his weak d3-pawn.
21 ... h6 22 Re3?!
Our will doesn’t always determine our destiny. If you consider entering a high-risk
venture without all the necessary data, you better be a good guesser! Moves like this either
bring short-term happiness, or if it goes sour, long-term pain.
 Exercise (critical decision): White’s last move was overly optimistic.
Black can play either 22 ... hxg5, 22 ... Bd5, or 22 ... Qf5. One line leads to
an advantage for White, another leads to dynamic equality, while a third
line leads to a winning position for Black. Which one should Tal play?

22 ... Bd5?!
This one only leads to an equal game.
Answer: Tal should have gone for 22 ... Qf5! 23 Rh3 e4! (principle: meet a wing attack
with a central counter), and if 24 Nxe4? Rxe4! 25 dxe4 Qxe4 (threatening both a back rank
mate on e1, and also ... Qxg2 mate) 26 Kf1 Qxg2+ 27 Ke2 g6 28 Qh4 Bg7 29 Rg3 Qc6 30
Rb1 Ne6 when Black has a winning attack.
Instead, 22 ... hxg5? 23 Rh3 costs Black his queen. After 23 ... Qxh3 24 Qxh3 Ne6 25
Bb2 Black doesn’t have enough for the queen.
23 Bxd5 Qxd5 24 Ne4 Re6!
Tal sees Rg3 coming and prepares to utilize lateral defence of h6 with his rook.
25 Rg3 Kh8
Covering against a Bxh6 threat.
26 Be3 a5!
It’s disorienting to see Tal in the role of counter-attacker, rather than attacker. Most of
White’s pieces are away on the kingside, so Tal creates counterplay on the other side of the
board, where White isn’t prepared for confrontation.
27 Bc5 Bxc5 28 Nxc5?!
Correct was 28 bxc5.
28 ... Rf6?!
Too aggressive. Black seizes the advantage after 28 ... Re7! 29 Rf1 Nc6 when ... axb4 is
threatened.
29 Rf1?
Tseshkovsky worries about his f2-square. With this move, White lays the foundation for
his future difficulties. White should try 29 Ne4! Rfa6 30 bxa5 f6! (30 ... Rxa5?? is met with
the crushing shot 31 Nf6! Qe6 32 Qxh6+! gxh6 33 Rg8 mate) 31 Rb1 Qf7 32 Qxf7 Nxf7 33
Rxb5 Rxa5 34 Rxa5 Rxa5 35 d4 exd4 36 Rd3 Ne5 37 Rxd4 Rxa3 and the game ends in a
likely draw.
29 ... axb4 30 axb4
Chess is a cruel endeavour, since failure on a single move may undo all the hard work of
the entire game.

Exercise (combination alert): It almost feels like an impossibility that Black has a
combination in this position. Yet it exists. Tal soon seizes upon a trivial-seeming
weakness, and then uses it to extort further concessions from his opponent. How?

Answer: Weak back rank.

30 ... Rxf2!!
A small force may overwhelm a larger one, only if it has the element of surprise on its
side. This shot came from nowhere. Now it feels like Tal is Black again! Doesn’t it strike
you as strange that we chess players wage war in the country of mind, and the only physical
contact we have with our opponent is the handshake at the beginning of the game? Yet being
victim to such shots almost brings on physical pain, thanks to the damage done to our
psyche.
31 Re1
The only move. White’s dilemma is represented in two opposing issues:
1. He must safeguard his own weakened king.
 2. He must continue to generate threats against Black’s king.
If:
a) 31 Rxf2?? Ra1+ 32 Rf1 Qd4+ 33 Kh1 Rxf1 mate.
b) 31 Kxf2 Qd4+ 32 Re3 (or 32 Kf3?? Qf4+ 33 Ke2 Ra2+ 34 Ke1 Qd2 mate) 32 ...
Ra2+ 33 Kg1 (33 Kf3?? Qf4 mate) 33 ... Qxe3+ 34 Kh1 Rf2 35 Rd1 Qf4 (threatening a back
rank mate once again) 36 h3 Nc6 and b4 falls as well, leaving White completely busted.
31 ... Rf4!
Tal keeps playing on White’s weak back rank. He attacks b4, realizing that his e5-pawn
is tactically defended.
32 h3
A move made with a feeling of profound relief, rather than:
a) 32 Qxe5?? Qxe5 33 Rxe5 Ra1+ and once again, White’s weak back rank proves to be
his undoing.
b) 32 Rxe5?? Qd4+ 33 Rge3 Ra1+ 34 Qd1 Rxd1 mate.
32 ... Ra1!
Tal’s dirty tricks continue.
33 Rxa1 Qd4+ 34 Kh2 Qxa1 35 d4?!

Exercise (critical decision): White’s last move, objectively incorrect, sets up a

pair of traps. Black has four reasonable looking responses here; two of them
work and two of them lose. Tal can play: a) 35 ... exd4, winning a pawn. b) 35 ... Qxd4,
winning a pawn. c) 35 ... Rf1, threatening mate on h1. d) 35 ... Nc6, calmly
bringing into play his only inactive piece. Which one would you play?

Answers: ‘c’ and ‘d’ win, while ‘a’ and ‘b’ both lose.
35 ... Nc6!
 Alternatively:
a) 35 ... exd4?? 36 Qe5! Rf6 37 Qe8+ Kh7 38 Nd7! and Black gets wiped out.
b) 35 ... Qxd4?? walks into 36 Rd3! Qa1 37 Rxd8+ Kh7 38 Nd7 and Black is done for.
However, line ‘d’ also works, after 35 ... Rf1! 36 Re3 Nc6 37 dxe5 Nd4 38 e6 Qc1!,
and if 39 Qe5 Nf5! when White must hand over heavy material to avoid mate.
36 Ne6!?

The knight passes his own jurisdiction into ungovernable land. Helplessness, coupled
with frustration, is a dangerous mix. A mistimed combination is the actor who rushes on
stage and delivers his lines in the wrong act and scene of the play. Technically this isn’t such
a great move, yet it’s a good practical try, forcing Tal to work out defensive issues, which
pile up in layers, like geological strata. Instead, 36 d5 is met with 36 ... Rf1 37 Re3 Qc1 38
Re4 Nd4 39 Qxe5 Nf5! which is similar to line ‘d’ from above.
36 ... Rf1!
Tal isn’t about to fall for 36 ... fxe6?? 37 Qe8+ Kh7 (the king stands on tiptoe and is
discouraged to find that his sister still towers over him) 38 Qg6+ Kg8 39 Qxg7 mate.
37 Rxg7
Suddenly this looks pretty scary for Black. Tal had everything worked out.
37 ... Rh1+ 38 Kg3 Qe1+ 39 Kg4 Qe2+ 40 Kh4
It looks like the king marches in grave procession to his own funeral.
40 ... Qxh5+?!
Of course such a move curtails White’s ability to attack. This wins, but Tal missed the
immediate game-ender 40 ... Qf2+! 41 g3 Qf6+ 42 Ng5 Kxg7.
41 Kxh5 Rf1!
Avoiding 41 ... fxe6? 42 Rc7 exd4 (42 ... Nxd4?? walks into 43 Kg6 when Black is
unexpectedly mated) 43 Rxc6 Kg7 44 Rxe6 when the game should end in a draw.
42 d5
 After 42 Kxh6 Rf6+ 43 Kg5 Rxe6 44 Rxf7 Nxd4 Black should convert with his extra
piece.
42 ... Nd8!?
Preying on White’s overloaded knight. Tal decides not to haggle over trivialities and
avoids 42 ... fxe6 43 Kg6 exd5 44 Rc7 Nd8 45 Re7 Kg8 46 Rxe5 Nc6 47 Re6 Nxb4 48 Rb6
d4 49 Rxb5 Nc6 50 Kxh6 d3 and Black is winning.
43 Kxh6 Nxe6 44 Rh7+
44 dxe6?? loses instantly to 44 ... Rf6+.
44 ... Kg8 45 dxe6
A bad harvest foreshadows increased market prices, in the future, since prices
invariably rise in times of scarcity.

Exercise (planning): Black to play and force the win:

Answer: Simplification.
45 ... Rf6+! 46 Kg5 Rg6+!
Not 46 ... Kxh7?? 47 Kxf6 fxe6 48 Kxe5 and it is White who wins.
47 Kf5 Kxh7 48 exf7 Kg7 0-1
Black wins the king and pawn ending after 49 f8Q+ Kxf8 50 Kxg6 (as in most wars, the
winning side remains as broke as the conquered; the material count may be even, but the
game isn’t as Black is faster in the race) 50 ... e4 51 h4 e3 52 h5 e2 (the undramatic 52 ...
Kg8 also wins) 53 h6 e1Q.
Chapter Three
The Later Years
In this chapter we cover the mature Tal’s games, from 1973 until his death in 1992. With the
exception of the Dvoretsky game, Tal’s play remains highly tactical. So I find it impossible
to explain that someone who played such high-risk chess also owns the two longest non-
losing streaks in chess history. In 1972, Tal surpassed Petrosian’s record non-losing streak,
by going 68 games without loss. And then in the early 1980’s he shattered his own record,
this time playing an astounding 80 games without a loss. Somehow Tal mastered the art of
stepping to the very edge of defeat, without falling off. He continued to assess positions non-
arithmetically, as if the material count was irrelevant.

In the late 1980’s I had a chance to watch Tal in real life at a major Los Angeles
tournament. I first rode up in the elevator with him. Now normally writers tend to say out
loud what others simply think. This case was the rare exception for your writer. I was too
intimidated to say what I thought: ‘Oh my God! Oh my God! You’re Mikhail Tal!!’ (actually
the guy next to me said exactly that, so I didn’t have to).
In the tournament hall, there were many demo boards for the top players, yet everyone –
and I mean virtually 99% of those spectating – crowded Tal’s board with eyes riveted (at
the time I found it incomprehensible that the spectators found my fascinating London System
demo board game a grey, featureless thing by comparison). It was as if all the other IMs and
GMs playing were unreal holograms who didn’t exist. I have never seen such open worship
of a player, ever. Even in the latest stages of his life, combinations kept returning to Tal’s
games like a lover.

Game 38
M.Tal-P.Keres
Tallinn 1973
Ruy Lopez

Keres, like Korchnoi, gave Tal a lot a trouble – but not in this game.
1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 d6 5 0-0
Also played here are 5 c3 and 5 Bxc6+ bxc6 6 d4 f6.
5 ... Bd7
This line is more solid than the strategically suspect 5 ... Bg4!? 6 h3 h5!? (6 ... Bh5 7 c3
Nf6 8 d4 b5 9 Bc2 Be7 10 d5 Nb8 11 a4 left White with a pleasant strategic advantage in
G.Klompus-S.Arzumanyan, USSR 1967). We all fall sway to extremes of stylistic bias, from
time to time. When we play a line as sharp as this one, which has fallen into theoretical
disrepute, our opening variation becomes inextricably entangled with our loyalty to logic.
 After 7 d4 b5 8 Bb3 Nxd4 (or 8 ... Bxf3 9 Qxf3 Nf6 10 c3 exd4, A.Gurevich-
E.Rompteau, correspondence 1966, when I like White’s chances after 11 Qe2 dxc3 12 Rd1
cxb2 13 Bxb2 with a kind of super-Danish Gambit for the two pawn investment; Black’s
king will never be safe and White leads big time in development) 9 hxg4 hxg4 (or 9 ... Nxb3
10 axb3 hxg4 11 Ng5 Qd7 12 Qd3 c6 13 c4 b4 14 c5 f6 15 Nd2 fxg5 16 Nc4 Qe6 17 Rd1
Qh6 18 Kf1 Qh1+ 19 Ke2 Qxg2 20 cxd6 Qf3+ 21 Qxf3 gxf3+ 22 Kxf3 when Black’s ragged
pawn structure and lag in development mean a busted ending, V.Krishnan-T.Taylor, South
California State Championship 2012; this win allowed my student to go on to win the State
Championship) 10 Ng5 Nh6 11 Bd5 c6 12 c3 cxd5 13 cxd4 Be7 14 Qd2 dxe4 15 dxe5 d5 16
Nc3 e3 17 Qxd5 Bxg5 18 Qc6+ Kf8 19 Bxe3 Bxe3 20 fxe3 and Black was fighting for
survival, N.De Firmian-J.Timman, Reykjavik 2000.
Returning to 5 ... Bd7:
6 c3 Nge7 7 d4 Ng6 8 Re1
8 Be3 is also played here.
8 ... Be7 9 Nbd2 h6
 Question: Why ... h6? Black already covers g5.

Answer: Black may be planning to ‘goodify’ his bad bishop with a future ... Bg5, in
which case he has the ... hxg5 option.
10 Nf1 Bg5
And here it is already. White isn’t so tempted to chop it, since that would open the h-file
for Black’s rook.
11 Be3
11 Ne3 is met with 11 ... Bxe3 12 Bxe3 when Black rids himself of his bad bishop and
can be satisfied with the opening’s outcome, M.Sagafos-T.Gareev, Cappelle la Grande
2007.
11 ... Bxe3 12 Nxe3 0-0 13 Bc2
The bishop backs up e4, since it isn’t of much use on a4.
13 ... Re8 14 Qd2 Rc8
The idea is to keep c7 covered in case of a future Nd5. Also logical is 14 ... Nh4 15
Nxh4 Qxh4 16 Nd5 Rac8. The swap on h4 helped Black, who was the more cramped side,
I.Kopylov-B.Kovernikov, correspondence 2005.
15 g3
Keeping Black’s knight out of f4.
15 ... Qf6
Attacking f3.
16 Nd5!?
 Hey, I said: “Attacking f3”! Tal’s last move is the spark which is the precursor to a
future fire which scorches the land. He intends to respond to ... Qxf3 with Bd1, trapping
Black’s queen.
On 16 Kg2 Tal probably was worried about the line 16 ... Bh3+!? which the comps like
for White’s side, after 17 Kxh3 Qxf3 18 Nd5 Qh5+ 19 Kg2 Qg5. This was given an exclam
by Keres, who assessed this position as even, while Komodo gives White an edge after 20
Qxg5 hxg5 21 Rad1 Nf8 22 Ne3.
16 ... Qxf3!
Tal’s entire premise must now be rethought and re-evaluated. Keres cleverly self-traps
his queen, having seen a resource.
17 Bd1
 Exercise (combination alert): It’s easy to pick off a stray soldier – but not so
easy when she is part of a military formation. It looks like Keres is about to lose
a miniature. His queen is trapped. Or is it? Black to play and not lose his queen:

Answer: Knight fork/clearance. Black’s queen is far from trapped.

17 ... Nh4!
Black’s forces are drawn to f3, as if to a loadstone. Any retreat runs counter to our sense
of honour, which in this case is even more apt, since Black has no viable retreat for his
queen! Even near the end of his life, Keres remained tactically astute, having foreseen this
trick which extricates his queen.
18 gxh4
Not 18 Bxf3?? Nxf3+ 19 Kg2 Nxd2 and White finds himself down two pieces for zero
compensation.
18 ... Qh3

Exercise (combination alert): For most of us, when we land in such weary looking
positions, our imaginations go inoperative. Not so with Tal. Now what? It looks as if
Tal’s position is collapsing. After all, his kingside structure is decimated, seemingly
without compensation. However Tal found a miraculous saving idea. What is it?

Answer: Line opening/knight fork.

19 Nf6+!!
The criminal knight melts discretely into the crowd. Amazingly. Tal is the only player I
know who found hidden ways of seizing the initiative, while fettered and gagged. I don’t
know about you, but when I play over Tal’s games, I get that disoriented, out-of-place
feeling, as if I’m at a party where I don’t know anyone.

Exercise (critical decision): No time to rest. Should Black chop

the knight, or should he give up the exchange with ... Kh8?

19 ... Kh8?
This attempt to quell the kingside commotion is an overreaction. Keres, undoubtedly
startled by Tal’s last move, unnecessarily hands over the exchange and now stands worse.
He should accept the challenge with:
Answer: 19 ... gxf6! (it’s not so easy to find your plan when there is an absence of
identifying characteristics within the position; here we only see a maze of seemingly random
lines) 20 Qxh6 exd4 21 Kh1 (threatening a nasty rook check on g1) 21 ... Ne5! (any
defensive formation is only as strong as its weakest square, and somehow Black manages to
hold the game by reinforcing g4; alternatively, 21 ... Bf5?? is met with 22 Rg1+ Bg6 23
Rxg6+ fxg6 24 Qxg6+ Kf8 25 Qxf6+ Kg8 26 Bh5 when there is no defence to the coming
Rg1+, while after 21 ... Rxe4 22 Rg1+ Bg4 23 Bxg4 Rxg4 24 Rxg4+ Qxg4 25 Rg1 Qxg1+
26 Kxg1 dxc3 27 bxc3 Re8 28 Qxf6 Re6 Black still has chances to erect a fortress draw,
despite Komodo’s completely winning assessment for White) 22 Rg1+ Bg4 (22 ... Ng4? is
met with 23 Bxg4 Bxg4 24 Rg3 Qxg3 25 hxg3 dxc3 26 bxc3 Rxe4 27 Qxf6 and White has
reasonable chances to convert) 23 Rg3 (White should probably avoid 23 cxd4 Nf3 24 Rg2
Rxe4) 23 ... Qf1+ 24 Rg1 Qh3 with a draw by repetition.
20 Nxe8 Rxe8 21 h5 Bg4
Keres goes after the h5 straggler, after which he gets a pawn for the exchange.
22 Qe3
Also promising was 22 f4!.
22 ... Qxh5 23 Kh1
 Tal, hoping to turn a problem into a virtue, plans to attack down the opened g-file.
23 ... Bxd1 24 Raxd1 Qh4
I would try 24 ... g5!? which clamps down on any f4 break, while discouraging White
down the g-file.
25 Qf3 Kg8
Covering f7, but in doing so, placing his king on a more vulnerable position on the g-
file. Black can also try to randomize the position with a line like 25 ... Rf8 26 Rg1 exd4 27
cxd4 f5!?.
26 Re3 Rf8 27 Qg3 Qe7?!
Black’s queen remains unassimilated to the position’s new reality. More accurate was
27 ... Qf6! (pressuring d4, while suppressing White’s f4 break) 28 d5 Ne7, in which case
White can play for a c4-c5 queenside pawn break.
28 f4! exf4 29 Qxf4

Not only did Tal fix his structure, but he also opened the f-file for his rooks.
29 ... Re8 30 Rg1
Threatening a cheapo on h6.
30 ... Kh7 31 Rg4!
This rook laterally covers e4, which frees the e3-rook to operate along the third rank.
31 ... Nd8
The long alienated knight wishes to return to society, hoping to make good use of itself
for kingside defence.
32 e5!
Tal immediately takes advantage of the fact that Black’s knight no longer controls e5.
32 ... d5
This move stabilizes the centre, which favours White, since he is the one attacking.
Keres didn’t like 32 ... Ne6 33 Qe4+ when White’s queen picked off b7 next.
 33 Rh3!
Threat: Rxg7+, followed by Qxh6+.
33 ... Qf8
33 ... Ne6?? covering g7, fails to cover g7!. White has 34 Rxg7+! with mate in two
moves, no matter which way Black recaptures.
34 Rf3
Tal prevents ... Ne6, by tying the knight to defence of f7.
34 ... Kg8??

I strongly suspect this is simply a ChessBase input error, and Keres actually played 34
... Kh8 in which case I owe both Keres and Tal apologies for giving them both unjust double
question marks!
35 Qf5??

Question: What the hell!? How did two world-class

GMs miss that White’s queen can take h6?

Answer: I agree that the players’ last two moves are as mysterious as fortune cookie
aphorisms. As I said before, either this is a typo on the ChessBase game score (the most
likely answer) or, the players were in terrible time pressure, producing a mutual freak-out of
double question marks. If this was the actual game, then both sides missed 35 Qxh6.
35 ... Qe7?!
35 ... Ne6 was necessary.
36 b4?!
Tal clamps down on ... c5 counterplay, when he had 36 Rh3! Qe6 37 Qh5 (re-threatening
h6) 37 ... Kh7 38 Qg5! Qg6 (38 ... Rg8? is met with 39 Rxh6+! Qxh6 40 Rh4, winning) 39
Qh4 Qb6 40 Qg3 Ne6 41 Qf2 Nd8 42 Qc2+ Kh8 43 Qg2 Ne6 44 b3, and if 44 ... c6?? 45
Qd2! threatening Rxh6+. There is no defence, since 45 ... Kh7 46 Rxg7+! forces mate.
36 ... Rf8 37 Qh5
Once again menacing h6.
37 ... Ne6
37 ... Kh7 38 Rf6 Ne6 transposes to the game’s continuation.
38 Rf6!?
Tal, of course, picks the flashy route to the mundane 38 Qxh6 which wins easily.
38 ... Kh7 39 Qf5+ Kh8 40 Qh5 Kh7 41 Rh4!
Threat: Rxh6+.
41 ... Kg8 42 Qg4!
Tal isn’t satisfied with a mere winning ending after 42 Rxh6 gxh6 43 Qxh6 Qxh4 44
Qxh4.
42 ... Ng5
Or 42 ... Kh7 43 Qh3! and there is nothing to be done about the coming Rfxh6+.

Exercise (combination alert): It appears as if Black’s once atrophied defence flutters

and threatens to revive. This is not the reality. Find Tal’s attacking continuation:

Answer: Annihilation of defensive barrier.

43 Rhxh6!
“Know this royal brat: Death doesn’t discriminate between the old and the young,” says
the approaching rook to the h6-pawn. Tal spares no expense to finance his attack, and now
the defence crumples like an article of clothing, tossed into the laundry basket.
43 ... Ne4
If 43 ... gxh6 44 h4 Kg7 45 hxg5 h5 46 Qf5! and there is no defence to the coming g6.
44 Qh5!
 Black can take his pick over which rook he wants. Unfortunately, whichever one he
takes, he loses.
44 ... gxh6 45 Rxh6 1-0
“Many in the past have tried to kill me, yet none have succeeded,” boasts Black’s king,
who never did understand the law of averages. Black is mated.

Game 39
M.Tal-M.Stean
Hastings 1973/74
Sicilian Defence

1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 a6 6 Bg5 e6 7 f4 Nbd7 8 Qf3 Qc7 9 0-0-0
The players enter the theoretically dense main line of the 6 Bg5 Najdorf.
9 ... b5
9 ... Be7 is more commonly played here. The main line runs 10 g4 b5 11 Bxf6 Nxf6 12
g5 Nd7 13 f5!?.
10 Bxb5!?

“How can an action which gives one such pleasure be a sin?” the bishop asks himself.
The move choices we make are simply reflections of our inner nature. Tal unleashes a wave
of confusion throughout the position, and for him, that was a good enough reason to
sacrifice. Tal’s last move is made with the philosophy: rarely is a monumental goal easily
accomplished. It is human nature to respond with aggression when we feel the sting of
wounded pride. But I ask: how did Stean’s play give Tal offence this early in the game? The
answer is that Tal was simply a man who considered anything but the most direct path a
weakness in himself.
 Question: Is this one sound?

Answer: This is actually a book line which may be sound, when you play through it with
the comps. Personally, I don’t like to play into lines like this from either side. Why? You can
enter a position, played hundreds of times, with thousands of players believing in it. Then
one day, someone finds a comp-generated new move/refutation, and suddenly, nobody wants
to play it again. I challenge my comp to refute the London System! 10 Bd3 is an obviously
calmer response.
10 ... axb5 11 Ndxb5
11 e5 Bb7 12 Ndxb5 Qb8 transposes to the game continuation.
11 ... Qb8 12 e5 Bb7
This move is necessary, since 12 ... dxe5?? walks into White’s cheapo 13 Qxa8! with a
winning position.
13 Qe2 dxe5 14 Qc4
Threat: Nc7+.
14 ... Bc5
GM Tony Kosten disliked this move and felt it lost by force. I disagree. This is the
natural move, which blocks White’s Nc7+. The move does, however, contain practical
difficulties for Black, who must thread the needle with only-moves to secure equality.
Black is better off playing 14 ... Be7! 15 Nc7+ Kf8 16 Rxd7 Nxd7 17 Rd1 Bxg5 18 fxg5
(White is down a full rook, yet Komodo evaluates at ‘0.00’!) 18 ... Ke7 19 Qb4+ Kd8 20
Nxa8 Qxa8 21 Nb5 Bd5 22 c4 Bc6 23 Nd6 Rf8 24 Nf5 Rg8 25 Qe7+ Kc7 ½-½,
F.Kunzelmann-J.Bokar, correspondence 2009. The game ends in perpetual check after 26
Qd6+ Kd8 27 Qe7+.
15 Bxf6! gxf6
The d7-knight must cover the c5-bishop.
16 Rxd7!
 Business is brisk for the lucky coffin makers and the journey to the cemetery becomes
more and more familiar for Tal’s pieces, who are dying off at alarming rates. As we all
understand by now, income disparity never plagued Tal’s mind during the game. Somehow
he treated his pieces like Monopoly money.

Question: Isn’t Tal pressing his attack with mounting recklessness?

Answer: On the contrary, it’s too late for second thoughts. His move – a not-so-subtle
attempt to ferret out Black’s king from his hole – is the only way to continue White’s attack.
The comps still evaluate at ‘0.00’.
16 ... Be3+
This zwischenzug saves the bishop.
17 Kb1!

Question: Why didn’t White play 17 Rd2 to get Black’s bishop for the rook?

Answer: The line works out badly for White after 17 ... 0-0! 18 Qe2 exf4 19 Kb1 Bxd2
20 Qxd2 Rd8. Black remains up on material, with a relatively safe king, P.Clerides-
J.Penttinen, Limassol 2000.
17 ... Kxd7 18 Rd1+
 Exercise (critical decision): For now, Black’s overall goal (to convert his extra
material and win) remains subordinate to the more urgent concern of survival.
Black has a choice of 18 ... Ke7, 18 ... Ke8, or blocking with 18 ... Bd4. Two of the
choices favour White, while only one saves Black. Which one would you play?

18 ... Bd4?
Once your plan loses its dominating principle (in this case, for Black, it is merely
survival), either paralysis or chaos sets in. The bishop had long prayed for a sign, and to his
chagrin, soon gets one. This is the worst of Black’s choices.
Instead, 18 ... Ke8?! works out in White’s favour after 19 Nc7+ Kf8 20 fxe5! Ba6 21
Qc6! (21 Nxa6 Qxe5 is okay for Black) 21 ... Ra7 22 Rd7! Rxc7 23 Rxc7 Kg7 24 exf6+
Kh6 25 a3 Rf8 26 g4. Advantage White: Black’s king remains under fire and White still has
his three connected passed queenside pawns for the piece.
Answer: Correct is 18 ... Ke7! which should save Black after 19 Qb4+ Ke8 20 Nd6+
Kf8 21 Nf5+ Kg8 22 Nxe3 Bc6 23 Qc5 exf4 24 Ng4 f5 25 Qxc6 fxg4 26 Ne4. White has
enough counterplay to save the game, but no more: 26 ... Ra5 27 Nd6 Qa8 28 Nb7 (threat:
Rd8+) 28 ... Rd5 29 Rxd5 exd5 30 Qf6 Qxb7 31 Qg5+ Kf8 32 Qd8+ with perpetual check.
19 fxe5 fxe5
19 ... Qxe5 fails to save Black: 20 Rxd4+ Bd5 21 a3 Rhc8 22 Qb4 Ra6 23 Rd1 h5 24
Nxd5 exd5 25 Re1 Ra4 26 Rxe5 Rxb4 27 Rxd5+ Kc6 28 Rd6+ Kxb5 29 axb4 and Black
can resign.
20 Nxd4 exd4 21 Qxd4+ Ke7
21 ... Bd5 eliminates White’s knight, which fails to help Black’s king after 22 Nxd5 exd5
23 Qxd5+ Ke8 24 Re1+ Kf8 25 Qc5+ Kg8 26 Qg5+ Kf8 27 Qh6+ Kg8 and the rook lift 28
Re3 is decisive.
 22 Qc5+!
More accurate than 22 Qd7+ Kf8 23 Rf1 Qe8 24 Qxb7, which is still very much in
White’s favour, Y.Krupenski-M.Kanep, Tallinn 2000.
22 ... Kf6
22 ... Ke8 allows White’s knight to enter with force after 23 Nb5. Now if 23 ... Bd5 (or
23 ... f5 24 Nd6+ Ke7 25 Nc8+! Kf7 26 Qe7+ Kg6 27 Qxe6+ Kg5 28 Nd6 and Black’s extra
rook won’t help him at all) 24 Nc7+ Kd7 25 Nxd5! exd5 26 Qxd5+ Ke8 27 Qc6+ Kf8 28
Qh6+ Ke8 29 Re1+ Kd7 30 Qh3+ Kc6 31 Qc3+ Kb6 32 Qf6+ Ka5 (32 ... Ka7 33 Re7+ is
of no help either) 33 Re5+ Qxe5 34 Qxe5+ and White’s three queenside passed pawns win
easily.
23 Rf1+
23 Qd4+? allows Black to escape after 23 ... Kg6 24 Qg4+ Kf6 25 Rf1+ Ke7. White
must agree to a perpetual check, S.Lacroix-T.Libersan, Montreal 2005.
23 ... Kg6 24 Qe7!

Tal’s awareness hovers over f6 and f7 with watchful intensity.

24 ... f5?
With time pressure looming, there must be a lowering of our standards, or we risk
flagging. Now the defence crumples like an unstrung puppet. 24 ... Qe5! is Black’s last
chance to offer resistance: 25 Qxf7+ Kg5 26 g4! Kh4! (Black’s king hopes to find his footing
on still unstable ground) 27 Qe7+ Qg5 28 Qxb7 Raf8 29 Rg1 Qe3 (it feels as if a light,
vagrant breeze of counterplay begins to emerge for Black, as he plays on White’s weakened
back rank; it’s still not enough in the end) 30 Qe7+ Kh3 31 Ne2! Rf3 32 Qg7 Kxh2 (threat:
... Qxg1+!) 33 Re1! Rhf8 (the would-be back rank shot 33 ... Qxe2?? fails miserably, since
34 Rxe2+ comes with check) 34 Qxh7+ Rh3 35 Qd7 Rhf3 36 Qd1 and White still has some
work to do before he scores the win.
25 Qxe6+ Kg7 26 Qe7+
 The comp found a quicker path with 26 Qd7+! Kh6 (26 ... Kg8 is met with 27 Rxf5) 27
Rxf5 Qe8 28 Rf6+ Qg6 29 Rxg6+ hxg6 30 Qxb7 which wins.
26 ... Kg6
White has two ways to finish here. Tal utilizes his kingside pawns as attackers in the
mating net.
27 h4!
27 g4! also wins after 27 ... f4 28 h4 (threat: h4+ and Qf6 mate) 28 ... Qf8 29 Qg5+ Kf7
30 Qxf4+ Kg8 31 Qc4+ Kg7 32 Qc7+ Kg8, and simplest is 33 Rxf8+ Rxf8 34 a4 with an
easy win.
27 ... Ra5
Black desperately covers f5, but his position is past saving. Instead, 27 ... Rf8 is met
with 28 g4! Qg3 29 gxf5+ Kh6 30 Ne2 Qg4 (or 30 ... Rae8 31 Qxf8+ Rxf8 32 Nxg3 and
White wins) 31 Rg1 Qxg1+ 32 Nxg1 which wins, since 32 ... Rxf5?? is met with 33 Qe6+.

Exercise (calculation/combination alert): It becomes painfully clear that

Black’s king will not be granted safe passage though hostile territory.
Tal worked out a forced mate in seven moves. How does it begin?

Answer: Attraction/mating net.

28 h5+! Kxh5
28 ... Kh6 29 Qf6+ allows White to deliver mate one move faster.
29 Qf7+ Kh4
Or 29 ... Kh6 30 Qf6+ Kh5 31 Rh1+ Kg4 32 Qh4 mate.
30 Qf6+ Kg3 31 Qg5+ Kh2 32 Qh4+ Kxg2 33 Rf2+ Kg1
The defence-starved king realizes in his dying thought: wealth is unable to relieve
hunger if there is no food to buy.
 34 Ne2 mate (1-0)
“Oh no, please, no, it was nothing,” murmurs the knight, as she bashfully poses for the
clicking cameras, over the black king’s body. Three against one doesn’t strike me as a fair
fight.

Game 40
B.Spassky-M.Tal
Tallinn 1973
Nimzo-Indian Defence

1 d4 Nf6 2 c4 e6 3 Nc3 Bb4 4 Bg5

Spassky’s favourite Leningrad line of the Nimzo-Indian.
4 ... h6 5 Bh4 c5 6 d5 b5!?

Question: Is this some kind of a Blumenfeld Gambit of sorts?

Answer: Tal’s move is more goading than a direct slap across the cheek. I guess you can
call it the Nimzo-Blumenfeld Indian. It seems to be a hybrid. Black is willing to offer a
pawn to seize central control. Normal in this position is 6 ... d6, or 6 ... Bxc3+ 7 bxc3 d6 8
e3 e5.
7 dxe6
There is also the tricky version 7 e4!? g5 8 Bg3 Nxe4 9 Be5 0-0 10 Qh5 d6 11 Bd3
Nxc3 12 Qxh6 Ne4+ 13 Kf1 dxe5 14 Bxe4 f5 15 Qg6+ Kh8 16 Qh6+ and the game ends in
perpetual check, A.Yermolinsky-A.Shabalov, Parsippany 1996.
7 ... fxe6 8 cxb5
White can also decline with 8 e4 0-0 9 e5 Qa5 10 Nge2 (but not 10 exf6?? Bxc3+ 11
bxc3 Qxc3+ 12 Ke2 Qxc4+ and White’s h4-bishop falls) 10 ... Ne4 11 Qc2 Bb7 12 f3 Nxc3
13 bxc3 Ba3 14 cxb5 a6! when White’s underdeveloped position appears to be in some
danger, E.Bareev-B.Gelfand, Moscow 1990.
8 ... d5 9 e3 0-0

Question: Do you think Black has full compensation for the pawn?

Answer: More than enough. For only one pawn, Black gets:
1. A development lead in an open position.
2. Central dominance.
3. An open f-file.

Question: Can Black try to disrupt with 9 ... d4?

Answer: This confrontation is premature. For example, 10 a3 Bxc3+ (or 10 ... Ba5 11
Bxf6 Qxf6 12 Qh5+ Qf7 13 Qxc5 and White wins material) 11 bxc3 Qa5 12 Rc1 dxe3 13
Qf3! 0-0 14 Qxa8 Qxa3 15 Ne2 Nbd7 16 Bxf6 exf2+ 17 Kxf2 Nxf6 18 Kg1. Black doesn’t
have enough of an attack to justify handing over a full rook.
10 Nf3
Spassky also tried 10 Bd3 d4 11 exd4 cxd4 12 a3 Ba5 13 b4 dxc3 14 bxa5 Bb7 15 Nf3
Qxa5 16 0-0 Nbd7 17 Qe2 Bxf3 18 Qxf3 Ne5 19 Qe2 Nxd3 20 Qxd3 Nd5. If anyone stands
better, it’s Black, since his knight is anchored powerfully on d5, B.Spassky-W.Unzicker,
Bath 1973.
10 ... Qa5!
A new move at the time, improving upon 10 ... Nbd7 11 Bd3 e5!? 12 Bf5! e4? 13 Be6+
Kh8 14 Bxd5 Nxd5 15 Bxd8 Nxc3 16 bxc3 Bxc3+ 17 Nd2 Rxd8 18 Rc1 Ba5 19 0-0 Ne5 20
Qh5 Bxd2 21 Qxe5 Bxc1 22 Rxc1 Bb7 23 h4 1-0, B.Spassky-H.Liebert, Sochi 1967.
11 Bxf6?!
Now the precarious balance of power begins to swing in Black’s favour. 11 Qc2! Ne4
was K.Sasikiran-N.Short, British Championship, Torquay 1998. Now, instead of playing the
rook to c1, Komodo found the shocking and unplayed trick 12 Bd3!! Nxc3 13 0-0!. Believe
it or not, White, down a full piece (for now) stands better, no matter how Black plays. For
example, 13 ... Nxb5 14 a3 regains the piece with an edge.
11 ... Rxf6

12 Qd2

Question: Can White try 12 Qc1!? intending a3 next?

Answer: It looks rather slow. For example, 12 ... a6! and if White accepts with 13
bxa6? then 13 ... Bxa6 14 a3 Bxf1 15 Kxf1 Nd7 16 Ne2 c4 17 Qc2 Bd6 looks highly
unpleasant for White, who has a problem unravelling his h-rook.
12 ... a6!
A thematic move in the position, seeking to increase his development lead.
13 bxa6?!
When we grab material when behind in development, we offer nothing, yet expect
something. As we all understand, the world doesn’t work that way. Spassky’s patience
begins to fray and now is in danger of a route. It’s high time to go into emergency concession
mode with 13 Be2 axb5 14 0-0, returning the pawn with a slightly inferior position.
13 ... Nc6 14 Be2
Kasparov suggested 14 Qc1, which may not help White after 14 ... d4 15 a3 Bxc3+ 16
bxc3 Bxa6 17 Bxa6 Qxa6. I still hate White’s position and I’m not sure he is any better off
here than with the game’s continuation.
 Exercise (planning): How did Tal seize the initiative?

Answer:
Step 1: Play upon the principle: Open the position and create confrontation when
leading in development.
14 ... d4!!
This move requires immense powers of calculation to pull off.
15 exd4
Step 2: Sacrifice the exchange by undermining a key defender of d4.
15 ... Rxf3! 16 Bxf3 cxd4 17 0-0!
It isn’t easy to make sense of the disorienting vista. Spassky makes the best of what he
has:
a) 17 Rc1? Bxa6!! 18 Bxc6 (or 18 Be2 Rd8 19 0-0 dxc3 20 Qe3 Nd4! 21 Bxa6 Qxa6 22
bxc3 Ne2+ 23 Kh1 Nxc1 24 Rxc1 with an extra piece for Black, plus initiative after 24 ...
Rd3 25 Qe4 Rxc3) 18 ... Rd8! leaves White helpless, since 19 Rd1 is met with the crushing
19 ... Qe5+.
b) 17 Bxc6?? loses instantly to 17 ... dxc3 when White lacks a reasonable response.
17 ... dxc3 18 bxc3
Also highly unpleasant was 18 Qd1 cxb2 19 Rb1 Rxa6 20 Rxb2 Bc3.
18 ... Bxc3 19 Qd6 Rxa6
In preparation for a coming pin.
20 Bxc6
20 Rad1? is met with 20 ... Nd4 21 Qg3 e5 with a dominating position for Black.
 Exercise (critical decision/combination alert): It looks like Black has two
promising moves to disconnect White’s queen from his bishop. Black can play
20 ... Bb4, or 20 ... Be5. It looks like both moves favour Black. This isn’t the case.
There are two trails, with one of them false. Which one would you play?

Answer: 20 ... Bb4!

20 ... Be5?? fails miserably to 21 Qe7! and the problem is 21 ... Rxc6?? is harshly met
with 22 Qe8+ Kh7 23 Qxc6 when White is up two exchanges.
21 Qb8 Rxc6
So Tal emerged with two bishops for Spassky’s rook and pawn – a clear plus for Black.
However, matters are not so simple for Black since he must still untangle and complete
development.
22 Rac1 Bc5
Black is forced to self-pin. Naturally not 22 ... Rxc1?? 23 Rxc1 when the pinned c8-
bishop falls.
23 Rc2?!
Now Spassky is unable to restore order. If you are assigned a clandestine mission, then
it’s unwise to call your enemy ahead of time to let him know you are coming. Aggression
from a position of inferiority is a sign that a player wants to distract from his own strategic
vices. Spassky hopes to build on the c-file pin, underestimating Tal’s coming tactical
resources.
It’s not all that easy for Black to unravel after the counterintuitive 23 Rfd1!!. Shockingly,
Black is unable to unravel without allowing White serious counterplay: 23 ... Kh7 (after 23
... Bxf2+ 24 Kh1! Bc5 25 Rc2 Bd6 26 Rxc6 Bxb8 27 Rxc8+ Kh7 28 Rxb8 Qxa2 White
stands no worse) 24 Kh1 Qa6 (24 ... Ba6 is met with the unpleasant 25 Rd8) 25 Rd8 Bd6 26
Qb1+ g6 27 h4! e5 28 Rxc6 Qxc6 29 Qb3, threatening mate on g8. Black has nothing better
than to take a draw with 29 ... Qc1+ 30 Kh2 Qf4+ 31 Kg1 Qc1+.
23 ... Qa4!
Intending to slip over to the kingside to generate threats against White’s king.
24 Qb3
Spassky avoids the trap 24 Rfc1?? Bxf2+!, winning.
24 ... Qf4!

How long can Tal continue to avert disaster along the c-file? A gradual accretion of
details offers us a better overall picture of Black’s correct plan: Tal intends to distract
Spassky’s c-file pin project with direct threats to White’s king.
25 Qg3
Spassky, exhausted from walking through a carnival of perils, has had quite enough of
Tal’s attempted cheapos and desperately tries to remove queens from the board. 25 Rfc1??
is once again met with the dirty trick 25 ... Bxf2+! 26 Kh1 (26 Kf1 Bg3+ 27 Ke2 Ba6+
forces mate) 26 ... Qxc1+ with mate in two moves.
25 ... Qf5 26 Rfc1 Bb7!
Both black bishops take aim at White’s king. From the position emerges a tumble of
unanswered questions, with each side toiling feverishly on concurrent and opposite
preoccupations:
1. Can White make anything of his c-file pin?
2. Can Black’s tied up forces still continue to generate threats on White’s king?
27 Qf3
White’s queen continues her erratic dance. 27 Qb8+? Kh7 28 Qxb7?? is once again met
with 28 ... Bxf2+! winning.
27 ... Qg5
Black’s pinned dark-squared bishop makes a mockery of White’s rooks, as long as he
goes unarrested and unhanged.
28 Qb3
Spassky alertly dodges yet another cheapo. If 28 Qg3?? Bxf2+! 29 Qxf2 Qxc1+! wins.
28 ... Rc7!
Triple purpose:
1. Threatening mate on g2.
2. Covering the b7-bishop.
3. Continuing to cover the c5-bishop.
Tal, probably the most mate-obsessed of all the world champions, was born with a rare
birth defect: an attack-addled brain. 28 ... Bxf2+? fails this time: 29 Kxf2 Rxc2+ 30 Rxc2
Qxg2+ 31 Ke3 when Black has no more than a draw.
29 g3
29 Qxe6+?? Rf7 30 g3 Bxf2+! overloads White’s c2-rook.

Exercise (combination alert): Spassky, who up to this point, managed to

turn back Tal’s violent challenge, soon discovered that it is impossible to snuff
out an approaching forest fire through force of will alone. How did Tal continue?

Answer: Attraction/annihilation of defensive barrier. Tal unveils his long-planned

subterfuge, as his long pinned fugitive bishop evades justice.
29 ... Bxf2+!
The wizard raises his arms, and a whirl of dead leaves begins to spin around his aura of
power.
30 Kxf2 Qf6+
Slightly more accurate was 30 ... Qf5+! 31 Kg1 Qe4! which leaves White helpless. Tal
finds this theme much later in the game.
 31 Ke1
The king quivers, like pudding with a face:
a) 31 Kg1 Qd4+ 32 Kf1 Rf7+ 33 Ke1 Qe4+ 34 Re2 Qh1+ 35 Kd2 Rd7+ 36 Kc2 Rc7+
wins.
b) 31 Ke2 Qe5+ 32 Qe3 Ba6+ 33 Kf2 Rf7+ picks off White’s queen.
31 ... Qe5+ 32 Kf1
For the remainder of the game, White’s king bounces around like a popped kernel of
corn in the microwavable bag.
32 ... Ba6+
Not a serious slip. Tal missed 32 ... Qf5+! 33 Kg1 (33 Ke2 Qe4+ 34 Kd1 Rd7+ 35 Rd2
Qh1+ 36 Kc2 Rc7+ wins) 33 ... Qe4!, winning on the spot.
33 Kg1 Qd4+ 34 Kg2 Qe4+?!
“Our fates have always been intertwined, my love,” whispers the queen, into the king’s
ear. Tal should play 34 ... Bb7+! 35 Kh3 Qe5! (threatening mate on h4) 36 g4 Qe4 37 a4 (or
37 Rxc7 Qg2+ 38 Kh4 Qxh2+ 39 Qh3 g5+ 40 Kh5 Qxh3+ 41 Kg6 Be4+ 42 Kf6 Qf3+ 43
Kxe6 Qxg4+ and Black wins) 37 ... g5! 38 a5 h5! 39 Rg1 hxg4+ 40 Rxg4 Rh7+ 41 Kg3
Qe1+ 42 Rf2 Qg1+ 43 Rg2 Qxg2 mate.
35 Kg1?
Instead:
a) 35 Kh3? Rxc2 36 Qxc2 Bf1+! overloads White’s rook.
b) 35 Kf2! is White’s best practical chance. After 35 ... Rf7+ 36 Kg1 Black must find 36
... Bd3! (threatening a big check on e3) 37 h4 Rf3 38 Kh2 Bxc2 39 Qxc2 Qe3 (threatening
both ... Rf2+ and also ... Rxg3) 40 Qc8+ Rf8 41 Qc6 Rf2+ 42 Kh3 Qe5 43 Rc3 Qf5+ 44 g4
Qb1 45 Rg3 Qxa2 46 Qh1 h5! when White’s king won’t survive.
35 ... Bb7!

Now Tal sees the correct idea.

 36 h4 Qh1+ 37 Kf2 Rf7+ 38 Ke2 Qe4+! 0-1
“I see very clearly that in the past I have been far too kind,” says the queen. “Kind?”
thinks White’s king, who wouldn’t describe his sister in that way. If the following lines look
familiar, it’s because we looked at similar themes in the earlier notes. Even Komodo was
exhausted from all the calculation which took place in this game:
a) 39 Qe3 Ba6+ 40 Kd2 Rd7+ wins White’s queen.
b) 39 Kd1 Rd7+ 40 Rd2 Qh1+ 41 Kc2 and simplest is 41 ... Rc7+.

Game 41
M.Tal-T.Petrosian
USSR 1974
Pirc Defence

The ideological space of interpretation is as vast as is possible, in a game between Tal and
Petrosian.

1 e4 g6 2 d4 Bg7 3 Nc3 d6 4 Nf3

Our stylistic biases normally find ready outlets in the openings we choose to play, so
this move, which exerts a calming influence on the position, comes as an odd choice for Tal.
It strikes me as surprising that he would play one of the quietest systems White has at his
disposal (Tal, as we all know, considered quiet positions akin to captivity), rather than risk
a more theoretically intense line like 4 f4, the violent Austrian Attack.
4 ... Nf6
A Modern Defence transposes to a Pirc.
5 Be2 0-0 6 0-0 Nc6

Petrosian goads Tal early in the game. Playable alternatives are found in 6 ... c6, 6 ...
Bg4, 6 ... Nbd7, 6 ... c5, and 6 ... a6.
7 d5
White’s most principled move. If your opponent provokes you, then it is only just that he
should be punished! White can also refrain with 7 h3 e5 8 dxe5 dxe5, E.Rozentalis-
J.Ehlvest, Koszalin 1998. White may claim a microbe of an edge after 9 Bc4.

Question: Why not gain a tempo here with 8 d5?

Answer: After 8 ... Ne7 I like Black’s game since White essentially gets an inferior
King’s Indian Defence, since his c-pawn sits on c2, rather than on c4. Black’s natural ... f5
break comes a lot quicker than White’s painfully slow c4-c5 push.
7 ... Nb8
Instead, 7 ... Ne5 8 Nxe5 dxe5 9 Bg5 looks a shade more comfortable for White,
G.Milos-R.Del Bosco, Sao Paulo 2000.
8 Re1
Tal plays for an eventual e5 break. After 8 a4 Bg4 9 h3 Bxf3 10 Bxf3 (possession of the
bishop-pair is normally our inviolable birthright in this line) 10 ... c6 11 a5 a6 12 Be3 Nbd7
13 Bd4 Re8 14 Re1 Qc7 15 Na4 White’s bishop-pair and extra space gave him the edge,
G.Kasparov-N.Short, Saint Louis (blitz) 2015.
8 ... e5
I would have expected 8 ... c6 to keep Black’s structure fluid.
9 dxe6!?
This move helps Black’s development, but as we all know, Tal favoured open positions.
I would be more inclined toward 9 Bg5 h6 10 Bh4 g5 11 Bg3 Nh5 12 Nd2 Nf4 13 Nc4 f5,
M.Brooks-J.Veal, Stillwater 2001. I prefer White after 14 Bf1 Nxg2 15 Bxg2 f4 16 h3 fxg3
17 fxg3 since he may be able to make use of the f5-square later on.
9 ... Bxe6
Petrosian correctly avoids 9 ... fxe6, N.Gurieli-M.Meyer, New Delhi 2000. Now White
can favourably play the disruptive 10 e5!.
10 Bf4 h6?!
 Question: What is the point of this move?

Answer: I’m not really sure. Perhaps Petrosian wanted to prevent Ng5, or perhaps Qd2
and Bh6. In any case, the loss of time isn’t justified when his queenside remains
undeveloped. The unravelling of an article of clothing begins imperceptibly, through the
fraying of a lone thread. Petrosian allows himself to get preoccupied in an obvious waste of
time, made that much more dangerous in a semi-open position. When we are not sure of the
correct plan, then the place we begin is to eliminate the discards. Black looks fine after 10
... Nc6, E.Geller-G.Kuzmin, Moscow 1974.
11 Nd4 Bd7
Dire need adds savour to an otherwise mundane project. More wasted time, but I don’t
see a preferable alternative. GM Alexander Volzhin suggested 11 ... Qd7. I think White still
has a large advantage after 12 e5! dxe5 13 Bxe5 when Black’s queen and bishop are
awkwardly placed.
12 Qd2
Menacing h6.
12 ... Kh7 13 e5!
 Principle: Open the game and create confrontation when leading in development.
13 ... dxe5 14 Bxe5
The surest way to lure your opponent into a trap is to tell him exactly what he wants to
hear. In this case the bishop pours a vial of odourless poison in the black king’s morning tea,
allowing Black a ... Ne4 discovery trick, after which Black’s counterplay – if you can call it
that – isn’t enough, and when it does come, it arrives in irregular intervals.
14 ... Ne4
Petrosian – seemingly oblivious to his development lag – indulges in an extravagance by
wasting yet more time in pursuit of the bishop-pair. Black’s unappetizing options:
a) 14 ... Ne8?! 15 Bxg7 Kxg7 16 Ne4 Bc6?? (this move hangs a queen; Black had to try
16 ... Nd6 17 Qc3! Kh7 18 Bf3 when he is under intense pressure) 17 Nf5+ 1-0, A.Kolybin-
Z.Janda, Decin 1998.
b) 14 ... Nc6! 15 Nxc6 Bxc6!.
 Question: Doesn’t this move just hang a pawn?

Answer: I think Black is better off handing over a pawn, rather than submitting to the
wretched 15 ... bxc6?! 16 Rad1, even if 16 Qxd8 Raxd8 17 Bxc7 Rd7 18 Bf4 is an extra
pawn for White in the ending. Even so, this may be Black’s best option, since a sliver of
hope – no matter how faint – is still better than no hope at all.
15 Nxe4 Bxe5 16 Nf3! Bg7!
Petrosian fends off persistent annoyances, the way a man brushes off summer flies
buzzing around his head. He finds the only move within a hive of losing moves:
a) 16 ... Bxb2?! (the last thing Black needs is more loss of time; this move exhausts
Black’s resources past tolerable limits) 17 Rad1! (threat: Nc5) 17 ... Qe7 18 Bc4 Kg7 19
Rb1 Bf6 20 Nxf6 Qxf6 21 Rxb7 and Black’s game is on the verge of collapse.
b) 16 ... Nc6?? hangs material to 17 Nxe5 Nxe5 18 f4! Nc6 19 Qxd7! winning a piece,
due to the f6-knight fork.
17 Rad1
Threat: Nc5.
17 ... Qc8
17 ... Qe7 fails to 18 Bc4 (threat: Neg5+ and Rxe7) 18 ... Kh8 (18 ... Kg8 fails to save
Black after 19 Neg5 Qf6 20 Nxf7! Rxf7 21 Ne5 Be6 22 Bxe6 Qxe6 23 Nxf7 Qxf7 24 Qd8+
Kh7 25 Qe8! Qxe8 26 Rxe8 Bf6 27 Rde1 and Black is hopelessly tangled up on his
queenside development) 19 Neg5 Qf6 20 Nxf7+! Rxf7 21 Bxf7 Qxf7 22 Ne5 Bxe5 23
Qxh6+ Qh7 24 Qxh7+ Kxh7 25 Rxe5 Threat: Re7+ and either rook takes d7. Now if 25 ...
Bc6 26 Re7+ Kh6 27 Rd8 when Black can resign.
18 Bc4 Be8
This awful contortion is not what Black needs to complete his development project. Yet
all of us are sometimes driven to extreme measures when we feel the pin prick of need.

Exercise (combination alert): White’s position experiences multiple states of well

being: 1. Black is missing a chunk of his army (his knight and a8-rook). 2. Tal’s hidden
attackers lean their ears closer to Black’s king, listening to everything he has to say. The
presence of number one on the list, means that Black’s king is insufficiently protected.
Petrosian plans to finally develop his queenside with ... Nc6 next move. Tal never gave him
the opportunity. How should White continue?

Answer: Annihilation of the defensive barrier. This sacrifice is justified since half of
Black’s army is AWOL on the queenside.
19 Neg5+!!
This knight weaves and dances like a candle’s flame, left near a draughty window.
19 ... hxg5
Declining is no better: 19 ... Kh8 20 Qf4 Qf5 21 Qxf5 gxf5 22 Rd8 Bd7 23 Nxf7+ Kh7
24 Re7 Rxd8 25 Nxd8 leaves Black hopelessly busted.
20 Nxg5+ Kg8 21 Qf4!
In a single move, Tal redeems the entirety of his previously unpaid debt. Threat: Qh4
and Qh7 mate. The black king’s deepest fear has come to fruition: he has fallen short of his
ruthless f4 sister’s expectations.
21 ... Nd7
 Exercise (combination alert): Black hopes to swing his knight
to f6, but never gets the chance. White to play and force mate:

Answer: Removal of a key defender.

22 Rxd7! Bxd7
22 ... Qxd7?? entombs Black’s king and allows White to force a quick mate after 23
Qh4.
23 Bxf7+ 1-0
23 ... Rxf7 24 Qxf7+ Kh8 25 Qxg6 Bf5 is met with 26 Nf7+ Kg8 (Black’s king, beset by
enemies, searches frantically for an improvised weapon; at last he found one: a plastic
McDonald’s coffee stir stick, which somehow still fails to compensate for the white queen
and knight’s war axe and spiked morning star) 27 Nh6+ wins.

Game 42
M.Tal-M.Dvoretsky
USSR Championship, Leningrad 1974
King’s Indian Defence

In open tactical fist fights, Tal was universally feared and his domination in chaotic
positions was absolute. The wise tried their very best to shun any kind of complicated
position when facing him. He was the ultimate hawk on the chess board, who fed
emotionally on conflict, until it became his sole reason for playing the game. So it’s natural
for us to sometimes forget that he turned into quite a skilled positional player in his mature
years, as this game demonstrates.
 1 d4 Nf6 2 c4 g6 3 Nc3 Bg7 4 e4 d6 5 Be2 0-0 6 Nf3 e5 7 Be3 c6 8 0-0 exd4 9
Bxd4!?

Slightly unusual. Tal wasn’t one of those players who blindly chased the dictates of
theoretical fashion. I tell my students that those who do so are members of a hive, who
blindly follow without individuality (which invariably annoys them!).

Question: Doesn’t it strike you as odd that you are a

writer of opening books who continually bashes theory?

Answer: I’m not anti-theory; I’m anti blind belief, and against memorization substituting
thinking. When we blindly follow theory’s fashions we return to each game in futile cycles,
like a scripture recited, re-recited and re-re-recited in rote fashion, with its essential
teaching still remaining uncomprehended. If you follow a fashionable line, then contemplate
it deeply, and follow your intuition, rather than what you are told to believe.
9 Nxd4 is certainly a more natural move, since it frees White’s f-pawn for f3, supporting
e4.
9 ... Re8 10 Qc2
Covering e4.
10 ... Nbd7
10 ... Qe7 is met with 11 Rfe1, which indirectly protects e4.
11 Rad1 Qe7 12 Rfe1 Ne5
Otherwise:
a) 12 ... Nc5 can be met with 13 h3 when e4 is immune, since if 13 ... Nfxe4? 14 Bxg7
Kxg7 15 Bd3 f5 16 Bxe4 Nxe4 17 Nd2 Kg8 18 Ndxe4 fxe4 19 Qxe4 Be6 20 Qd4. Black is
under heavy pressure, with Ne4 in the air.
b) 12 ... Nxe4?? 13 Bxg7 Kxg7 14 Nxe4 Qxe4 15 Bd3 wins serious material.
 13 h3 Bh6 14 b4
Tal proceeds with his kingside initiative.
14 ... b6?!

An inaccuracy which allows Tal to inflict damage upon Black’s structure. 14 ... Nfd7
was a superior choice.
15 c5!
This move is the first intimation that something isn’t quite right in Black’s position.
15 ... bxc5?
Black minimizes his disadvantage with 15 ... Nxf3+ 16 Bxf3 dxc5 17 bxc5 bxc5 18 Be3
Bxe3 19 Rxe3.
16 Nxe5! dxe5
After 16 ... cxd4 17 Nxc6 Qd7 18 Nxd4 White wins a pawn.
17 Bxc5 Qb7 18 Na4!
Stressing his control over the c5 hole. Not only that, but Black must eternally watch over
his weak c6-pawn.
18 ... Be6 19 Bd6 Nd7 20 Nc5 Nxc5 21 bxc5!
 Question: Doesn’t this move nullify the
pressure on Black’s formally backward c6-pawn?

Answer: It’s a matter of switching one advantage for another. It’s true that this move
helps Black with the c6-pawn, yet White secures multiple advantages with the move:
1. He secures d6 as an outpost for either his bishop or a rook.
2. On d6, White’s bishop controls b8, which in turn means that Black is unable to
challenge White on the b-file. So Black now is threatened with a future invasion on b7.
3. By recapturing on c5 with a pawn, Tal eliminated any ... a5 simplification ideas.
The rote 21 Bxc5?! is met with 21 ... a5 22 a3 axb4 23 axb4 when Black’s position
doesn’t look so bad anymore.
21 ... Bf8
Necessary, otherwise Black is unable to play ... Rb8 to challenge the b-file.
22 Rb1 Qd7 23 Red1 Bxd6 24 cxd6!
 This deeply entrenched passed pawn exerts an asphyxiating effect on Black’s game.
24 ... Rab8 25 a4
Freeing White’s queen from guard duty to a2.
25 ... Rxb1 26 Qxb1 Qd8
Perhaps thinking about playing ... Qb6.
27 Qc2 Qd7
On 27 ... Qb6 Black gets pushed back after 28 Qc3 Bd7 29 a5 Qb7 30 Bd3! (intending to
seize control over the b-file with Rb1 next) 30 ... Qb8 31 Qa3 Re6 32 Bc4 Re8 33 Qc5 Kg7
34 Rd3 Qa8 35 Rb3 f6 36 Qb4 (threat: Qb7) 36 ... Bc8 37 a6 with complete domination.
28 Qc5 f6 29 a5 Kg7?
Black’s position – much like ageing – is so gradual, that you only see differences when
you look back ten moves (or years!) or so, and realize to your dismay, that your position has
gone noticeably downhill. After this error, Black’s slow drip of strategic complaints turns
into a major leak down the b-file. Dvoretsky refuses to grab hold of his offered lifeline. It
was absolutely crucial for Black to seize the file first with 29 ... Rb8.
30 Rb1
Of course. Now White threatens a6, followed by Rb7.
30 ... Rd8 31 a6! Kh6
If 31 ... Qxd6 32 Qxa7+ Kh6 33 Rb7 Bd7 34 Qe3+ g5 35 a7 c5 36 Qf3! (forcing Black’s
king into a second rank pin) 36 ... Kg6 37 Qh5+ Kg7 38 Bb5 and Black can resign.
32 Qe3+!
Tal forces further weakness in the pawn front around his opponent’s king.
32 ... g5
Not much to think about here since 32 ... Kg7?? walks into 33 Rb7.
33 Qf3! Qf7
Black can only passively witness the build-up around his king with growing anxiety.
 34 Rb7 Rd7
34 ... Bd7 35 Rxa7 is resignable for Black.

Exercise (combination alert): For Black, obligation arrives in

full force, like a half-forgotten promise, made long ago. Dvoretsky’s
defenders look as immovable as a cathedral’s gargoyles, who watch
threateningly, but only as harmless symbols. How did Tal finish?

Answer: Overloaded defenders.

35 Bc4!
A friend will offer help with no thought of reward; if a stranger offers help with no
return, be careful. You may be encountering a con-artist. Of course, to Tal, such a
combination represented a trivial sub-routine in his computations. He must have visualized
this shot long ago.
35 ... f5
If your life has been one full of anguish, you barely notice if a little more is added. This
final errant strand of counterplay is easily brushed aside. White’s bishop can’t be touched.
If:
a) 35 ... Bxc4?? (if a dog chases a car, the worst thing that can happen is for him to catch
the car) 36 Rxd7 Qxd7 37 Qxf6+ Kh5 38 g4+ Kh4 39 Qh6 mate. “I have discovered the
whereabouts of my Judas,” declares the queen, as she points an accusing finger at her
cowering brother.
b) 35 ... Kg7?? 36 Bxe6 Qxe6 37 Qf5! wins on the spot.
36 exf5 Qxf5
36 ... Bxc4 is met with 37 Qxc6 Rxb7 38 d7+ Kg7 39 Qxb7 Bxa6 40 Qc7 and White
forces promotion.
 37 Bxe6 1-0
After 37 ... Qxe6 there follows the familiar mating pattern: 38 Rxd7 Qxd7 39 Qf6+ (the
queen follows Black’s king the way a hungry seagull follows a fishing boat) 39 ... Kh5 40
g4+ Kh4 41 Qh6 mate.

Game 43
M.Tal-L.Portisch
Biel Interzonal 1976
Ruy Lopez

1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Be7 6 Re1 b5 7 Bb3 d6 8 c3 0-0 9 d4

Fischer also briefly flirted with this line, which allows Black to play ... Bg4. 9 h3 is
White’s main line.
9 ... Bg4
This move increases pressure on d4.
10 Be3
10 d5 Na5 11 Bc2 c6 12 h3 Bc8 13 dxc6 Qc7 14 Nbd2 Qxc6 is considered satisfactory
for Black.
10 ... exd4

Question: Why did Black give up the centre?

Answer: This is actually Black’s most dynamic option, where he soon gets a Benoni-
like structure. Instead, after 10 ... Nxe4?! (nothing of value is free in life) 11 Bd5 (skewering
both black knights; however, the story isn’t over yet) 11 ... Qd7 12 Bxe4 d5 13 Bc2 e4 14 h3
Bh5 15 Ne5! Bxd1 16 Nxd7 Bxc2 17 Nxf8 Rxf8 18 Bf4 Black is just short of full
compensation for the exchange in the ending, but 10 ... Na5 is the safer route.
11 cxd4 Na5 12 Bc2 Nc4
12 ... c5 is also played here.
13 Bc1
White’s bishop and b2-pawn were simultaneously attacked, so there wasn’t much choice
but to retreat to the bishop’s place of origin.
13 ... c5
Portisch stakes out a portion of central control for himself.
14 b3 Nb6 15 Nbd2
15 d5? is premature. Black can meet it with 15 ... Nfxd5! 16 exd5 Bf6 when White loses
material, M.Keller-C.Renner, Abensberg 1987.
15 ... Nfd7
Black gets out of the way of an oncoming e5, and also opens the possibility for ... Bf6.
16 h3 Bh5 17 g4!?

Question: Isn’t Tal risking overextension with this move?

Answer: Not just yet, although the move clearly contains the potential to do so later on.
As usual, Tal opts for the huffiest option, rejecting the calmer 17 Bb2.
17 ... Bg6 18 Nf1
Once again, I think it’s in White’s best interest to keep the centre fluid. If 18 d5 Bf6 19
Rb1 Re8 and maybe it’s stylistic, but it feels to me like Black achieved a nice looking
Benoni and I already prefer his side, L.Dominguez Perez-V.Bologan, Poikovsky 2005.
18 ... d5!?
Portisch seeks to clarify the central tension. Black had a reasonable alternative in 18 ...
Rc8.
 19 e5 Rc8
The slightly more accurate 19 ... Bxc2! prevents Tal’s next move.
20 Bf5!
This move is annoying for Black, since he won’t be inclined to trade on f5.
20 ... c4!?
Portisch relies on his newly acquired queenside pawn majority, which would probably
favour him in an ending. However, we are a long way from any ending, and Tal now builds
up his kingside majority into a potential attack. The trouble with such a move is that it
stabilizes White’s centre, enabling him to attack on the kingside, without too much fear of a
distracting central counter with ... cxd4.
21 Ng3 Rc6 22 Kg2 Re8 23 Rh1!

A dark undertow tugs Tal’s thoughts back to his opponent’s king. He intends to force
Black’s hand with a coming h4 and h5.
23 ... f6
Tal’s attempted entry into the kingside sends alarms sounding. Principle: Counter in the
centre when assaulted on the wing.
24 h4 fxe5!?
When we engage in an irrevocable alteration of our position, with it comes optimistic
joy, mingled with a disquieting fear of the unknown. This move creates mutual super
majorities on king and queenside. Portisch gets use of c5, while for Tal, d4 opens as a
potential square. In essence, each party’s wing attack accelerates.
25 dxe5
The structure, until now, fits no predetermined form.
25 ... Bc5
Opening up an attack on e5.
26 Re1
 I would be more inclined to leave the h1-rook where it stands and develop with 26 Bb2!
which also adds protection to the d4-square.
26 ... Bb4 27 Re3 d4!?
This little combination looks like it wins an exchange for a pawn. However Tal had seen
deeply into the coming complications.
28 Qxd4! Bc5?!
Portisch should have probably admitted his project was flawed and played 28 ... Qc7!
when he might just have adequate central pressure for the pawn.
29 Qe4!
The point. Black’s c6-rook also hangs.
29 ... Bxf5 30 Nxf5
Tal always relied on his pieces over his pawns, rejecting the also playable 30 gxf5.
30 ... Bxe3?!
30 ... Qa8 minimizes the damage.
31 Qxc6 Bxc1 32 e6!?

Your writer – much like Tal’s e-pawn – gets irritated when driving, and some guy ahead
of me rudely blocks my lane, because he has the gall to drive at the speed limit. Tal gets
fancy and throws away his advantage, but at the same time concocts a devilishly deep snare
for Portisch. His move is made with the philosophy: in irrational positions, an irrational
solution is sometimes the most plausible of our choices.

Question: Isn’t Tal’s last move a double question mark, since 32 ... N7b8 wins a piece?

Answer: Please see the game continuation! There are two kinds of lies: by commission
and by omission, the latter of which is the more subtle. Tal’s move is also the latter in this
case, where he knowingly or unknowingly abandoned a promising continuation to set up a
very, very deep trap. How can I fault him when Portisch, a world-class player, walked right
into it? I hate to complain about Tal’s objectively bad moves which turn out so beautifully
against human opponents.
Komodo frowns on Tal’s move, which to be fair, is basically an incredibly deep cheapo.
Still, doesn’t it strike you as bizarre, that truth chooses to inhabit an insentient artificial life
form like Komodo, rather than in us humans, whose collective hearts bleed from our patent
inferiority to the comps? Objectively correct is Komodo’s suggestion 32 Rxc1! Nxe5 33
Qb7 (threatening mate) 33 ... Ned7 34 bxc4 bxc4 35 Rd1 when Black is badly tied up. Note
that 35 ... Qc8?? is impossible due to 36 Qxc8 Rxc8 (or 36 ... Nxc8 37 Rxd7) 37 Ne7+
winning a rook.
32 ... Nb8?
Now the defence clenches like a spasming muscle. It’s easy to confuse our needs with
our desires. In this manner Black appropriates a piece, yet loses the game. Tal had foreseen
the consequences with greater depth and clarity. 32 ... Qf6! 33 Rxc1 Rxe6 leaves the game
dynamically balanced, and White’s excavations turn up nothing.
33 Qb7!
The deepest of traps is when we convince the opponent that we are the marionette, and
he or she pulls the strings – when in reality, the opposite is the case. The queen veers
sharply, as if suddenly recalling an urgent appointment. Mate is threatened, as well as
Black’s bishop. Now in case you think Tal’s sacrifice was simply material being swept into
a black hole, just keep watching.
33 ... Bb2

Question: This move saves the extra piece, while simultaneously

covering the g7 mate. Isn’t this an absolute refutation of Tal’s idea?
 Answer: Unfortunately for Portisch, Tal had prepared a refutation of the refutation!
34 Qf7+ Kh8 35 Rd1!
Playing on Black’s weak back rank.
35 ... Qc8 36 Ng5!
When separate people who share a common enemy gather, it begins to take on the
appearance of a conspiracy. In this instance a number of witnesses come forth to falsely
testify against Black’s king. Black is curiously helpless to prevent Tal’s coming
combination.
36 ... Bf6
Alternatively:
a) 36 ... h6 37 Qg6 hxg5 38 Qh5+ Kg8 39 Qf7+ Kh8 40 hxg5 leaves Black helpless
along the newly opened h-file, to a coming Rh1, forcing mate.
b) 36 ... Rf8 is met with the nasty 37 Nh6!! forcing mate, similar to the game’s
continuation.

Exercise (combination alert): Tal’s following combination is hauntingly similar to the

one he played on Smyslov in the first chapter. White to play and force mate:

Answer: Smothered mate/overloaded defender/interference. White threatens Qg8+ and

Ngf7 mate.
37 Nh6!! 1-0
“Arrest that prancing miscreant!” screams Black’s king. This sudden shift reveals data
which had been previously concealed. One can’t help but be awed by Tal’s dexterity in such
chaotic situations. This striking idea (so similar to Tal’s final combination against Smyslov,
in Chapter One) makes an indelible mental imprint into our inner database. The rays of Tal’s
focus converge upon a single square: g8.
 After 37 ... Rxe6 (37 ... Re7 is met with 38 Qxe7! Bxe7 39 Ngf7 mate, or if 37 ... gxh6
38 Qxh7 mate) 38 Rd8+!! Qxd8 (the act of blinking is either one to moisten dry eyes, or as a
gesture of disbelief; in this instance, the black queen’s incessant blinking is due to the latter
reason) 39 Nxe6 Qc8 40 Nf8! (this deadly interference trick forces mate) 40 ... Qc6+ 41 f3
Qd5 42 Ne6 Qd2+ 43 Kh3 Black’s luck runs out and he is mated.

Game 44
L.Polugaevsky-M.Tal
Riga Interzonal 1979
English Opening

One of Tal’s last really great triumphs was his victory in the 1979 Riga Interzonal. He
defeated a normally difficult opponent in Polugaevsky. Then Tal’s first opponent in the
Candidates’ quarterfinals was Polu, who there defeated Tal soundly. After the match, Tal
jokingly said about his decline: “Now I am Polu-Tal!” which translates to ‘semi-Tal’!
1 Nf3
My next book will be on the Sveshnikov Sicilian. If I have the room, I may add short
chapters on the following rarely played lines, which are similar to the opening in Tal’s
game: 1 e4 c5 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 e5 6 Ndb5, and now I plan to cover
both 6 ... Bc5!?, and also 6 ... h6!?, Ulfie’s line. The positions reached are similar to the one
Tal gets in this game.
1 ... c5 2 c4 Nf6 3 Nc3 d5 4 cxd5 Nxd5 5 e4!?

If you believe a person to be your enemy, he will become one, even if he was not your
enemy before. This is White’s most aggressive option.

Question: Doesn’t this move create a gaping hole on d3?

 Answer: It weakens d3, but Black’s trouble is his knight isn’t all that stable on d3,
unless he is willing to make a radical concession (and, of course, Tal does just that in the
game). 5 g3 and 5 d4 are the more positional routes.
5 ... Nb4 6 Bc4
6 Bb5+ and 6 d3 are also played here, and White must avoid the trap 6 d4?! cxd4 7
Nxd4?? which hangs a piece to the attraction/knight fork trick 7 ... Qxd4!.
6 ... Be6!?

Once Tal decided to complicate, order was virtually impossible to restore. Wow. There
is a difference between shaping a position, as opposed to radically altering it. Tal decides to
follow a hunch, over tried methods. When a top player advocates a potentially shady line, he
gives the rest of us moral authority to follow suit.

Question: Why on earth did Tal agree to the decimation of his own structure?

Answer: Tal’s style fails to fit the impress of a more standard mould. With the exchange
of bishops, he seizes control over d3, gambling that his knight, soon to arrive on d3,
compensates his damaged structure. A normal continuation would be 6 ... Nd3+ 7 Ke2 Nf4+
8 Kf1 Ne6 9 b4!? (this is sort of a Wing Gambit for the English Opening) 9 ... cxb4 10 Ne2
Nc7 11 d4 when White’s central domination and development lead compensate him for the
pawn, G.Kasparov-R.Gruenberg, Hamburg (simul) 1985.
7 Bxe6 Nd3+ 8 Kf1
After 8 Ke2?! Nf4+ 9 Kf1 Nxe6 Black gets all the benefits of White’s d4 hole, without
having to pay a price himself.
8 ... fxe6
Even from their youth, the twins were considered homely children.
 9 Ng5!
I think this is a better try for the initiative than the more cautious 9 Ne1 Nc6 10 Nxd3
Qxd3+ 11 Qe2 Qd7 12 d3 g6 13 Be3 b6 14 h4 0-0-0 15 Rd1 Bg7. Black’s domination of d4,
and pressure on d3 more than compensate for his ugly doubled e-pawns, Y.Seirawan-
V.Ivanchuk, Monaco (rapid) 1993.
9 ... Qb6?!
In chess it’s not easy to escape the past – which is synonymous with theoretical
precedence. This was a new move at the time, and an attempted improvement (which it
clearly isn’t!) over GM Stean’s game from the previous year. This wasn’t one of those cases
where Tal forgot the continuation of the previous game (as we age, our mental database
begins to corrupt and we can’t remember a damn thing, and remembrance of the book move
tends to slip aside when our conscious mind attempts to articulate it in our own personal
style).
Stronger is Stean’s move 9 ... Nc6! 10 Nxe6 Qd7 11 Nxc5! Nxc5 12 Qh5+ g6 13 Qxc5,
J.Timman-M.Stean, Amsterdam 1978. Black’s development lead and open f- and d-files
offer him full compensation for the pawn if he plays 13 ... e6!.
10 Qe2?!
This meek response justifies Tal’s last move. White should play the energetic 10 Qf3!
threatening both the d3-knight and f7 infiltration: 10 ... h6 (10 ... Ne5 11 Qh3! Nbc6 12 f4
Nf7 13 Nxe6 favours White) 11 Nxe6! Qxe6 12 Qxd3 Nc6 13 Qb5 b6 14 h4 when Rh3 is
coming and I don’t believe in Black’s full compensation for the pawn.
10 ... c4

Tal buys his d3-knight some time on its square.

11 b3
Preparing to undermine the support post of d3.
11 ... h6 12 Nf3 Nc6!
 Tal offers a pawn, hoping his development lead and dominating d3 outpost compensate.
13 bxc4 0-0-0 14 g3?
This is just too slow. Polugaevsky should have taken firmer action with 14 Ne1 Nce5 15
Ba3 Nxe1 16 Rxe1 Qa5 17 Bb2 Nd3 18 Rb1 g5 19 Nd1 Rg8 20 Bc3 Qxa2 21 Ra1 Qxc4 22
Rxa7 Kb8 23 Ra3 and Black only stands a shade better after 23 ... Bg7.
14 ... g5!
Tal prepares to develop his last undeveloped piece, and then double rooks down the f-
file. Soon, we see that Tal’s bead on f2 is the invisible hand which manipulates all which
transpires.
15 Kg2 Qc5!
Preventing Ba3, while menacing ... Qxc4.
16 Rb1
Slightly more accurate was 16 Ne1 Nce5 17 Nxd3 Nxd3 18 Rf1 Bg7 19 Rb1.
16 ... Bg7 17 Nb5?
This move forecloses on White’s avenue of defensive opportunity. Polugaevsky’s
counterplay – a story in a still unread novel – lies in dormancy. This violent attempt to seize
the initiative from a position of inferiority, fails. White should just suffer quietly with a line
like 17 Ne1 Nce5 18 Nxd3 Nxd3 19 Rb3 Qxc4 20 Ba3 when he stands only slightly worse.
17 ... Qxc4 18 Qe3
Attacking a7.
18 ... Rhf8!
Hey, I said: “Attacking a7”! Tal ignores the threat and begins his own attack on f2.
19 Rf1
19 Nxa7+?? Nxa7 20 Qxa7 Qxe4 21 Qe3 Qd5 22 h3 h5 threatens ... g4, and if White
plays 23 g4 he is mated after 23 ... Rxf3! 24 Qxf3 Nf4+ 25 Kg3 h4+.
19 ... g4 20 Nh4
 Exercise (combination alert): White has two promising
continuations in this position. What would you play here?

Answer: Overloaded defender.

20 ... Nxf2!
Answer no.2: Even stronger was 20 ... Qxa2! 21 Ba3 Rxf2+! (overloaded defender) 22
Rxf2 Qxb1 (both b5-knight and f2-rook hang) 23 Nxa7+ Nxa7 24 Qxa7 Ne1+! 25 Kh1 Nf3+
26 Kg2 Qg1 mate.
Answer no.3: Kasparov suggested 20 ... Rxf2+! 21 Rxf2 Nxf2 22 Qf4 (threatening mate
on c7) 22 ... Be5 23 Na3 Qe2! 24 Qe3 Qxe3 25 dxe3 Nxe4 with an easy win.
21 Ng6
21 Rxf2 is met with 21 ... Rxf2+ 22 Kxf2 (forced, since White’s queen must cover e4)
22 ... Rf8+ 23 Nf5 exf5 with a winning attack.
21 ... Rd3!
Tal’s impertinent pieces begin to take a haranguing tone with White’s queen.
22 Na3
22 Qe2?? is met with the crushing 22 ... Rxg3+ 23 hxg3 Qxe2.
22 ... Qa4 23 Qe1
The queen checks herself into a battered women’s centre.

Exercise (combination alert): In this position, we are reminded of high

school physics class: light and matter may enter a black hole’s aperture,
but never exit. Now what? White threatens Rxf2, as well as Nxf8. White’s
kingside is nothing more than a well-tended graveyard. What did Tal play?
 Answer: Exchange sacrifice/zwischenzug.
23 ... Rdf3!
Tal ruled in positions which required theoretically abstracted ideals.
24 Nxf8 Nd3!
This deadly zwischenzug crumples the defence.
25 Qd1
Abandoning e4 leads to mate. However, 25 Qe2 walks into 25 ... Nd4.
25 ... Qxe4
Threatening horrible discoveries. Now e4 – the last line of White’s defensive hopes –
finally falls, like the cherished final tooth of an elderly man.
26 Rxf3
26 Kg1 Bd4+ is of no help for White.
26 ... gxf3+ 27 Kf1
27 Qxf3 walks into the fork 27 ... Ne1+.

Exercise (combination alert): How did Tal force mate?

Answer: There is no defence to the infiltration of h3.

27 ... Qf5!
Answer no.2: 27 ... Qg4! also forces mate.
28 Kg1 Bd4+ 0-1
“It is foolish to pretend that our roles in life have not altered,” the bishop tells White’s
king. Polugaevsky doesn’t need to be shown the finish: 29 Kh1 Nf2+ 30 Kg1 Nxd1+ 31 Kf1
Qd3+ 32 Ke1 Qe2 mate. The queen offers a look of baffled innocence when asked about her
brother’s mysterious death, while only she was in the room.
Game 45
M.Tal-Y.Rantanen
Tallinn 1979
Sicilian Defence

1 e4 c5 2 Nf3 Nc6 3 Bb5

The Rossolimo Sicilian, so popular now, was a rare line at the time of the game.
3 ... e6
Also commonly played are 3 ... g6 and 3 ... d6, the latter of which I advocate in Anti-
Sicilians: Move by Move.
4 0-0
I generally chop the c6-knight in this position.
4 ... Nge7 5 Nc3
Today, 5 c3 and 5 Re1 are more commonly played.
5 ... a6

6 Bxc6

Question: What does Tal get for handing Black the bishop-pair?

Answer: What he loves most: a development lead. Chances are approximately balanced,
according to theory.
6 ... Nxc6 7 d4
Principle: A development lead takes precedence over the opponent’s bishop-pair, so
it’s correct for White to open the position.
7 ... cxd4 8 Nxd4 d6 9 Re1
 Question: What is the point of placing the rook on e1?

Answer: It serves two purposes:

1. White can play for Nd5 tricks.
2. White can contemplate Nxc6 and e5 ideas, later on.
9 ... Bd7
After 9 ... Be7 10 Nxc6 bxc6 11 Qg4! (this move proves to be slightly awkward for
Black) 11 ... Kf8 12 b3 Qa5 13 Bb2 e5 14 Qe2 Be6 15 Na4! f6 16 c4 I slightly prefer
White, since Black’s bishops don’t work all that well in this structure, E.Safarli-A.Korobov,
Baku 2012.
10 Nxc6
Tal makes another concession, handing Black greater central influence, if he can
increase his development lead in doing so.
10 ... bxc6
10 ... Bxc6 is met with the annoying 11 Nd5!, ensuring that Black can’t hang on to the
bishop-pair.
11 Qg4!
Black may have both bishop-pair and strong pawn centre, yet he clearly suffers from
developmental issues on the kingside.
11 ... e5 12 Qg3 f6?!
 Question: I don’t understand this move. What is Black’s defensive idea?

Answer: The move contains two ideas:

1. Black reinforces e5, just in case White plays f4.
2. Black plans to castle by hand, with ... Kf7, ... Be7, ... Re8, and ... Kg8. The obvious
problem with this plan is that it is very slow. And as we all know by now, allowing Tal time
to generate mischief is a dangerous strategy.
12 ... g6 looks like an improvement, J.Moreno Carnero-S.Alonso, Buenos Aires 1993.
Now White can try 13 b3 Be7 14 Bh6 g5!? 15 Nd1 Rg8 16 h4 Rg6 17 hxg5 Bxg5 18 Bxg5
Qxg5 19 Qxg5 Rxg5 20 Ne3. Black looks okay to me in this ending.
13 Be3
Also reasonable would be to keep options open for the bishop’s placement with 13 b3
Kf7 14 f4.
13 ... Kf7 14 Rad1 Be7 15 Na4
The idea is to clamp down on d6 with c4, worrying Black about a potential c5 push later
on. 15 f4 is the alternative.
15 ... Rb8 16 b3 Re8 17 c4 c5!?
Now d6 is condemned to lifetime backwardization (you may not want to Google this
‘word’) for life, and Black also created a self-inflicted hole on d5, but if 17 ... Be6 18 c5!
d5 19 exd5 cxd5 20 f4! e4 21 Qf2. White still holds a clear advantage, since Black’s central
majority is firmly blockaded on the dark squares.
18 Nc3 Bf8 19 f4 Kg8
An example of the hidden dangers inherent in Black’s position: 19 ... a5 20 Rd3 a4?! 21
fxe5 fxe5? (forced is the highly unpleasant 21 ... Rxe5 22 Bf4 Re6) 22 Bg5! Qc8 23 Rf1+
Kg8 24 Nd5 Bg4 25 Qh4! axb3 26 axb3 Be2 27 Nf6+! gxf6 (27 ... Kf7 28 Qxh7 Bxf1 29 Rf3
gives White a winning attack) 28 Rg3! and White whipped up a winning attack.
20 Rf1 Qc8 21 fxe5 fxe5
21 ... dxe5 is met with 22 Rxf6! Bg4 23 Rxf8+ Rxf8 24 Rd5 Be6 25 Rxe5 Qd7 26 Bxc5
Qd2 27 h3! (White picked up three pawns for the exchange, but he must avoid 27 Bxf8??
Rxf8 28 h3 Qc1+ 29 Kh2 Rf1 30 Qd3 Rh1+ 31 Kg3 Qe1+ when Black gets a winning
attack) 27 ... Qc1+ 28 Kh2 Rf1 29 Qe3 and White stands clearly better.
22 Nd5
Threat: Nf6+.
22 ... Kh8?!
This move, which transgresses the tolerable limits Black’s position is able to bear,
allows White’s rook to seize the seventh rank, adding another attacker against Black’s king.
Black had to try 22 ... Bg4 23 Rd2 Kh8 24 Rf7 Bh5 25 Rc7 Qg4 26 Qxg4 Bxg4 27 Bg5 with
a very favourable ending for White.
23 Rf7! Bg4
Not 23 ... Be6?? 24 Rc7 Qd8 25 Bg5 and Black’s queen has no place to go.

Exercise (combination alert): How to navigate the staggering columns

of calculations? White has multiple promising paths. Find one of them:

Answer: Tal opts for the flashiest line.

24 Nf6!
I strongly suspect that Tal willingly entered such lines (where multiple pieces hang!)
with no true idea if the position he entered into was favourable or unfavourable for him. He
basically trusted that interaction, coupled with an alert, receptive mind, would eventually
lead him to the position’s truth. And if he failed to discover the truth, then he could always
fall back on a good, old fashioned cheapo!
 Answer no.2: The comps have to ruin our fun by pointing out the even stronger line 24
Qh4! Qe6 (or 24 ... Bxd1?? 25 Nf6 h5 26 Qg5! and there is no defence to the dual threats
Qg6 and Qh6+) 25 Rdf1 Qg6 26 Nc7 Rec8 27 h3 Be6 28 Nxe6 Qxe6 29 R1f6! Qxf6 30
Rxf6 gxf6 31 Qxf6+ Kg8 32 Qe6+ Kh8 33 Bg5 Bg7 34 Qxd6. White picks up too many
pawns and is winning.
24 ... gxf6
Short-lived relief (or at least the illusion of it) gives way to apprehension. 24 ... Bxd1??
is met with 25 Qh4 h5 26 Qg5!. Black is defenceless against the dual threats of Qh6+ and
Qg6, both with the aim of delivery of checkmate on h7.
25 Qh4
Black’s king is rightfully disturbed when he notices his sister Googling ‘undetectable
poisons’. Tal menaces immediate mate on h7.
25 ... Bg7
When we are forced to make a series of only moves, we become the chess equivalent of
zombies, devoid of free will.
26 Bh6 Bxd1??
In a calculation of this sort, there is zero allowance for variance, since each move is the
only move. Black’s seemingly eternal reserve of patience is finally breached. Ideally, we try
to pick the perfect moment to strike. In reality, most of us strike when we are compelled by
events, much like a ship tossed in a storm. When you go over Tal’s games, it feels as if his
opponents’ eyes are only sporadically open to the truth. The second they close them, he finds
a way to exploit the lapse.
Black had to try 26 ... Qe6! (only with this move does Black kindle a few sparks of
hope) 27 Bxg7+ Kg8 28 Rxf6 Bxd1 29 Rxe6 Rxe6 30 Bf6 Bh5! and Black has chances to
defend the position.
27 Bxg7+ Kg8
 Exercise (combination alert): When you don’t know your enemy’s location,
you live with the uncomfortable feeling that he or she constantly lurks behind
you. This is clearly the case for Black, whose resistance is as impermanent as a
dying person’s final breaths. The g8 dandy is no match for the muscular thugs
which surround him, and Tal found a pretty mate. Do you see it?

Answer: Clearance/attraction.
28 Bh8!!
The bishop plunges down on Black’s king with the finality of the guillotine’s blade.
When our opponent’s combination works, it’s like looking at a reflection of your worst fear,
and seeing it come to life.
28 ... Kxf7
Apparently, ‘free’ rooks are Tal’s fastest growing export.
29 Qxf6+ Kg8
Black’s king sighs forlornly when he contemplates the fact that he was born ten minutes
before his twin f6 sister, thinking: “That was the last time I beat her at anything.”
30 Qg7 mate (1-0)

Game 46
B.Spassky-M.Tal
Montreal 1979
Queen’s Indian Defence

1 d4 Nf6 2 c4 e6 3 Nf3 b6 4 e3 Bb7 5 Bd3 d5 6 b3

 Question: Isn’t this line a bit of a passive choice for White?

Answer: I don’t think so. Spassky goes old school, trotting out the Zukertort Colle,
which is anything but a quiet line (actually, it’s a Zukertort versus a Zukertort set-up). White
often plays for an early Ne5, followed by f4 and then plays for mate on the kingside. Some
of my students play the same openings over and over again, and I can clearly see that doing
so barricades them into preconceived notions and automatic opening play without thought.
For me, it’s refreshing to play new lines, even if it costs me my hereditary ‘+=’ as White.
6 ... Bd6
Black’s most common continuation. Some players avoid posting the bishop on d6, so that
they can reply to White’s future Ne5 with ... Nxe5, without getting forked after dxe5.
7 0-0 0-0 8 Bb2 Nbd7 9 Nbd2
9 Nc3 posts the knight more aggressively, at the cost of diminished control over the key
e5-square.
9 ... Qe7
Alternatively, 9 ... Ne4 10 Qe2 f5 11 cxd5 exd5 12 Ne5 Qe7 13 f4 c5 14 Nxe4 fxe4,
F.Bellia-E.Postny, Bad Wiessee 2014. I prefer White’s game after 15 Bb5!.
10 Rc1 Rad8
A move like this betrays Tal’s subliminal aggression. A positional player would have
posted rooks on c8 and d8, or played ... Ne4. Tal reserves his f8-rook for possible central
and kingside action later on.
11 Qc2
Spassky too takes aim at Black’s king, while retaining control over e4. His move is
more accurate than 11 Qe2 Ne4!.
11 ... c5 12 cxd5
Dramatically sharpening the position by inducing Tal into a hanging pawns centre.
12 ... exd5
Of course, Tal takes up the challenge. He isn’t the type to go for a line like 12 ... cxd4 13
Nxd4 Bxd5 when White holds a slight and safe edge.
13 dxc5 bxc5
Komodo, who may be having a bad day, suggests 13 ... Nxc5?! 14 Bxf6 Qxf6 15 Bxh7+
Kh8 16 Bd3 claiming nearly full compensation for a pawn for Black. To my human eyes, I
just see an extra pawn for White.
14 Qc3!?
The idea is to tie Black’s f6-knight down, due to the g7 mating threat. However, Spassky
may have been goading Tal into a premature ... d4 thrust. And he later got his wish.
14 ... Rfe8 15 Rfd1 d4!?
 Well, we all understand that restraint isn’t Tal’s strong suit. He just wasn’t the type to
tease out an opponent’s intentions with endless probing, when instead, for the ridiculously
low cost of a pawn, he could introduce chaos. With this move, multiple, seemingly diverse
ideas all boil down to a single theme: the lack of safety for White’s king.

Question: Isn’t this premature?

Answer: According to the comps, it certainly is, and once again, any other player than
Tal would get a ‘?!’ for the move. Yet Tal was able to confuse a former world champion into
going astray quickly. Everyone else in the world would have played 15 ... h6.
16 exd4 cxd4 17 Qa5?
This move may actually be an unspoken compliment to Tal’s tactical abilities. Spassky
uncharacteristically backs down and fate intervenes on Tal’s behalf. Correct was to accept
with 17 Nxd4! Be5 (maybe Spassky feared 17 ... Bxh2+?, but then comes 18 Kxh2 Ng4+ 19
Kg3 Qe5+ 20 f4 Qe3+ 21 N2f3 Bxf3 22 Bxh7+ Kxh7 23 Qxe3 Nxe3 24 Rh1+ Kg8 25 Kxf3
Nd5 26 Nf5 with an extra pawn and an initiative for White) 18 Qc2 g6 19 N4f3 Ng4 20 Bb5
when I don’t see a way for Black’s attack to proceed, and he remains down a pawn.
17 ... Ne5!
Now Black generates a dangerous build-up around Spassky’s king, without even a single
pawn invested.
18 Nxe5 Bxe5 19 Nc4
Kasparov gave his move a question mark, but it feels to me like White has no good
options. For example, 19 Nf1 Nh5! 20 Ng3 Nf4 and if 21 Bf1? Nh3+! 22 gxh3 Bxg3! is
decisive, since White must deal with the killing threat of ... Qe4.
19 ... Rd5!
A move freighted with not-so-hidden intent. Tal’s potential attackers methodically
reconfigure themselves, just like our computer, when it updates a new version of a program.
20 Qd2

Exercise (combination alert): White’s game, assailed with acute anxiety, isn’t an
idyllic scene, of the kind portrayed by Norman Rockwell in his paintings. Expose
your weakness to your enemy, and you irresistibly draw him to you. In this
instance, White’s king is insufficiently protected. A great trial is about to be
visited upon Spassky, one which he fails to survive. How did Tal continue?

Answer: Annihilation of defensive barrier.

20 ... Bxh2+!
Idle speculation is pushed aside in favour of concrete calculation. This is a stock
sacrifice, made deeply unusual from the fact that its victim is a former world champ.
21 Kxh2?
A move played with the philosophy: if I am destined to die, then I will fight to the last
and die unrepentant. When we unwisely grab material when under attack, we forget an
essential truth: life is more valuable than material wealth. This is a misguided attempt to
avoid an uncomfortable truth. Acceptance walks into a forced mate. White can still play 21
Kf1!. There is no need for an eeny, meeny, miny, moe when it comes down to an only move.
White is still busted after 21 ... Rh5 22 Re1 Bf4! 23 Rxe7 Rxe7 24 Kg1 Bxd2 25 Nxd2 h6,
but at least he can still play on.
21 ... Rh5+!
Attackers, thick as a stagnant pond’s mosquitoes, swarm White’s king. This is more
accurate than 21 ... Ng4+ 22 Kg3 Rg5 which is also lost for White.
22 Kg1
Or 22 Kg3 Ne4+ 23 Bxe4 Qh4+ 24 Kf3 Qxe4+ 25 Kg3 Qh4 mate.
 22 ... Ng4! 0-1
If 23 Qf4 Qh4. “Oh you amusing, little man. Did you really believe you could defeat me,
when I am the original, while you are a mere copy?” the queen asks her younger brother.
Black forces mate.

Game 47
M.Tal-L.Portisch
Montreal 1979
French Defence

1 e4 e6 2 d4 d5 3 Nd2 c5 4 exd5 exd5

Today, 4 ... Qxd5 is a popular alternative to the isolani position.
5 Bb5+ Bd7
Karpov tested Korchnoi in their Candidates’ final/World Championship match with 5 ...
Nc6 6 Ngf3 Bd6 7 dxc5 Bxc5 and Korchnoi held draws each time. I just wonder if chess
history would have been different if Korchnoi had stuck with the French, rather than his
disastrous outings with the Dragon and Petroff’s Defence.
6 Qe2+

Question: What is the purpose of this check?

Answer: The idea is to induce ... Be7 and then play dxc5, followed by Nb3, stubbornly
hanging on to the c5-pawn.
6 ... Be7
6 ... Qe7 is also played, but doesn’t make much sense to me, since the side with the
isolani should generally avoid too many swaps.
 7 dxc5
This pawn isn’t as free as it looks since Black generates with a development lead, and
White expends energy to win the pawn, and even more to try to hang on to it.
7 ... Nf6 8 Nb3 0-0 9 Nf3 Re8
Intending ... Bxc5.
10 Be3 a6

Portisch was probably influenced by Korchnoi’s difficult ending against Karpov in their
Baguio World Championship match of 1978, which went 10 ... Bxc5 11 Nxc5 Qa5+ 12 Qd2
Qxb5 13 0-0-0 b6?! (Black still looks okay to me after 13 ... Bg4) 14 Nxd7 Nbxd7 15 Kb1
Ne4 16 Qd3! (Karpov isn’t about to fall for 16 Qxd5?? Nc3+, winning the queen) 16 ...
Qxd3 17 Rxd.3 White’s position follows the Shaolin Kung Fu principle: the best victory is
when you overcome an opponent without need of force, which should be your last resource.
Karpov achieved an ideal Tarrasch isolani ending, with all the winning chances. However,
Korchnoi, one of the all-time great defenders, held the draw here as well, A.Karpov-
V.Korchnoi, 22nd matchgame, Baguio City 1978.
11 Bd3
The most combative move, keeping pieces on the board. 11 Bxd7 helps Black’s
development and here the players agreed to a very likely prearranged draw, M.Tal-
A.Karpov, Montreal 1979.
11 ... Ba4
 Question: This move looks very odd. What is its purpose?

Answer: Portisch clears d7 for his b8-knight, followed by ... Rc8, after which he wins
back his temporarily sacrificed pawn.
12 Nfd4 Nbd7 13 0-0-0!
Sharper than 13 0-0 Bxb3 14 Nxb3 Nxc5 where Portisch achieved dynamic equality,
B.Larsen-L.Portisch, Tilburg 1978.
13 ... Nxc5
I prefer 13 ... Bxb3! 14 Nxb3 a5! when Black begins to generate serious queenside play,
Y.Yakovich-Y.Shereshevsky, USSR 1985.
14 Nf5 Bf8 15 Nxc5
Following 15 Qf3?! Nxd3+ 16 Rxd3 Bb5 17 Rdd1 g6 18 Nfd4 Bd7 (threat: ... Bg4) 19
h3 h6! I like Black’s attacking chances, since ... a5 and ... a4 are in the air.
15 ... Bxc5 16 Qf3
Tal escapes the e-file pin and his queen creeps closer to Portisch’s king.
16 ... Bxe3+ 17 Nxe3
For once, Tal behaves himself, avoiding the structurally risky 17 fxe3!? which opens the
f-file, seizes control over d4 and keeps his knight on the advanced post, at the cost of
weakening his structure.
17 ... Rc8
Houdini suggests 17 ... d4!? 18 Ng4 Nxg4 19 Qxg4 Rc8 20 Kb1 Qf6 and Black looks
okay here as well.
18 Bf5 Rc5!?
18 ... Qc7 is playable, when White must avoid 19 Bxc8?? Rxe3! 20 Qf5 Re2 21 Rd2
Rxd2 22 Kxd2 Ne4+ and Black wins.
 19 Rd4
Tal plans to double rooks and add pressure to d5.
19 ... Bc6!?
This is a clear case of provocation. Portisch – clearly not a believer in the thought: the
more grandiose your plan, the less likely it’s going to work – dares Tal to play 20 b4,
chasing the almost trapped c5-rook.
20 b4!?

Each side wants war, but for different reasons. Portisch is willing to sacrifice material,
while Tal is willing to do the same with structure. Okay, the pretence of positional play is
over. Tal is certainly cavalier about the strategic obligation which this weakening thrust
entails. The idea is to chase Black’s rook to oblivion, at the cost of weakening the pawn
front around his king. For those unwilling to risk, 20 Rhd1 is the move.
20 ... Rb5
The only move. 20 ... Rc3? 21 Kb2 costs Black an exchange for insufficient
compensation.
21 a4!
More accurate than 21 Bd3?! Rxb4! 22 Rxb4 d4 23 Qf4 dxe3 24 fxe3 Qa5. This time
Black keeps queens on the board and gets promising play for the exchange, due to White’s
shaky king and even shakier pawns.
21 ... Rb6!
One way to fool an opponent is to feign meek submission for the moment, while secretly
preparing an insurgency. Portisch hedges his idea with a cautionary interpolation. The idea
is to goad White’s pawn to a5, after which Black generates play with ... b6!.

Question: Can Portisch sacrifice an exchange

for an attack with 21 ... Rxb4? 22 Rxb4 d4?
 Answer: This line would be promising for Black, if not for the dirty trick 23 Qd1!
which removes queens from the board. After 23 ... dxe3 24 Qxd8 Rxd8 25 fxe3 Bxg2 26
Rd1 Re8 27 e4! g6 28 Rxb7! Black lacks full compensation for the exchange.
22 a5 Rb5
22 ... Rxb4?! 23 Rxb4 d4 24 Qd1 dxe3 25 Qxd8 is the same old story. Black falls short
of full compensation for the exchange when White forces queens off the board.
23 Qf4!
The position is a patchwork of conflicting ideas. Threat: Bd3. In this way Tal avoids ...
Rxb4! and ... d4 counterplay. As we have seen in the notes earlier, White should avoid the
greedy 23 Bd3 Rxb4! 24 Rxb4 d4 when Black gets the promising queens-on-the-board
version of the exchange sacrifice.
23 ... b6!
Portisch issues a little warning to Tal: when you face a multifaceted idea and enjoy a
small victory (Tal will win the exchange), just remember: you merely destroyed a copy of a
copy. The original idea still lives. In this instance Portisch inflicts as much damage as he
can before he loses the exchange.
24 Bd3
The b5 outlier is surrounded. Black should generate enough play for the exchange.
24 ... bxa5 25 Bxb5 axb5!
More accurate than 25 ... Bxb5 26 Rhd1 axb4 27 Nxd5 Nxd5 28 Rxd5 Qa5 29 Qd4
when Rd8 is coming and I don’t believe in Black’s full compensation.
26 Rhd1! axb4 27 Kb2!
Avoiding 27 Rxb4?! Qa5 28 Kb2 Ne4! 29 Rb3 Ra8 when Black’s attack is serious.
27 ... Qc8
27 ... Qa5 is met with 28 Ra1.
 28 Rxb4 Ne4

Exercise (planning): How can Tal seize the initiative?

29 Rd3?!
Refugees crowd the streets of White’s queenside. This is not the moment for the safe,
slow path. After this unnecessary defensive move, Black gets decent compensation for the
exchange, with one pawn and an active position.
Answer: Tal should return the exchange. Black struggles after 29 Rxd5! Bxd5 30 Nxd5
f5 (or 30 ... Nc5 31 Rxb5 Ne6 32 Qd6 Nd4 33 Qc7!; White is up a clean extra pawn, with
the initiative) 31 f3 Nf6 (31 ... Nc5 32 Rxb5 leaves White up a pawn with all the play) 32
Nxf6+ gxf6 33 Rxb5 Qa6 34 Qb4 Ra8 35 Rb8+ Rxb8 36 Qxb8+ Kg7 37 Qc7+ Kg6 38 c4
and White has decent chances to win the pawn up queen and pawn ending.
29 ... Nc5?!
Slightly inaccurate. Black should play 29 ... Qe6 30 Rbd4 Nc5 31 Ra3 Na4+ 32 Kc1
Qe7 33 Rb4 f6 with dynamic equality.
30 Ra3 Na4+
This move cuts off the a3-rook.
31 Kc1 Bd7?
Shifting into overdrive isn’t of much use when you only have a few drops of gas in the
tank. Correct was 31 ... Qe6.
32 Qd6
It is in the nature of a siege that day by day, the defender grows weaker and weaker
through slow starvation. Except in this case, I’m not sure who is the attacker and who is
defending. This is the second best continuation. Komodo suggests 32 Nxd5! allowing 32 ...
Re2 33 Qc7! and Black must trade queens, since 33 ... Qe8?? is met with 34 Raxa4! when
the a4-rook can’t be taken due to Rb8.
32 ... Bc6
After 32 ... Re6 33 Qxd5 h6 34 Qd2 White is on his way to consolidation.
33 Rd3?
Tal continues with uncharacteristic caution. Once again, correct was 33 Nxd5! Bxd5 34
Qxd5 Nc3 (or 34 ... Re2 35 Qd3 and everything is covered) 35 Qd2 Ne4 36 Qd4. White’s
powerfully centralized major pieces ensure a safe king, while pocketing an extra exchange.
33 ... h6 34 Rf4! Qe6?!
The ending isn’t holdable. Portisch still had practical chances with the line 34 ... Qa8 35
Qa3 Qb8 36 Qb4 Nb6 37 Rg4! Nc4 38 Nxc4 dxc4 39 Qc3 g6 40 Rh3 and White is winning.
35 Qxe6 fxe6

Question: Why did you give Portisch’s last move a dubious

mark. It looks to me like Black has a pretty good ending.

Answer: It appears as if Portisch generated good play for the exchange, since his centre
rolls with ... e5 next. Tal’s next move dispels this illusion.
36 Ng4!
One of the greatest psychological shocks in chess is when our opponent assaults us at a
point we considered invulnerable. Tal intends to blockade with Ne5 next, which would be
intolerable. Portisch’s problem: Black is unable to support an ... e5 push without
overextending.
36 ... e5
This is a case of the cure being worse than the original complaint. Portisch couldn’t
stomach 36 ... Re7 37 Rg3 Kh7 38 Rf6! (threat: Rxh6+, followed by Nf6+ and Rg8 mate) 38
... Kh8 39 Rf8+ Kh7 40 Rd8! Be8 (40 ... e5? loses to 41 Rd6 Be8 42 Rxh6+) 41 Ne5 with a
dominating position.
37 Rf5
The players fight for control of e5, like rival siblings squabbling for a toy.
37 ... Nc5
37 ... e4 38 Rg3! Re6 39 Nf6+ Kh8 40 Nxd5 wins.

Exercise (combination alert): How gratifying when our

once risky idea (Tal’s 20 b4!?) is verified with a winning
position later on. Find a few accurate moves and it becomes
clear that Black’s game is destitute of all forms of counterplay.

Answer: Double attack/overloaded defender.

38 Rc3! Bd7 39 Rxe5! 1-0
“Choose your words carefully. I have a literal mind and would not like to see you suffer,
through an unfortunate ‘misunderstanding’,” threatens White’s rook. Black pawns continue to
vanish without a trace, as if behind a magician’s veil.

Game 48
M.Tal-J.Van der Wiel
Moscow Interzonal 1982
English Opening

1 c4 Nf6 2 Nc3 e6 3 Nf3 b6 4 e4

 We enter one of those hybrid English/something else/no-name hybrids. Instead, 4 g3 Bb7
5 Bg2 gets us into a Queen’s Indian, or a Hedgehog.
4 ... Bb7 5 Bd3

Question: Doesn’t this block White’s queenside development?

Answer: White’s development isn’t hampered by this move. The idea is borrowed from
the Kopec Sicilian, where White plays Bd3, c3, Bc2 and then d4. The only difference in this
case is that Tal’s c-pawn is pushed two squares instead of one.
5 ... c5
5 ... d6 prevents e5 ideas, and if 6 Bc2 c5 7 d4 cxd4 8 Nxd4 when we reach a Hedgehog
structure.
6 0-0
Tal prefers rapid development over immediate confrontation with 6 e5 Ng4 7 h3 Bxf3 8
Qxf3 Nxe5 9 Qxa8 Nxd3+ 10 Ke2 Nf4+ 11 Kf1 Nc6 12 Qxd8+ Kxd8. For the exchange,
Black got a pawn, control over d4, and a development lead, V.Korchnoi-L.Polugaevsky,
Evian 1977.
6 ... Nc6 7 e5!?
Tal decides to risky an overextended e-pawn, rather than get bogged down into
something like 7 Bc2 e5 with a closed, manoeuvring game to follow, M.Nikolov-
I.Georgiadis, Rethymnon 2011.
7 ... Ng4 8 Be4
The e-pawn is safe, for now.
8 ... Qc8
Alternatives:
a) 8 ... f5 9 exf6 Nxf6 10 Bxc6 Bxc6 11 d4 Bxf3 12 Qxf3 cxd4 13 Nb5 Rc8 14 Bf4 a6
15 Nxd4 Rxc4 16 Rfd1 and White’s development lead compensated for his missing pawn,
J.Speelman-N.Short, Barcelona 1989.
b) 8 ... Qb8 fails to win the e-pawn after the simple 9 Re1 f6 10 exf6 Nxf6 11 Bxc6
Bxc6 12 d4 cxd4 (12 ... Bxf3?! 13 Qxf3 cxd4 14 Nb5, intending Bf4 next, gives White a
dangerous development lead) 13 Nxd4 Qb7 14 Bg5 Be7 15 Nxc6 Qxc6 16 Qd3 0-0 17
Rad1, Le Quang Liem-A.Riazantsev, French Team Championship. 2012. Black stands only a
shade worse after 17 ... Rf7.
9 d3!?

A new move enters Tal’s field of vision. He virtually begs his opponent to eat the e5-
pawn. Tal hopes to improve upon a Korchnoi-Polugaevsky game (either that, or Tal forgot
the book move and made up his own; when we are unable to remember our book lines, we
are tormented by the thought that our own treacherous brain stole valuable knowledge from
us) 9 Re1 d6 10 exd6 Bxd6 11 d4 cxd4 12 Nb5 Bc5 13 Nfxd4 h5, V.Korchnoi-
L.Polugaevsky, Buenos Aires 1980. I slightly prefer White after 14 Nxc6 Bxc6 15 Nd6+
Bxd6 16 Qxd6 Bxe4 17 Rxe4 Nf6 18 Rd4 Qd7 19 Bf4.
9 ... Ngxe5!?
Van der Wiel is unintimidated by his legendary opponent and decides to grab the pawn
and test Tal. His move is made with the thought: a cause, no matter how noble, seems
unimportant to those who suffer on an empty stomach. Only when we are well fed, do we
turn our attention to career, love and philosophy. Far safer is 9 ... f5 (Black steers clear of
the pawn grab, the way I do when I’m driving my shrimpy Prius hybrid and get out of the
way of my son’s lumbering four-wheel drive pickup truck) 10 exf6 Nxf6 when I don’t think
Black is so badly off, I.Stohl-V.Salov, Groningen 1983.
10 Nxe5 Nxe5
 Question: I see a relatively closed position, with zero
compensation for the pawn. Is this sacrifice sound?

Answer: It takes a certain level of wilful cognitive disassociation for me to award some
of Tal’s sacrifices with an ‘!?’, rather than the ‘?!’ many of them truly deserve, when
evaluated objectively. Essentially his sacrifices were simultaneously weak/strong, if we
factor in the practical chances he earned from the chaos. In this instance, however, I feel that
his sacrifice was 100% sound and perhaps should even have been declined by Van der
Wiel. On the surface it feels as if White’s initiative is so barely there and insubstantial, that
it feels as if it’s a shadow of a shadow. Now watch the transformation in just a few moves.
11 f4! Nc6?!
Perhaps already a slight inaccuracy. When under attack, every piece exchanged tends to
help the defender. That is why Black should have tossed in 11 ... Bxe4! 12 Nxe4 Nc6 13 f5.
I think this is a better version, than what Black got in the game. After 13 ... exf5 14 Rxf5 Be7
15 Qg4 g6 16 Nf6+ Bxf6 17 Rxf6 0-0 18 Bh6 d6 19 Qf4 Nd4 20 Qxd6 Re8 21 Raf1 Qd8!
Black looks okay to me. Now White should avoid 22 Qxd8 Raxd8 23 Rxf7? Nf5!, as this
interference/double attack wins material.
12 f5!
 White plays f5 before Black does it himself.

Question: What did Tal get for the pawn?

Answer: Tal’s move is saturated with dire implication. For just one pawn, Tal got the
following:
1. A dangerous development lead.
2. Open e- and f-file potential.
3. White’s advanced f-pawn ensures that Black’s king will be dangerously unsafe in
either centre, or kingside. The problem is Black is a long way from castling queenside.
4. White’s queen may swoop into g4 or h5.
Conclusion: White has more than enough for the pawn.
12 ... g6?!
The idea is to force a resolution on f5 (which it doesn’t!). There will be a heavy price to
pay on the kingside dark squares for this weakening move. Instead:
a) 12 ... Be7? was suggested by Tal. White looks almost winning after 13 Qh5 Rf8 14
Bg5 Qd8 15 Bf4.
b) 12 ... h5! is logical, keeping White’s queen out of both g4 and h5. This may be
Black’s best shot at survival.
13 Bg5!!
Tal ignores Black’s threat and simply continues to develop.
13 ... gxf5!?
It isn’t a great idea to call an opponent’s bluff, if your opponent isn’t bluffing. I can’t
fault Van der Wiel for avoiding the admittedly nauseating looking 13 ... Bg7 14 Nb5
(threatening a fork on d6) 14 ... 0-0 15 f6 Bh8 and I have a bad feeling Black is getting
mated at some point.
 14 Bxf5!
“God observes you at all times – as I do,” the bishop warns Black’s king. Tal can also
delay his idea a move with 14 Qh5! Bg7 15 Bxf5! with a powerful attack, no matter how
Black responds.

Exercise (critical decision): Should Black accept Tal’s offer of the piece and take
on f5? Or should he decline by playing 14 ... Be7? Both lines are trouble for Black,
but he is far better off in one, rather than the other. Which one should he play?

14 ... Be7?
Now Black suffers without even being up a piece.
Answer: He had to accept and pray after 14 ... exf5!. In this instance, facing the
inevitable is a far better course for Black than postponing the inevitable. Black’s last move
is made with the philosophy: if you are in hell, then it’s a waste of energy to plead with the
devil for a way out. After 15 Qe2+ Ne7 16 Rae1 Rg8 17 Bxe7! Rxg2+ 18 Qxg2 Bxg2 19
Bh4+ Be4 20 Nxe4! fxe4 21 Rxe4+ Be7 22 Rxe7+ Kd8 23 Rfxf7 Kc7 (freedom! Black’s
king wanders about with the air of a lost child at the mall, in search of his mother) 24 Bg3+
Kc6 25 Rxh7 a5 the defenders are undermanned, with many wounded, yet Black may
survive his ordeal.
15 Qh5!
White attackers crowd the kingside as tightly as cucumbers in the pickle jar.
15 ... Bxg5
Acceptance is too late. 15 ... exf5?? 16 Rae1 Kd8 17 Rxf5 when Nd5 and Rxf7 are
threatened, and Black is hopelessly tangled.
16 Qxg5 Ne7
Threatening ... Rg8, rather than:
 a) 16 ... exf5?? 17 Rae1+ Kf8 18 Qh6+ Kg8 19 Nd5 (threatening mate on f6) 19 ... f6 20
Nxf6+ Kf7 21 Nh5 Qf8 22 Rxf5+ Kg8 23 Rxf8+ Rxf8 24 Qg7 mate.
b) 16 ... Qd8 17 Qg7 Rf8 18 Nd5! with a decisive attack. If 18 ... exd5?? 19 Rae1+ Ne7
20 Bg4! threatening mate, starting with Rxf7!, and if 20 ... d6 21 Be6! fxe6 22 Rxf8+ Kd7 23
Rxd8+ Rxd8 24 Qh6 when Black is down too much material.
17 Be4
Threat: Nb5.
17 ... Bxe4
17 ... Rg8 is simply met with 18 Qf6, menacing f7.
18 Nxe4 Qc6

Exercise (combination alert): Continue Tal’s attack. If you

correctly plot the focal point, you will find the idea.

19 Rxf7!
Answer: Annihilation of defensive barrier. It’s never a good sign when our opponent’s
moves match our suspicions. Now f7 is the breakthrough point.
19 ... Kxf7 20 Qf6+!
This in-between check (zwischencheck?) is crucial to the combination, since either
retreat is fatal for Black’s king.
20 ... Kg8
We note that Black’s king spends an awful lot of time apologizing, for a king, or if 20 ...
Ke8 21 Qxh8+ Kf7 22 Qxh7+ Ke8 23 Qh8+ Kf7 24 Rf1+ Nf5 25 Qf6+ Ke8 26 g4 and wins.
21 Qxe7
Threat: Nf6 mate. The blade skims by the black king’s right cheek, leaving an angry, red
memento of his e7 sister’s visit.
 21 ... Rf8

Exercise (planning): Black’s exhausted position has the worn, wrinkled look of an
old leather sofa, in dire need of either upholstery, or the junkyard. Find Tal’s finish:

Answer: Removal of a key defender.

22 Rf1! 1-0
Black’s king is helpless after 22 ... Rxf1+ 23 Kxf1 when he must give up his queen to
stave off mate.
Answer no.2: A student found 22 Qg5+! Kf7 23 Rf1+ Ke8 24 Nf6+! Kd8 (24 ... Rxf6 25
Qxf6 Rg8 26 Qf7+ wins) 25 Ng8+!.

Game 49
M.Tal-B.Abramovic
Moscow 1982
Queen’s Gambit Declined

1 c4
Tal’s attacking set-up is standard issue for the London System. Compare it to this game:
1 d4 Nf6 2 Nf3 e6 3 Bf4 c5 4 c3 d5 5 e3 Be7 6 Nbd2 0-0 7 Bd3 b6 8 Ne5 Bb7 9 Qf3! Nbd7
10 Qh3! Re8?! 11 Ndf3 a6?? (11 ... Nxe5 was necessary) 12 Nxf7! Qc8 (12 ... Kxf7 is met
with 13 Qxe6+! when taking on e6 walks into 14 Ng5 mate, but if 13 ... Kf8 14 Ng5, forcing
mate) 13 Qxe6 Rf8 14 Nd6+, C.Lakdawala-R.Bruno, San Diego (rapid) 2010. I have won
about ten very similar games to this one, using exactly the same attacking themes. When we
repeat a combinational pattern, we are the thief who walks by the place where he
successfully pulled off the last job – this time in broad daylight – just for the joy of the
memory.
1 ... e6 2 Nc3 d5 3 d4 Be7
The idea behind this move is to deprive White’s bishop of early g5 options.
4 cxd5
Tal enters a version of the Exchange Queen’s Gambit Declined. 4 Nf3 is also played
here.
4 ... exd5 5 Bf4
This is a game which warms my London System-loving heart.
5 ... Nf6 6 e3 0-0 7 Bd3 b6!?
In my opinion the Queen’s Indian style set-up isn’t a good fit for this position, since
White gets attacking chances. 7 ... c6 and 7 ... c5 are played more often.
8 Nf3 c5 9 Ne5!

Question: Doesn’t this knight jump simply encourage a swap?

Answer: White made the move with two ideas:

1. It clears the way for Qf3 and possibly later Qh3, when White’s queen enters the attack
zone.
2. White isn’t afraid of a future ... Nc6 (or ... Nbd7) and ... Nxe5, since White may
recapture with his d-pawn, chasing away the f6-knight, a key defender.
9 ... Bb7 10 Qf3!
As usual, Tal chooses the kingside as his preferred arena for settling disputes.
10 ... Nbd7 11 0-0 a6?!
Necessary was 11 ... Qc8 12 Bf5 Nxe5 13 Bxe5 Qc6 with a manageable disadvantage.
12 Qh3!
I have played this attacking plan so often from similar positions in London and Exchange
Slav, that when I see someone else play it, it feels like looking in a mirror. I discovered
early on that you can only control that which is already known and understood (I’m not
interested in a fair fight in crazy complications!). This attacking theme is so worked out in
detail, that it’s a bit like the domesticated pet who is almost part human, having lost its
ability to be an animal in the wild.
12 ... g6!
The only move, probably made with great hesitation, since the display of weakness
psychologically marks you as prey. Black must weaken the dark squares around his king, and
also give White a line-opening hook with f5 later on:
a) 12 ... Nxe5? 13 dxe5 Bc8 14 g4! Bxg4 15 Qg2 and White wins a piece.
b) 12 ... Bc8?? is met with the overloading shot 13 Nxd5!.
c) 12 ... h6? 13 Bf5! (even stronger than the immediate sacrifice on h6) 13 ... Nxe5 14
dxe5 Ne4 15 Nxe4 dxe4 16 Bxh6 is decisive.
d) 12 ... Re8?? walks into the trap, which three unfortunates fell for in my database: 13
Nxd7 Qxd7 14 Bxh7+ Kf8 15 Bf5 Qd8 16 Qh8+ Ng8 17 Bh7 Bf6 18 Bg5! winning.
13 Bh6

Question: Is this really a tempo gain if Tal chased

Black’s rook to a good square on an open file?

Answer: Tal’s last move was not played to gain a tempo. Instead, he got:
1. By chasing the rook to e7, Black’s f7-pawn is weakened.
2. By playing Bh6, Tal cleared the way for his next move, f4, which backs up his e5-
knight, and also prepares an assault down the f-file with a future f5.
13 ... Re8 14 f4
 Black faces the Stonewall from hell and one look tells us that Black isn’t likely to
survive White’s coming kingside attack.
14 ... Bd6
This has the appearance of a wasted move, in a position which certainly can’t afford it.
Even at this early stage, it feels to me as if Black’s defensive resources are close to
exhausted:
a) 14 ... Bf8 15 Bg5 (stronger was to eliminate the defender of the dark squares around
Black’s king with 15 Bxf8! Nxf8 16 f5, with a dangerous attack brewing around Black’s
king) 15 ... Bg7 16 Qh4 Qc7 17 f5! cxd4, H.Hernandez Carmenates-Y.Gonzalez Vidal, Santa
Clara 2007, and if 18 Nxd7! Qxd7 19 Na4! Qxa4 20 Bxf6 Bxf6 21 Qxf6 Qc6 22 Qxd4 Qc5
23 Rae1 Qxd4 24 exd4 with only a minimal edge for White in the ending, due to Black’s
slightly bad bishop.
b) 14 ... Qc7?! 15 f5! cxd4 16 Nxd7! Qxd7 17 exd4 Bc6, J.Bartos-L.Jinova, Zdar nad
Sazavou 2007. White has a winning attack after 18 Bg5 Ne4 19 Bxe7 Rxe7 20 Bxe4 dxe4 21
d5 Bb5 22 Qh6! when the threat of f6 is decisive.
c) 14 ... b5 may be Black’s best hope to survive: for example, 15 f5 cxd4 16 exd4 Nxe5
17 dxe5 Bc5+ 18 Kh1 Rxe5 19 Bf4 Qe7 20 Bxe5 Qxe5 21 Rae1 Ne4 22 Bxe4 dxe4 23 fxg6
hxg6 24 Rxf7! Kxf7 25 Qh7+ Ke8 26 Qxb7 Ra7 27 Qc8+ Kf7 28 Nxe4 Be7 29 Qc1 Kg7.
15 Bg5
Threat: Nxd7.
15 ... Be7?!
The scariest way to lose is when your position gets worse and worse, without an easily
discernible cause. The problem with this move is that it takes the pressure off Tal’s e5-
knight, allowing him f5. Komodo suggests desperation mode by allowing White his intent
with 15 ... b5!? 16 Nxd7 Qxd7 17 Bxf6 Qxh3 18 gxh3 Rxe3 19 Bc2. Black only has two
pawns for the piece and should avoid taking the third with 19 ... Rxh3? 20 Rad1 c4 21 Rde1
with a dominating position and extra material for White.
16 f5
Such promising Stonewall-style attacks are so self-explanatory, that even a normally
attack/initiative-challenged player like your writer plays them with surprising competence.
16 ... cxd4

Exercise (combination alert): That which has been accomplished

isn’t nearly as important as that which needs to be accomplished.
Black’s arrow-bristled defensive shield can’t hold any longer and
Tal has two promising attacking ideas in the position. Find one of them:

Answer: Annihilation of defensive barrier.

17 fxg6!
The weakness of f7 and h7 are the central facts, from which all other arguments flow.
Tal’s sacrifice comes with all the subtlety of a bell’s clanging.
Answer no.2: Also crushing was 17 Nxf7! (this intrusive knight reminds me of my
gigantic Akita, Emma, who is actually closer to a nanny, than a dog; each morning she barks
at daybreak to get me up, when my preference would be to get up at the crack of noon) 17 ...
Kxf7 18 exd4 (White has time for this calm recapture, since Black’s king isn’t going
anywhere) 18 ... Nf8 19 fxg6+ hxg6 20 Qh6 (threat: Rae1, followed by Bxg6+!, and then
Qh7+) 20 ... Ke6 21 Bxg6 Nxg6 22 Rxf6+! Bxf6 23 Qxg6 Rf8 24 Re1+ Kd7 25 Qf5+ Kc7
(or 25 ... Kc6 26 Re6+ Kc7 27 Bf4+ Kc8 28 Rc6 when rook and queen deliver the dreaded
double checkmate; the red-faced black king is unable to look his sister in the eye, and
pretends to study the ceiling with minute interest) 26 Bf4+ and wins.
17 ... Nxe5
Well, why not? You can only be damned once. I wouldn’t describe this knight as a
coiled, rattlesnake-like defender, striking without conscience. To attempt to use an
optimistic piece grab to eradicate an otherwise miserable position, is similar to attempting
to bring a terminally ill patient back to health with positive affirmations. 17 ... hxg6 18
Nxg6! fxg6 19 Bxg6 is crushing. If 19 ... dxc3 (or 19 ... Nf8 20 Bxf6 Bxf6 21 Bxe8 dxc3 22
Qg4+ Bg5 23 Bf7+ Kg7 24 h4 Bc8 25 Qg3 and White wins) 20 Rxf6 Nxf6 21 Bxf6 forces
mate.
18 Bxf6!
Clearing away the final defender of h7.
18 ... Bxf6
Instead:
a) 18 ... fxg6 19 Qe6+ Nf7 20 Bxd4 Rf8 21 Bxg6 forces mate.
b) 18 ... hxg6?? makes life easy for White after 19 Qh8 mate. “It seems that your code of
chivalry has been rendered obsolete,” mocks the queen.
19 Qxh7+ Kf8
The king feels as if he is about to cry. So he does.
20 exd4
Tal goes after the sole defender of f7.
20 ... Nxg6 21 Bxg6 fxg6 22 Rf3

In preparation for Raf1. Tal, at the moment is uninterested in table scraps, and doesn’t
bother taking the hanging b7-bishop with 22 Qxb7.
22 ... Re6 23 Raf1 Rc8
23 ... Bc6 is met with 24 Qxg6.
24 Qxb7 1-0
What? A pragmatic decision from Tal? At this stage Black playing on is as pointless an
exercise as sweeping the floor after your house burned to the ground: 24 ... Kg8 is met with
the overload trick 25 Rxf6! Rxf6 26 Rxf6, winning a piece.
 However, it’s shocking that Tal rejected the flashier continuation 24 Qxg6! Rcc6 25 Rh3,
which leaves Black defenceless.

Game 50
M.Tal-V.Korchnoi
Montpellier Candidates 1985
Sicilian Defence

Everyone has a boogie man out there. You know the one I mean: the guy is rated about the
same as you, yet he beats you to a bloody pulp every time you play him! Korchnoi one of the
greatest defensive players of all time, normally gave Tal fits. Here is a game where Tal
overcame his great rival.

1 e4 c5 2 Nf3 d6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 Nc6 6 Bg5 e6 7 Qd2 Be7 8 0-0-0 0-0 9
Nb3 a5

Openings like the Richter-Rauzer Sicilian are so venomous, there is the danger that the
snake will thrash about wildly and accidentally bite itself. If you recall, Koblencs played 9
... Qb6 on Tal in the first chapter of the book.

Question: Isn’t Black’s last move strategically

reckless? Korchnoi just gave Tal control over b5.

Answer: It does appear on the surface that Black’s ill-equipped army roars defiance by
banging their weapons on their shields, the way Mel Gibson’s army did, when facing
Edward Longshanks’ troops, in Braveheart. Black also gets something in return: the
strategic threat ... a4 and ... a3, creating a puncture on the dark squares around White’s king.
 Question: But why can’t White simply play a4 to prevent this problem?

Answer: That is exactly what Tal did, and it doesn’t come without cost. First, it allows
Black a future ... Nb4, where it may be inconvenient for White to move his c3-knight and
then play c3. Secondly, Korchnoi’s ... a5 induced a weakening in the pawn front around
Tal’s king.
10 a4
Tal isn’t about to allow the black a-pawn to run amok, and fixes it on its square.
10 ... d5 11 Bb5!
I think this is White’s only shot at an edge. 11 exd5 got Karpov nothing after 11 ... Nxd5
12 Bxe7 (not 12 Nxd5?! Bxg5 and Black stands slightly better) 12 ... Ncxe7 13 Nb5 Bd7 14
Be2 Nf5 15 N3d4 Nxd4. The players agree to a draw in a position in which I already prefer
Black, A.Karpov-G.Kasparov, World Championship (Game 37), Moscow 1985.
11 ... dxe4?!
A mistake must be recognized, before it can be corrected. I don’t like this move, which
opens the game when lagging in development. Black is better off with 11 ... Nb4 12 e5 Nd7
13 Bxe7 Qxe7 14 f4 b6 15 Rhe1 Nc5 16 Nd4 Bd7 and maybe White holds a touch of an
edge due to his grip on d4 and b5, Y.Balashov-A.Khalifman, Minsk 1986.
12 Qxd8 Bxd8!?
Perhaps the idea is to transfer the bishop to either c7 or b6. Black also fails to equalize
after the more natural 12 ... Rxd8, M.Steadman-T.Doyle, New Zealand 1997. Now 13
Rxd8+ Bxd8 14 Bxc6 bxc6 15 Nxe4 also looks like a tough ending for Black.
13 Rhe1 Na7
Korchnoi fears further damage to his structure with Bxc6.
14 Bc4
Missing 14 Nxe4!, and if 14 ... Nxb5 (or 14 ... Nxe4 15 Bxd8 Nxf2 16 Rd2 Nxb5 17
axb5 Ng4 18 Nc5! and White has pressure, since 18 ... Nxh2? can be met with 19 Be7 Re8
20 Na4! f6 21 Nb6 and wins, as 21 ... Ra7 is met with 22 Bc5) 15 Nxf6+ gxf6 16 Bh6 Re8
17 Re3! Kh8 18 axb5 Bb6 19 Rg3 Bxf2 20 Rf3 Bb6 21 Rxf6 when Black is in deep trouble.
14 ... h6 15 Bxf6! gxf6
There is much hand-wringing and teeth gnashing in Black’s kingdom, mainly due to his
inability to develop his queenside pieces. After 15 ... Bxf6 16 Nxe4 Be7 17 f4 Nc6 (17 ...
b6 is met with 18 c3 Bb7 19 Rd7 Bh4 20 g3 Bc6 21 Rxa7 Rxa7 22 gxh4 Bxa4 23 Nd4
which also favours White) 18 Bb5 Black experiences serious problems completing his
queenside development.
16 Nxe4 f5 17 Nd6 Bc7 18 g3
 Exercise: (combination alert): White’s position looks promising, yet it
seems like one where Tal can only build, rather than actually strike. His
sensitive eye spotted a geometric anomaly, leading to a winning position. If
you find his next move, profits await White. How would you continue here?

18 ... b6?
A small taste of freedom only whets your appetite for more. Black doesn’t have the time
to slowly unravel his queenside, and his last move is a bit like attempting to parry a
belligerent bar drunk’s intended blow with reasoned argument. Korchnoi should have tried
18 ... Rd8! 19 Nxc8 (or 19 Nb5 Nxb5 20 Bxb5 b6 21 Bc6 Rb8 22 Nd4 Bd6 23 Nb5 Be7 24
Rxd8+ Bxd8 25 Rd1 Be7 and White stands better, yet has no easy inroad into Black’s well-
fortified camp) 19 ... Rdxc8 20 Be2 Bb6 21 f4 Rc7 22 Rd6 Nc6 23 Red1. White retains
pressure, but with no imminent breakthrough.
Answer: Clearance.
19 Nxf5!!
“Soon,” murmurs the knight, as he plots petty revenges against his many enemies. Tal
decides that his position, having reached its apex, isn’t going to get any readier, so now is
the time to strike.
19 ... exf5 20 Bd5! Be6!
The bishop’s lips move silently, reciting a prayer. This concession (it’s psychologically
difficult to return an insult with silence), can’t be described as morale building for Black,
yet it is necessary. Korchnoi, a master of defence, makes a pragmatic lesser-evil choice,
shorn of sentiment, correctly rejecting 20 ... Rb8? 21 Re7. The seventh rank isn’t such an
uncrossable barrier after all. Black must return the piece with a completely lost position.
21 Bxa8 Rxa8
 After 21 ... Bxb3 22 cxb3 Rxa8 23 Rd7 Rc8 24 Kb1 Bb8 25 Rc1 Re8 26 Rb7 Kf8 27 f4
Kg7 28 Rxb6 Black’s pieces are completely tied down to the defence and he won’t survive.
22 Nd4

Question: Why is Tal’s combination so great?

He only got a rook and pawn for two bishops.

Answer: I gave Tal the double exclam, not for the depth of his combination, but for his
ability to assess the aftermath. An assessment is often a subjective mental construct, and it’s
possible for two players of even strength to come to opposite conclusions of a single
position. But not in this one. Komodo assessment: ‘+1.63’. The problem is Black’s light-
squared bishop must move, after which White’s rook infiltrates e7, and black pawns begin
to fall at an alarming rate.
22 ... Bd5
Alternatively:
a) 22 ... Re8 23 Nxe6! fxe6 24 Rd7 Bb8 25 Rb7 Nc6 26 Rxb6 Nd4 27 c3 Nf3 28 Rexe6
Rxe6 29 Rxe6 (threat: Re8+ and Rxb8) 29 ... Ba7 30 Kd1! is hopeless for Black.
b) 22 ... Bc4 is probably Black’s best shot at resistance: 23 b3 Bd5 24 Nb5 Nxb5 25
Rxd5 Nc3 26 Rd7 Rc8 27 Ree7 Bb8 28 Rxf7 Be5 29 Rxf5 Re8 30 Rff7 and Black can play
on, but is still losing in the long run.
23 Re7
Our opponent’s progress makes our own failures all the more painfully apparent. Tal’s
threats spring up like unwanted mushrooms (the kind you can’t put on a pizza!) in my
backyard.
23 ... Rc8
Or 23 ... Bd8 24 Nxf5 Be6 25 Re8+ Kh7 26 Ne3 Nc6 27 Rd6 Be7 28 Rxe6 Rxe8 29
Rxc6 with two extra pawns and an easy win.
24 Nb5! 1-0
What a blessed feeling when we get an accounting for past losses against us.

Black’s tangled defenders act the way some of our not-so-wonderful friends do when
they agree to help you, but actually have no intention of doing so: 24 ... Nxb5 25 Rxd5 Nd6
(after 25 ... Na7 26 Rdd7 White’s rooks gaze at c7, a7 and f7 hungrily, like beggars outside
a bakery window) 26 Rxc7! Rxc7 27 Rxd6 Rb7 28 Rxh6 once again leaves White up two
pawns.

Game 51
M.Tal-J.Speelman
Reykjavik 1988
Pirc Defence

1 e4 d6 2 d4 g6 3 Nf3
This is an annoying move order for Modern Defence players.

Question: Why?

Answer: Because White can keep Black guessing about possible transpositions to
queen’s pawn lines by tossing in c4.
3 ... Bg7 4 Be2 Nf6
This move transposes to Pirc. I have faced 4 ... a6 5 Nc3 Nd7 6 Bf4 e6!? 7 Qd2 b5 8 a3
h6 9 0-0 Ne7 10 Rae1 g5!? with the messy kind of position the higher-rated player generally
likes, when playing down, J.Choi-C.Lakdawala, San Diego (rapid) 2013. I was more afraid
of 5 c4 where I may be entering a King’s Indian set-up with an early ... a6, which I may not
need.

Question: How would you have continued if your opponent had played this way?

Answer: I don’t play the straight KID, so I would have probably attempted to muck
things up with something like 5 ... Bg4 6 Nc3 Nc6 7 Be3 e5, and if 8 d5 Bxf3 9 Bxf3 Nd4
with a strange KID hybrid position, C.Milton-C.Lakdawala, San Diego (rapid) 2006.
5 Nc3
I already mentioned this earlier in the chapter, in his win against Petrosian, that I found
odd Tal’s attraction to this very quiet white set-up against the Pirc and Modern Defences.
5 ... 0-0 6 0-0 c5
The Schmidt Benoni is one of Black’s most ambitious tries against White’s set-up. This
move was played in Korchnoi’s ill-fated final game of his 1978 Baguio match loss to
Karpov, when the score stood at a nail-biting 5-5, with many draws. Also played are 6 ...
c6, 6 ... Bg4, 6 ... Nc6, 6 ... Nbd7, and 6 ... a6.
7 d5!
This is the only real path to a white advantage, since White gets a limp Dragon if he
allows ... cxd4. Korchnoi famously accused GM Ray Keene, one of his seconds in the
Baguio match, of “selling” 7 d5! to Karpov.
7 ... Na6
 Question: What is Black’s intent behind this move?

Answer: Black’s source of counterplay stems from queenside counterplay based on a ...
b5 push. So Black usually plays ... Nc7, ... Rb8, ... b6, ... a6, and if necessary ... Bd7 to
ensure his ... b5 break. The obvious trouble with this plan is its slowness. If handed all this
time, White generally enforces a successful e5 central break. Sometimes Black actually
fianchettos his light-squared bishop to take the sting out of White’s intended e5 break.
8 Re1
Alternatives:
a) 8 Bf4 Nc7 9 a4 b6 10 Re1 Bb7!?.

Question: Why did Black just fianchetto his bishop into a pawn wall on d5?

Answer: This can be considered an anti-strategy. The idea is to pile up on d5, to make
White’s e5 break difficult. If White does nothing, then Black can generate play by later
striving for an ... e6 break. 11 Bc4 (reinforcing d5) 11 ... Nh5 12 Bg5 Nf6 13 Qd3 a6 14
Rad1 Rb8 was A.Karpov-V.Korchnoi, World Championship (Game 32), Baguio City 1978.
Korchnoi found himself caught in Karpov’s icy cold embrace and lost the game. White at
this point could have played an immediate 15 e5!.
b) 8 h3 Nc7 9 a4 b6 10 Re1 Bb7 11 Bc4 a6 12 Qd3 e6!? 13 dxe6 Nxe6 14 Bd2 Qc7 15
Rad1 Rfd8 was the later A.Karpov-E.Torre, Brussels 1987. Black would be in trouble if
Karpov had found 16 Ng5! Bc8 17 Bc1 when 17 ... h6 could be met with 18 Nxe6 Bxe6 19
Bxe6 fxe6 20 e5 dxe5 21 Qxg6 with a winning structural advantage for White.
8 ... Nc7 9 Bf4?!
This may be inaccurate, due to Speelman’s next move. 9 a4! suppresses the ... b5 break,
and is White’s path to an edge.
9 ... b5!

Speelman alertly seizes upon Tal’s inaccuracy. He kind of gets a mild Benko Gambit
idea, without having sacrificed a pawn.
10 Nxb5
Instead:
a) 10 a3 b4 11 axb4 cxb4 12 Na2 a5 13 Bc4 Nh5 14 Bc1 also looks fine for Black, who
forced several of Tal’s pieces to retreat, while he gained valuable queenside space.
b) 10 Bxb5?? fails miserably to 10 ... Nh5! when White loses material to the dual threat
... Nxf4, and also ... Bxc3, followed by ... Nxb5.
10 ... Nxe4 11 Nxc7 Qxc7 12 Bc4 Nf6
 Black swapped away a wing pawn for White’s more valuable e-pawn.

Question: Does White get pressure on Black’s backward e7-pawn?

Answer: A little, but not enough to worry Black, who can cover the pawn as many times
as White can attack it.
13 h3 Re8 14 Rb1 a5
White can’t really make any use of b5.
15 Qd2 Qb6!?
15 ... a4 looks more consistent.
16 Re3
Tal prepares to load up heavy pieces on e7, rather than 16 a4! Ba6 17 b3 Rab8 18 Bxa6
Qxa6 19 c4 when White will be tied down to b3, just as much as Black is to e7. I still
slightly prefer Tal’s side here, due to his extra space.
16 ... Ba6 17 Bxa6 Qxa6 18 Rbe1 Kf8
After 18 ... Qc4! 19 Rxe7 Rxe7 20 Rxe7 Qxa2 Black stands better, since if 21 Bxd6?!
Bf8! 22 Bxc5 Bxe7 23 Bxe7 Qxb2 24 c4 Qxd2 25 Nxd2 Nd7. I think Black’s extra exchange
and passed a-pawn will beat White’s two connected passed pawns.
19 Ng5!?
A spy’s job is to observe, while remaining unseen.

Question: This almost looks like a random move. What does Tal’s knight threaten?

Answer: A good salesperson sells you what your imagination wants, and not what he or
she actually has. In this instance Tal pedals paranoia for Black’s king. The knight isn’t really
threatening anything – yet. This is the kind of move which made Tal’s opponents nervous. He
hopes to provoke the weakening ... h6. And if Black refuses, as in the game continuation,
then Tal keeps his eye open for future sacrificial possibilities on f7 or e6. I haven’t really
decided if overconfidence (Tal’s last move!) is a shortcoming or a strength in chess. For Tal,
it was both a curse and blessing. His veins flowed with the blood of a natural con-artist,
who lives for the joy of fleecing his marks.
19 ... Qb7
Now Black’s position drifts like smoke carried by the vagaries of the wind. It’s wrong
to allow White c4. The edge may be swinging to Black if he finds 19 ... Qb5! 20 b3 (20 c3?!
is met with 20 ... c4 which applies pressure to d5) 20 ... Qb4! 21 Rd3 Qxd2 22 Bxd2 c4!
with queenside pressure for Black.
20 c4 Qb4 21 Qe2
Tal, of course, isn’t about to allow a queen trade when he has a potential attack brewing.
21 ... h6?
Black should try 21 ... Nxd5! 22 cxd5 Qxf4 23 Ne6+! (knight fork/attraction/pin) 23 ...
fxe6 24 Rf3 Qxf3 25 Qxf3+ Kg8 26 Re2. Black got a rook, bishop and pawn, plus a compact
position for the queen, and my feeling is he should hold the draw without too much
difficulty.

Exercise (combination alert): It’s one thing to be outmanoeuvred or outplayed

by an opponent’s superior ideas, and quite another when we bring on our own
suffering deliberately. On his last move – which is akin to looking into the wrong end of
the telescope to see the planets, moon and stars – Black provokes Tal
into a sacrifice on f7. Should Tal play it, or is there something better?

Answer: The sacrifice works.

22 Nxf7!
 White’s earlier Ng5 clearly passed the threshold of a warning, and now turns,
uncomfortably for Black, into a direct threat. Oh, the bliss, when our once abstract concept
gels into a reality on the board.
22 ... Kxf7
The stunned black king’s sceptre slips from his nerveless fingers, clattering to the floor.
Abundance can sometimes be deadly. If you don’t believe me, then just ask Marie
Antoinette. This extra piece is just borrowed money, which must be repaid later with
extortionate levels of interest.
23 Rb3!
A crucial element in the combination, which allows Tal’s rook to swing to the seventh
rank. Speelman may have expected 23 Rxe7+? Rxe7 24 Qxe7+ Kg8 when White doesn’t
have enough for the piece.
23 ... Qa4
When you weigh your options, it’s not such a good sign that you have only one.
24 Qe6+
An unfulfilled ambition is not the same thing as a broken vow. This wins. Tal may (I
stress the ‘may’!) have missed an easier version with 24 Rb7! Kg8 25 b3 Qa3 26 Bc1.
Black’s queen is trapped, without sufficient compensation after 26 ... Qb4 27 Rxb4 axb4 28
Qc2 Kf7 29 Re3 and Black won’t survive. Then again, Tal may have seen this line and
rejected it in favour of his own continuation, since in that one he gets to attack with greater
gusto.
24 ... Kf8 25 Rb7 Qxc4
Black is unlikely to survive 25 ... Ng8 26 Re3!.

Exercise (combination alert): Continue Tal’s attack:

 Answer: Clearance. The black defenders, partially awake, partially dressed, and only
partially armed, are unable to offer meaningful resistance.
26 Bxd6!
White must be careful not to play the rote 26 Rxe7? Rxe7 27 Qxe7+ Kg8, after which
Black’s king has excellent chances of slipping away.
26 ... Ng8
Naturally not 26 ... exd6?? 27 Qf7 mate.
27 Re3!
The final attacker is lifted in.
27 ... Bf6 28 Rf3 Kg7

Exercise (combination alert): Proceed with Tal’s attack:

Answer: Removal of a key defender.

29 Bxe7! Rxe7
29 ... Bxe7 30 Rf7+ Kh8 31 Qe5+ (this mates faster than 31 Qxg6) 31 ... Nf6 32 Rbxe7
with a forced mate in five moves.
30 Rxe7+ Nxe7
30 ... Bxe7 31 Rf7+ Kh8 32 Qxg6 and Black has nothing better than 32 ... Qc1+ 33 Kh2
Qf4+ (“The weak thrive on petty revenge” thinks White’s king) 34 Rxf4 Rf8 with an easy
win for White.
31 Qxf6+
Komodo prefers 31 Rxf6.
31 ... Kg8 32 Qf7+ Kh8 33 Qxe7 Qxd5
 Exercise (combination alert): I assure you the solution to this problem
won’t tax the majority of readers. White to play and force the win:

Answer: Seize the seventh rank with the rook – a solution so obvious that it doesn’t
even deserve an exclam.
34 Rf7 1-0
Black must give up his queen to avoid mate.

Game 52
M.Tal-I.Rogers
San Francisco 1991
French Defence

1 e4 e6 2 d4 d5 3 Nd2 dxe4 4 Nxe4 Nd7 5 Nf3 Ngf6 6 Nxf6+ Nxf6 7 Bd3 c5 8 dxc5 Bxc5
9 Qe2 0-0
Perhaps Black should dodge the coming Bg5 pin by first playing 9 ... Qc7 10 0-0 (10
Bg5 Bb4+! 11 Nd2 Nd5 is considered okay for Black, A.Fedorov-V.Yandemirov, Sochi
1997) 10 ... 0-0 11 Bg5 b6! 12 Ne5 (12 Bxf6 gxf6 13 Qe4 f5 and now White should avoid
14 Qxa8? Bb7 15 Qxa7 b5! when he doesn’t get enough for the queen) 12 ... Bb7 13 Bxf6
gxf6 14 Bxh7+ Kxh7 15 Qh5+ and the game ends in perpetual check, V.Ivanchuk-A.Karpov,
Las Palmas 1996.
10 Bg5 Qa5+
10 ... h6 is more commonly played.
11 c3 Be7 12 Ne5
Cutting off ... Bd7 and making it difficult for Black to complete his development.
 12 ... b6?!

Independent thinking and heresy are usually only a sliver apart. With this move Rogers
entertains an almost insane idea. When the word ‘courage’ is used by an annotator, it is
generally in the context of a virtue. But courage, when pushed to recklessness, can be the
source of a player’s undoing. In my one game against GM Rogers, I was probably winning,
and he made just such a bizarro move, confusing the daylights out of your easily confused
writer, and won (it’s a cruel thing to make a positional player calculate, since it forces us to
do our least favourite thing in the world: maths!). Rogers, whose optimism outruns the
position’s truth, hopes that a line involving great dangers, generally implies great rewards,
as well. He hopes to complete his development, while luring Tal into misadventures on c6.
The flaw with the plan is that White is by no means obliged to comply with an immediate
jump into c6.
Correct was 12 ... Rd8, keeping White’s edge to tolerable levels.
13 Bb5!
In order to solve a riddle, it first becomes necessary to speak the language in which it is
asked. It isn’t easy to fathom the position’s inner workings, and such a level of
complications is frost on a window, which occludes vision of the world outside. If either
side overlooks a single minuscule detail, he risks complete ruin. Now Nc6 becomes a very
real threat, whereas if:
a) 13 Nc6? (the bait always looks delicious to the fish; the problem, of course, is the
hook hidden within the worm) 13 ... Qxg5 14 Nxe7+ Kh8 15 Nxc8 Qxg2! (zwischenzug) 16
0-0-0 Raxc8 and I don’t believe in White’s compensation for the pawn, even with the open
g-file, since g7 is too easy to defend.
b) 13 Bxf6 Bxf6 14 Nc4 Qg5 15 Qe4 (double attack) 15 ... Qf5! and White must avoid
16 Qxa8?? Qxd3. Black wins, since 17 Ne3 is met with the crushing 17 ... Ba6.
13 ... a6
 13 ... Bb7?? 14 a4! a6 15 Bxf6 gxf6 16 b4 Bxb4 is met with 17 Qg4+ Kh8 18 cxb4 and
Black’s queen is trapped.
14 Bc6 Ra7 15 b4!
Black’s queen is close to trapped.
15 ... Qa3 16 Qc2!

“If this yapping dog makes another sound, then beat it,” orders the queen, as she points to
her a3 sister. White’s pieces crowd around Black’s queen like the faithful entering a temple
on prayer day. Threat: Bc1, and also Nc4, in both cases trapping Black’s wayward queen.
16 ... Nd5!
Altered circumstances forces us all to bend our previous intent. Rogers finds Black’s
only method of playing on, with his pincushion-like position still absorbing all of Tal’s
pokes.
17 Bxd5! exd5
17 ... Bxg5?? is met with 18 Nc4, trapping the queen.
18 Bc1 Bf5!
In times of desperation, ‘later’ is synonymous with ‘never’. The only move. By some
form of arcane magic, Black’s defence appears to remain erect, but this is still an illusion
upon an illusion. “I was beginning to feel confident for the first time in the game, but Tal
hadn’t finished with the tactical tricks,” wrote a dismayed Rogers.
19 Bxa3!
It may feel as if Tal’s unfulfilled desires continue to stock up. If we look closer, we see
that in reality, Tal sells each concession dearly. He may not have been successful in his
queen-trapping project, but he didn’t come away empty handed either.
19 ... Bxc2 20 Nc6
Temporary relief returns to apprehension for Black.
20 ... Bf6!
 The best practical chance. 20 ... Rc7? loses to 21 Nxe7+ Rxe7+ 22 Kd2 Ba4 23 b5,
winning the exchange, in a simplified position.

Exercise (critical decision): Black’s move is made with the thought: you can
only be put to death once. Making a move like this one, Black becomes the condemned
death row inmate, who is now convicted of a second murder,
and a second death sentence. Should Tal play 21 Nxa7, which wins a piece?
Or should he first toss in 21 Kd2? Only one line leads to a white advantage.

Answer: He should take a7 immediately.

21 Nxa7!
The zwischenzug 21 Kd2? fails to 21 ... Rc7 22 b5 Rfc8!. Now White must avoid 23
Kxc2? (correct is 23 Bd6 Rd7 24 Be5 axb5 25 Bxf6 Rxc6 26 Bd4 Bf5 with a likely draw)
23 ... axb5 which regains the lost material advantageously, due to White’s insecure king
position.
21 ... Bxc3+
After 21 ... Ba4 22 Bb2 Re8+ 23 Kf1 Ra8 24 g3 Rxa7 25 Kg2 White should eventually
consolidate his extra exchange.
22 Ke2 Re8+ 23 Kf3 Re6
Tal feared 23 ... d4! intending ... Be4+ and ... Ra8. The comp says White still wins after
24 Rac1 Be4+ 25 Kg3 Ra8 26 Rhd1! h5 (26 ... Rxa7?? loses to the dirty back rank trick 27
Rxd4) 27 Rxc3! dxc3 28 Rd7 when Black was unable to regain his lost piece.
24 Rhc1! Be4+ 25 Ke2
White’s king escapes.
25 ... d4
25 ... Bxg2+ 26 Kd3 d4 27 Rxc3! dxc3 28 b5 and White’s once stranded knight escapes.
 26 f3 1-0
This is no shared misfortune, as there is no way to trap White’s knight, since b5 is
coming.

Game 53
M.Tal-V.Akopian
Barcelona 1992
Sicilian Defence

A Buddhist nun once told me: “If you begin to see yourself as important, then just look up at
the stars at night to get a perspective of just how small and insignificant we all are.” Old age
comes for us all, in order to mock us. Tal, far from his past glory, was nearing the end of his
life when he played this game. Yet even in his almost unrecognizably weakened state, we
catch a glimpse of his old tricks.

1 e4 c5 2 Nf3 Nc6 3 Bb5 d6 4 0-0 Bd7 5 Re1 Nf6 6 c3 a6 7 Ba4

I’m surprised that Tal avoided the messy gambit 7 Bxc6 Bxc6 8 d4 Bxe4 9 Bg5.

Question: This line seems tailor made for Tal? Why would he reject such a gambit?

Answer: Tal wanted complications, but not the kind where a theoretically well-prepared
opponent may have out-worked him in pre-game prep. His opponent, the then reigning world
junior champion, was (and is) known for his excellent opening preparation.
Instead, 7 Bf1 Bg4 is another path for White.
7 ... c4!?
The idea is to suppress d4. More common is 7 ... b5 8 Bc2 Bg4.
 8 d4!?
This move allows Black to equalize rather easily. Probably White’s only shot at an edge
lies in the line 8 Bc2, intending to pressure the c4-pawn with b3, Na3 and Qe2.
8 ... cxd3 9 Bg5 e6 10 Qxd3 Be7 11 Bxf6!?
A move made with the thought: strip a person of his theoretical armour, and he becomes
someone else. Tal was the merchant who dealt with likelihoods and possibilities. In his
younger days, he nearly always got the better of the bargain when he haggled with such
unbalancing moves.
11 ... gxf6

Question: Is Black okay here after the damage to his structure?

Answer: It’s debatable if Black’s structure has been damaged at all, since he gets
greater central influence and an open g-file. This looks to me like a rather lame version of a
Richter-Rauzer, for Tal’s side, since White has a pawn, rather than a knight on c3 and his f-
pawn, normally on f4, sits limply on f2. So I think Black is certainly no worse here.
12 Bxc6!?
Calmer is 12 c4 Ne5 13 Bxd7+ Qxd7 14 Qe2 Rc8 15 b3 Rg8.
12 ... Bxc6
I would be more inclined to take back with the b-pawn.
13 c4
For now, Tal’s knights seem to be holding their own against Black’s bishop-pair. The
trouble with such positions is sooner or later, the game tends to open, favouring the bishops.
13 ... 0-0
Intending ... Kh8, ... Rg8 and at some point a possible ... f5 break.
14 Nc3 Kh8 15 Rad1 Rg8 16 Qe3 Qf8
 Black can maybe load up major pieces along the g-file.
17 Nd4 Rc8
17 ... Bd7 18 a4 keeps the bishops behaving for now.
18 f4!?
I would have played 18 Nxc6 Rxc6 19 b3 Qg7 20 g3 Bd8 21 Ne2, with an
approximately balanced game.
18 ... Bd7!
I think it’s a good idea to retain both bishops, now that Tal committed to the loosening
f4.
19 b3 Bd8!
Threatening to swing over to b6.
20 Nf3
Cutting off ... Bb6.
20 ... b5!
The game continues to open.
21 Qa7?!

Now White’s counterplay trickles through his fingers like water. Here we see Tal’s old
weakness: an overestimation of his chances. Sometimes a little fear is actually a sign of
wisdom, since fear tends to keep us alive.
21 ... Bc7!
Target: f4. Also promising was 21 ... Ba5! 22 Qxd7 Bxc3 23 Re2 d5! 24 Qa7! (not 24
exd5? Qc5+ 25 Kf1 Rc7 when unwanted suitors line up to woo White’s queen; this move
traps the queen for insufficient compensation after 26 dxe6 Rxd7 27 exd7 Ba5) 24 ... dxc4.
Black stands clearly better since 25 Qxa6? Qc5+ 26 Kf1 Rcd8 is busted for White, as ...
Rd3 is coming.
22 Qxa6 bxc4 23 b4
 This smacks of desperation, but after 23 bxc4 d5 24 Kh1 f5! (threat: ... Qg7, with
simultaneous attacks on g2 and c3) 25 cxd5 Qg7 26 Rg1 Qxc3 27 d6 Qa5 28 Qe2 Bb6 29
Ne5 Rg7! 30 Nxd7 Bxg1 31 Rxg1 (forced; 31 Kxg1?? Qa7+ picks up the loose knight) 31 ...
Qd8 32 Ne5 Qxd6 White is unlikely to survive.
23 ... Qg7
Threatening mate. 23 ... d5! is promising for Black, as 24 exd5 is met with 24 ... Qxb4
25 dxe6 Bxe6 26 Nb5 Bxf4 when Black’s bishops come alive, while after 24 Qb7 Qh6 I
don’t think White will survive Black’s bishops.
24 g3 d5!
Uncovering an attack on f4. Akopian’s move is more dynamic than 24 ... f5 25 e5 d5.
25 exd5
Clearly Tal is a long way from pleased in this souring position. There is nothing better.
25 e5 is met with 25 ... Qg4 with a double attack on f3 and f4: 26 Kf2 fxe5 27 Nxe5 Bxe5
28 Rxe5 Qh3 and White is lost. If 29 Kg1? Rxg3+! 30 hxg3 Qxg3+ 31 Kh1 Qxc3 leaves
White’s king hopelessly exposed.
25 ... Bxf4 26 Kf2!
After 26 Ne2 Be3+ 27 Kh1 Qh6 White won’t survive the attack.

Exercise (critical decision): Tal has been completely outplayed, but be careful: when a
position grows more complex, the side who stands better – obviously Black, in this instance
– has more to lose. This is that disorienting moment when the ocean sucks its own water
from the beach, just moments before the tsunami. Black’s candidate moves: a) 26 ... Bxg3+
sacrificing a piece to expose White’s king. b) 26 ... f5 opening an attack on White’s loose
c3-knight. c) The calm 26 ... exd5, opening the game for his bishops. Only one of the
possibilities leads to a decisive advantage for Black. Which?
 26 ... f5?
An old template shouldn’t be applied to new circumstances. When we miss a decisive
combination, we are the person who is unexpectedly bequeathed a fortune by a distant
relative. But then we irrationally reject the money, thinking: “The best things in life are
free.”
Answer: Correct was ‘c’, 26 ... exd5! (principle: open the game when you own bishops
versus knights):
a) 27 gxf4?? Qg2+ 28 Ke3 Rge8+ and White is slaughtered.
b) 27 Nxd5?? Bxg3+ 28 hxg3 Qxg3+ 29 Ke3 Rc6! 30 Qa7 (inept white defenders fly
like dust from a beaten rug) 30 ... Bg4 when it becomes obvious that White’s king has no
prayer.
c) 27 Ne2 Bh6 28 Rxd5 Be6 29 Rc5 Ra8 30 Qc6 Rxa2 31 Rxc4 (31 Nfd4 Bd2 32 Rd1
Qh6 33 Kg2 Be3 is decisive) 31 ... Bxc4 32 Qxc4 Rga8 with an extra exchange and a
sustaining initiative.
Instead, Black’s worst choice is variation ‘a’: 26 ... Bxg3+? 27 hxg3 Qxg3+ 28 Ke2
Qg2+ (Black’s main problem is that he is unable to open the e-file with 28 ... exd5?? 29
Qxf6+ Rg7 30 Rg1 Re8+ 31 Kd2 Qf2+ 32 Kc1 Reg8 33 Rxg7 Rxg7 34 Rg1 and wins) 29
Ke3 Rce8 and the trouble is White can clog things with 30 d6 e5 31 Rg1 Qc2 32 Qa3. White
should survive this attack and consolidate.
27 gxf4! 27 Qxc3
Black regained his piece and looks like he is just winning, until we see Tal’s next move.
28 Qd6!

Attacking Black’s bishop, while leaving open possibilities of a queen swap on e5. Black
no longer stands even an iota better.
28 ... Ba4 29 Rd4!
Interfering with the black queen’s coverage of the a1-h8 diagonal.
 29 ... Rg7?
In time pressure, those normally steady neurotransmitters in our brain speed up. The
result? We either play brilliant, inspirational chess, or, we go haywire and short circuit, as
is the case here. 29 ... Rg6 was necessary to maintain the balance: 30 dxe6 Rcg8! 31 Ng5 f6!
32 Nf7+ Kg7 33 Ng5! (threat: Qe7+ and Qxh7 mate) 33 ... Kh8! (33 ... fxg5?? 34 Qe5+ Kh6
35 fxg5+ Kh5 36 Rh4+! Kxh4 37 Qxc3 picks off Black’s queen) 34 Nf7+ is drawn by
perpetual.
30 dxe6 Bc6
Threatening mate on the move. This looks scary, but Tal has everything under control.
31 Ng5!
A lock snaps shut on Black’s attack, as Tal clogs the g-file. Akopian may have counted
on 31 Re3? Rg2+! 32 Kxg2 Qxe3 33 Qe5+ Qxe5 34 fxe5 fxe6 35 Kf2 Bd5 36 Rd1 c3 37 a3
Ra8 38 Ra1 c2 39 Nd4 Rxa3 40 Nxc2 Rb3. White’s pawns are loose and he is unlikely to
save himself in the ending.
31 ... Rxg5
Alternatively:
a) 31 ... fxe6 and now the simplest is 32 Qe5 (threat: Rd8+ and Qxc3; the comp likes the
inhuman 32 Rxe6 and wins after a few impossible-to-find moves) 32 ... Re8 33 Rd6 Qxe5
34 Rxe5 Bd7 35 Nxe6 with a lost ending for Black, due to White’s two connected queenside
passed pawns, plus extra pawn.
b) 31 ... f6 allows 32 Qd8+!, and if 32 ... Rxd8 33 Rxd8+ Rg8 34 Nf7+ Kg7 35 Rg1+.
Now Black only gets two spite-blocks with 35 ... Bg2 36 Rxg2+ Qg3+ 37 Rxg3 mate.

Exercise (combination alert): How did Tal force the win of heavy material?

Answer: Queen trap/discovered attack/zwischenzug. Black has no defence to the coming

Rd8+ and Qxc3.
 32 Qe5+!
“Knock, knock!” sings the queen. Neither Black’s king or queen, however, are in any
mood to say “Who’s there?”. I hope nobody chose 32 fxg5?? Qf3+ 33 Kg1 Qg2 mate.
32 ... Rg7
When we get swindled, we find ourselves playing the lead role in a comedy, which is
funny to everyone else but us.
33 Rd8+ Rxd8 34 Qxc3
One can never bow low enough to satisfy White’s queen. The rest is easy.
34 ... f6 35 e7 Ra8 36 Qxf6 Be4 37 Rg1 Rxa2+ 38 Ke1! 1-0
Black faces two (three, if you count under-promotion with e8R+!) mate threats.
Index of Complete Games
Aronson.L-Tal.M, USSR Championship, Moscow 1957
Averbakh.Y-Tal.M, USSR Championship, Riga 1958
Bleks.I-Tal.M, Latvian Championship, Riga 1952
Bobotsov.M-Tal.M, Varna 1958
Botvinnik.M-Tal.M, World Championship (Game 17), Moscow 1961
Botvinnik.M-Tal.M, World Championship (Game 6), Moscow 1960
Donner.J-Tal.M, Zurich 1959
Gurgenidze.B-Tal.M, USSR Championship, Moscow 1957
Kapengut.A-Tal.M, Latvian Championship, Riga 1965
Polugaevsky.L-Tal.M, Riga Interzonal 1979
Ragozin.V-Tal.M, Riga (simul) 1948
Smyslov.V-Tal.M, USSR Team Championship, Moscow 1964
Spassky.B-Tal.M, Montreal 1979
Spassky.B-Tal.M, Tallinn 1973
Spassky.B-Tal.M, USSR Championship, Riga 1958
Tal.M-Abramovic.B, Moscow 1982
Tal.M-Akopian.V, Barcelona 1992
Tal.M-Averbakh.Y, USSR Team Championship, Riga 1961
Tal.M-Botvinnik.M, World Championship (Game 1), Moscow 1960
Tal.M-Botvinnik.M, World Championship (Game 11), Moscow 1960
Tal.M-Botvinnik.M, World Championship (Game 19), Moscow 1960
Tal.M-Botvinnik.M, World Championship (Game 7), Moscow 1960
Tal.M-Botvinnik.M, World Championship (Game 8), Moscow 1961
Tal.M-Dvoretsky.M, USSR Championship, Leningrad 1974
Tal.M-Fischer.R, Bled/Zagreb/Belgrade Candidates 1959
Tal.M-Fuster.G, Portoroz Interzonal 1958
Tal.M-Gligoric.S, Bled/Zagreb/Belgrade Candidates 1959
Tal.M-Hecht.H, Varna Olympiad 1962
Tal.M-Keres.P, Tallinn 1973
Tal.M-Koblencs.A, Riga 1957
Tal.M-Korchnoi.V, Montpellier Candidates 1985
Tal.M-Larsen.B, 10th matchgame, Bled 1965
Tal.M-Larsen.B, 6th matchgame, Bled 1965
Tal.M-Letelier Martner.R, Havana 1963
Tal.M-Lisitsin.G, USSR Championship, Leningrad 1956
Tal.M-NN, Riga (simul) 1958
Tal.M-Panno.O, Portoroz Interzonal 1958
Tal.M-Petrosian.T, USSR 1974
Tal.M-Portisch.L, 4th matchgame, Bled 1965
Tal.M-Portisch.L, Biel Interzonal 1976
Tal.M-Portisch.L, Montreal 1979
Tal.M-Rantanen.Y, Tallinn 1979
Tal.M-Rogers.I, San Francisco 1991
Tal.M-Simagin.V, USSR Championship, Leningrad 1956
Tal.M-Smyslov.V, Bled/Zagreb/Belgrade Candidates 1959
Tal.M-Speelman.J, Reykjavik 1988
Tal.M-Stean.M, Hastings 1973/74
Tal.M-Tolush.A, USSR Championship, Leningrad 1956
Tal.M-Tolush.A, USSR Championship, Moscow 1957
Tal.M-Van der Wiel.J, Moscow Interzonal 1982
Tal.M-Vasiukov.E, USSR Championship, Kiev 1964
Thorbergsson.F-Tal.M, Reykjavik 1964
Tseshkovsky.V-Tal.M, Sochi 1970