Morphy _Move by Move - Zenon Franco [2016].OCR Opt

Zen6n ranco

'

move by move

001

e

ranco

Morphy m,ove by move

EVERYMAN CHESS www.eve ryma11c:hess.c0im

002

Translated by Phil Adams First published in 2016 by Gloucester Publishers Limited, Northburgh House, 10 Northburgh Street, London EC1V 0AT Copyright © 2016 Zenón Franco Translated by Phil Adams The right of Zenón Franco 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: 978-1-78194-362-5 ISBN epub: 978-1-78194-363-2 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., Central Books Ltd, 50 Freshwater Road, Chadwell Heath, London , RM8 1RX. All other sales enquiries should be directed to Everyman Chess, Northburgh House, 10 Northburgh Street, London EC1V 0AT email: [email protected]; 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.

For Yudania Everyman Chess Series Chief advisor: Byron Jacobs Commissioning editor: John Emms Typeset and edited by First Rank Publishing, Brighton. Cover design by Horatio Monteverde.

003

About the Author Zenón Franco is a Grandmaster from Paraguay, now living in Spain. He represented Paraguay, on top board, in seven Chess Olympiads, and won individual gold medals at Lucerne 1982 and Novi Sad 1990. He’s an experienced trainer and has written numerous books on chess. Also by the Author: Anand: Move by Move Rubinstein: Move by Move Spassky: Move by Move Test Your Chess

004

Contents About the Author Bibliography Introduction 1 Paul Morphy’s Style of Play 2 Games Against Family Members 3 The First American Chess Congress 1857 4 The Visit to London and the Match Against Löwenthal 5 The Journey to Paris and the Match Against Harrwitz 6 The Match Against Anderssen 7 Retirement Paul Morphy’s results Index of Complete Games

005

Bibliography Books Morphy’s Games of Chess, Philip W. Sergeant (Dover Publications 1957) My 60 Memorable Games, Robert J. Fischer (Simon & Schuster 1969) My Great Predecessors, Part 1, Garry Kasparov (Everyman Chess 2003) Paul Morphy: A Modern Perspective, Valeri Beim (Russell Enterprises 2005) Periodicals New in Chess Magazine 1984-2015 Databases Paul Morphy: Genius and Myth, Thomas Eichhorn, Rainer Knaak & Karsten Müller (Chessbase 2003) Mega Database 2015 (Chessbase 2015) Chess Engines Houdini Rybka

006

Introduction It is a pleasure to write about a chess player whose career was as unusual as that of Paul Morphy, who was considered without question to be the best in the world in his day. He became number one, not through a ‘normal’ process of experience and learning, but almost overnight. In fact his career was the shortest of all those who can be considered the number one of their time. Between his first game in the first USA Championship in 1857 and the last game of his match against Adolf Anderssen, there was a space of only one year and three months. Even so, Morphy’s superiority over the rest was overwhelming. Morphy is possibly the easiest to understand of the players who have been the best of their time. As Max Euwe pointed out, his style was based on three basic elements: 1. rapid development of the pieces; 2. control of the centre; and 3. open lines. It seems simple, yet it can’t be so, since the other masters of the day couldn’t manage it when they confronted him; only Morphy grasped it. He was ahead of his time. One of his rivals Henry Bird said: “When one plays with Morphy the sensation is as queer as the first electric shock, or first love, or chloroform, or any entirely novel experience”. Morphy read all the literature that was available in those days, but this didn’t amount to much; that alone wasn’t enough. As a standard of comparison, let’s keep in mind that Robert Fischer said in 1968 that part of his own mastery was due to his having read around a thousand books and having taken the best out of each of them. All the masters agree that Morphy’s beautiful combinations were based on the fact that he was the first positional player, even though that sounds paradoxical. His teachings were perfected later by Steinitz, but Morphy was the first, the most revolutionary. And we are left with an obvious question: How did he discover it? How did he learn? As with two other world number ones whose careers I have had the good fortune to study, Akiba Rubinstein and Boris Spassky, Morphy’s written legacy is almost non-existent. Unfortunately, he left little or no evidence of how he chose his moves, how he evaluated the positions, etc. Thus we have to let his games speak for him. The Structure of this Book In the first chapter I examine Morphy’s style of play, with testimony from various world champions, along with my own conclusions. Then some practical examples of his style are given. The rest of the book features a selection of his games, ordered chronologically, with some biographical data. Extracts from games and some supplementary games are added. Zenón Franco Ponteareas, September 2016 With special thanks to Jonathan Tait for his very useful suggestions and improvements.

007

Chapter One Paul Morphy’s Style of Play I believe that brilliance is something that all we chess players associate with the name of Morphy, from the very first time that we see some of his games, which as a rule takes place not long after learning to play. Those spectacular sacrifices of material leading to mate seemed magical to us. Lasker was one of several champions who emphasized the positional basis of Morphy’s game; he wrote: “Morphy discovered that the brilliant move of a master is essentially conditional not on a sudden and inexplicable realization, but on the placing of the pieces on the board. He introduced the rule: brilliant moves and deep winning manoeuvres are possible only in those positions where the opponent can be opposed with an abundance of active energy.” Capablanca commented something in the same vein: “The greatest stylist was Morphy. He did not look for complicated combinations, but he also did not avoid them, which really is the correct way of playing ( ... ) His main strength lay not in his combinative gift, but in his positional play and general style. Morphy gained most of his wins by playing directly and simply, and it is this simple and logical method that constitutes the true brilliance of his play, if it is considered from the viewpoint of the great masters.” Alekhine pointed out that: “ ... at that time, finally, when Anderssen was alive, and with brilliance alone it was hardly possible to surprise anyone. The strength, the invincible strength of Morphy – this was the reason for his success and the guarantee of his immortality!”. Morphy’s invincible strength, in the opinion of Alekhine was his “deeply considered positional play, primarily of an aggressive character”. Euwe agreed with the world champions who preceded him: “Many people consider Morphy to be the greatest combinative player of all time. But Morphy owed his remarkable achievements not only to his admittedly tremendous combinative talents, but also to the fact that he was the first perfect positional player – in so far as one can speak of perfection in a human being. Moreover, he had excellent technique. “In the maturity of his positional insight, Morphy was half a century ahead of his time. While others racked their brains for the right procedure, Morphy simply knew what to do. Unlike his contemporaries who played for complications on the off-chance of finding something, he struck only when the position was ripe. When he did strike, the blow was hard and often settled the game outright”. Botvinnik emphasized that: “To this day Morphy is an unsurpassed master of the open games. Just how great his significance was is evident from the fact that after Morphy nothing substantially new has been created in this field.” What Botvinnik said was confirmed by Adolf Anderssen himself, who admitted that he could not compete with Morphy in open positions – none other than Anderssen, the protagonist of the ‘Immortal’ and the ‘Evergreen’ games. Fischer was not backward in giving praise: “Morphy was perhaps the most accurate chess player who ever lived. He had complete sight of the board and never blundered, in spite of the fact that he played quite rapidly, rarely taking more than five minutes to decide a move. (His opponents, in those days before chess clocks, often took hours.)” Smyslov noted that: “He played with inspiration, without striving to penetrate into the psychology of the opponent; he played, if one can express it so, ‘pure chess’. His harmonious positional understanding and deep intuition would have made Morphy a highly dangerous opponent even for any player of our times.” As Kasparov pointed out, almost all world champions (Botvinnik is the exception) saw their own game reflected in Morphy’s, their own vision of the game and their own strength; at times they seemed to be talking about themselves. For this to be so general is something surprising and unique. Kasparov explained Morphy’s strength in this way: “I think that it was a combination of a unique natural talent and brilliant erudition. His play was the next, more mature stage in the development of chess. Morphy had a well-developed ‘feeling for position’, and therefore he can be confidently regarded 008

as the ‘first swallow’ – the prototype of the strong 20th century grandmaster.” Fischer also mentioned the only fallibility shown by Morphy: “Perhaps his only weakness was in closed games like the Dutch Defence. But even then, he was usually victorious because of his resourcefulness.” There can be little doubt that if Morphy had continued to play he would also have excelled in closed positions. The five-time World Champion Viswanathan Anand was asked who he would play against, considering the whole history of chess, he answered: “I’d probably go for Morphy. He was the first player who made everything seem effortless. In fact I was shocked to see a collection of all his games, because usually we see only the best ones in which the chess looks like effortless. If you saw all his games you realized, first of all he was human, and second, his opponents were not all that bad. “Of course there are the ones [where] they look amateurish, but there are also the ones where they look ... suddenly you can see the beginning of modern chess. You suddenly see pawn structures and you think: they only discovered this in the 1920s. Suddenly you see the early versions of this and so on. ( ... ) Somehow they were knocking at the doors but they hadn’t worked out the details.” Nigel Short wrote: “My first hero was Paul Morphy. ( ... ) I think he was just way ahead of his time. I liked his economy. He didn’t waste time with his moves: they were very purposeful. In a sense they are classic games: they teach you to attack, not mess around. ( ... ) He was a hero and then at some stage I thought his games were not very sophisticated and his opponents were bad, and I should look at something better. I now think that he was so much in advance of his contemporaries, just a genius. Some of the ideas were quite deep and he had a fantastic calculating ability.”

On another occasion he added: “Morphy had this tremendous feeling for the initiative. ( ... ) All this playing against weak pawns and shuffling around is nonsense; it’s checkmate that wins the game. Morphy demonstrates economy of development, control of the centre and the use of open lines.” I wish to add what I believe was an essential characteristic of Morphy’s practical play, and that is his ability to maintain the tension, to abstain from committal decisions except when he had a concrete reason to do so. That characteristic was vital for him to succeed in the many games he played ‘at odds’; i.e. granting his opponent a material advantage at the start of the game. In some games, almost always simultaneous played at odds, rather than in ‘serious’ ones, Morphy seemed almost to be playing a game of ‘chicken’, from which he almost always emerged the winner. In those games the competitive and/or practical element was more important than objectivity. Finally, let us note Anderssen’s observation: “Morphy treats chess with the seriousness and conscientiousness of an artist ... For him a game of chess is a sacred duty”; and he said about his play: “Morphy’s play seemed to me like something from another world.” Morphy on the Attack

P.Morphy-E.Rousseau New Orleans 1849

White to play

Here we have a typical King’s Gambit scenario; White has sacrificed a piece to expose the enemy king. It is not possible to regain the piece yet with 10 Bxg8? in view of 10 ... h5!. Nevertheless, White has very reasonable compensation. So far he has only one pawn for the piece, but the f4-pawn looks as if it will fall soon. However, the most important factor is that it will take the black king some time to 009

reach a safe haven. Morphy was only twelve years old when he played this game. With greater maturity he would surely have gone for 10 d4!, a developing move which also builds a strong centre. Then 10 ... Qxd4? would allow 11 Bxg8, since the g6-square is now accessible to the white queen; while after 10 ... Kd8 White could play 11 Bxf4, not fearing 11 ... d5 on account of 12 Qg3 dxc4? 13 Be5. The move chosen by the young Morphy, in his first meeting with a well-known master, is inferior, but is testament to his tremendous imagination and his quest for the initiative at all costs. 10 e5? With this surprising method of opening the e-file Morphy wants to win by direct attack. There will be no material compensation. This move is reminiscent of a boxer, staking everything in search of a knockout in the final round. 10 ... Qxe5+ 11 Kd1 Kd8 This move is fine, although 11 ... Nf6 was also adequate, enabling the black queen to retreat to h5. 12 Re1 With the c8-bishop defended, 12 Bxg8? would fail to 12 ... d6 (or 12 ... d5) 13 Qg6 Bf5 14 Qf7 Bg4+ and wins; of course Morphy’s idea was not to take on g8 but to attack. 12 ... Qc5? So Morphy emerges the winner from this game of “chicken”. After 12 ... Qg7! or 12 ... Nf6!, he would have met insoluble difficulties in justifying his daring 10th move. 13 Bxg8 d5?? This natural move is actually a fatal mistake; 13 ... d6 was better, and then 14 Qg6 or 14 Be6 with some advantage.

Not even the mature Morphy could have handled this phase of the game with greater skill. 14 Re8+! An attractive way to draw out the enemy king, based on the key follow-up on move 16. 14 ... Kxe8 15 Qxc8+ Ke7 16 Nxd5+! This is the essential follow-up to 14 Re8+!, rather than 16 Qxb7+? Nd7 17 Qxa8, which surrenders the initiative and hands Black a mating attack after 17 ... Qg1+ 18 Ke2 Qxg2+ 19 Ke1 Qg1+ 20 Ke2 f3+! 21 Kxf3 Ne5+ 22 Ke2 (or 22 Kf4 Qf2+ 23 Kxe5 Bg7 mate) 22 ... Qg2+ 23 Ke3 Qf3+ 24 Kd4 Qf2+ 25 Kxe5 Bg7 mate. 16 ... Kd6 17 Qc7 mate J.Schulten-P.Morphy, Blindfold game, New York 1857

010

Black to play White is two pawns up, which can be reduced to one by 17 ... Qxd5, attacking f3 and with ideas of ... Nb6, or by 17 ... cxd5, opening the c-file. While these are both good moves, Morphy chooses the most direct way, concentrating more forces against White’s main weakness, which is his king. 17 ... Nb6! With this fresh pawn sacrifice we see a scenario which will be repeated time and time again. Morphy gains a superiority in forces in the area that becomes the focus of the struggle, not flinching from further sacrifices of material. Simple, isn’t it? Perhaps, but it has to work, and this will only happen if it is supported by precise analysis. Let’s not forget that the knight on h5 is not carrying out any function, either in attack or defence. From a “normal” position such as f6 it would defend the rook on e8, which as we shall see would be very useful. 18 Bxc6 Rac8 This “human” move, bringing more pieces into the attack, is an inaccuracy, or even an error. The unlikely-looking 18 ... Rec8! is preferred by the analysis engines, and it is the best move, for purely tactical reasons: in the ensuing tactical duel it will be important for Black not to have an undefended rook on e8. 19 Kd2? In this case it makes no difference which rook goes to c8. Instead, 19 Kb2! was better, with the possible continuation 19 ... Rxc6 20 dxc6 Bxe2 and now not 21 Rxe2? when 21 ... Na4+! wins, but the fact that the rook on e8 will be undefended grants White a vital tempo – he can play 21 Rb1!, preparing a retreat for the white monarch. Then after 21 ... Na4+ 22 Ka1 Nc3 23 Qd2, or 21 ... Ng3 22 Qd2, Black would have some compensation for the sacrificed material, but no clearly advantageous continuation. 19 ... Rxc6! Here there are equally strong alternatives, such as the simple 19 ... Nxd5 20 Bxe8 Qb4+ 21 c3 Rxc3! and wins. 20 dxc6 Bxe2 21 Rxe2 Qxd4+ 22 Ke1 Qg1+ 23 Kd2 Rd8+ 24 Kc3 Qc5+ Naturally 24 ... Qxd1 wins, but Morphy chooses a slightly quicker way. 25 Kb2

011

The ending after 25 ... Rxd1 26 Re8+ Qf8 27 Rxf8+ Kxf8 28 Be3 Rxa1 29 Kxa1 (or 29 Bxb6 Re1) 29 ... Ke7 is winning for Black, but Morphy had another idea in mind. 25 ... Na4+! 0-1 There is a mate after 26 bxa4, while 26 Kb1 sees Black conclude with 26 ... Nc3+ 27 Kb2 Nxd1+ etc.

P.Morphy-NN Blindfold simultaneous, New Orleans 1858

White to play

This position is very similar to the one arising from the Fegatello (or “Fried Liver”) Attack in the Two Knights Defence after 1 e4 e5 2 Nf3 Nc6 3 Bc4 Nf6 4 Ng5 d5 5 exd5 Nxd5?!. The differences are that Black has an extra pawn on d4 and his bishop is already developed on e7. White’s task is to demonstrate that what his own gains are more important: he is already castled and can quickly bring his c1-bishop and f1-rook into play. 8 Nxf7! In this version of the Fried Liver Attack, White has greater piece activity, but speed is essential. Black must not be allowed to castle. Here the knight sacrifice takes place in more favourable circumstances than on move six in the Two Knights Defence. 8 ... Kxf7 9 Qf3+ Ke6 Handing the piece back with 9 ... Bf6 10 Bxd5+ would be dismal: White has a big advantage after 10 ... Be6 11 Bxe6+ Kxe6 12 Bf4. 10 Nc3! The priority is development, bringing the remaining pieces into the attack as quickly as possible. One of the greatest admirers of Morphy’s play also reached this position in a simultaneous display. The finish was similar: 10 Re1+ Ne5 11 Bf4 Bf6 12 Nc3! (rapid development, just like Morphy) 12 ... 012

c6 13 Rxe5+ Kf7 (or 13 ... Bxe5 14 Re1) 14 Nxd5 Be6 15 Rxe6 Kxe6 16 Nxf6+ Ke7 17 Re1+ Kf8 18 Qa3+ 1-0 R.J.Fischer-T.Rouse, Chicago (simul) 1964. 10 ... dxc3 11 Re1+ Ne5 12 Bf4 Bd6 13 Bxe5 In order to win it would be sufficient to regain the sacrificed material with 13 Qxd5+ Kd7 14 Bxe5. Morphy chooses a more forceful way; all the white pieces are in the game except the rook on a1. Morphy brings this piece into play and is able to outnumber the enemy forces in the centre of the battlefield. 13 ... Bxe5

14 Rxe5+! Completing the three preconditions cited by Euwe: after rapid development of the pieces, control of the centre, and opening of lines comes the mating attack.

14 ... Kxe5 15 Re1+ Kd4 16 Bxd5 Rf8 17 Qd3+ Here 17 Qxc3+ was slightly quicker, with mate in three moves, but this does not matter – the black king cannot escape. 17 ... Kc5 18 b4+ Kxb4 19 Qd4+ 1-0 It is mate in four: 19 ... Ka5 20 Qxc3+ Ka4 21 Re4+ Kb5 22 Rb4+ Ka6 23 Qa3 mate. P.Morphy-J.Löwenthal London 1859

White to play

White is a pawn up and with the simple 22 Rfe1 his advantage would be beyond doubt. Another idea is to return the pawn to open lines with 22 e5! dxe5 23 Qe3. Morphy chooses something even more forceful. 22 Rb1! 013

In a few moves all the white pieces will be attacking the black king. 22 ... b6 23 Rfc1! Another attacking piece comes into play, threatening 24 Rxb6 axb6 25 Nxb6+ mating. 23 ... Qf6 24 Qe3 Ng4

25 Nxb6+! Destruction of the defence, opening lines in decisive fashion. 25 ... axb6 Or 25 ... Ke7 26 Rc7+! Kf8 27 Nd7+. 26 Rc7+! Kd8 If 26 ... Kxc7 then 27 Qxb6+ Kc8 (or 27 ... Kd7 28 Qa7+) 28 Qb8+ and mates. 27 Qxb6 Qxf2+ 28 Qxf2 Nxf2 29 Ra7 Nh3+ Something similar to the game happens after 29 ... Kc8 30 Kxf2. Note that 30 ... Rxe4 loses simply to 31 Ra8+. 30 gxh3 Kc8 31 Kf2 1-0

Black is defenceless against the plan of Ke3 and Rbb7, followed by the advance of the a-pawn to a6, and Ra8 mate. Morphy and the Initiative

P.Morphy-P.Ross Simultaneous game, New Orleans 1858

014

White to play

White has the advantage – his pawn structure is not favourable for an endgame, but “before the ending the gods have placed the middlegame” (Tarrasch) and White has all his pieces in play, whereas Black’s queenside remains dormant. Based on what we’ve seen so far, we can assume that the thirteen-year-old Morphy would have been reluctant to settle for the quiet move 12 Bd2, even though this guarantees a clear advantage, and would look for something better. 12 e5! With such a big dynamic advantage, opening lines should be the priority. 12 ... f5 Calculation confirms that opening the position with 12 ... fxg5 helps White: he can respond with 13 Nxg5, when 13 ... h6 is refuted by the nice shot 14 Qxg6!, exploiting the weakness of Black’s back rank. Developing a piece with 13 ... Na6 is no better, in view of 14 Rxf8+ Nxf8 15 Rf1 Ng6 16 Rf3, threatening 17 Qxg6 hxg6 18 Rh3 mate, among other things; in this line 14 ... Qxf8 runs into 15 Re4, with similar ideas. 13 Nd4!

White’s g5-bishop is now safe and Black cannot prevent further lines being opened with gain of time; the safety of the black king is becoming an ever greater problem. 13 ... f4 14 e6! dxe6 15 Nxe6 Bxe6 16 Rxe6 Qc8 The relief that Black gains from the piece exchange does not compensate for the opening of the e-file, as Morphy highlights with his next move. 17 Rxg6! hxg6 18 Qxg6 Qf5 If 18 ... Qe8, the confinement of the black king means that White can play 19 Qxe8 Rxe8 20 Rxf4 g6 (or 20 ... Kh7 21 Bf7) 21 Bf6+ and mates. 015

19 Rxf4! The prettiest and, at the same time, the quickest finish. 19 ... Qxg6 There is no salvation in 19 ... Qc5+ 20 Kh1, threatening 21 Qh5 mate and 21 Rh4 mate. 20 Rxf8+ Kh7 21 Bg8+ Kh8 22 Bf7+ Kh7 23 Bxg6+ Kxg6 24 Bf4 1-0 A.Meek-P.Morphy Mobile 1855

Black to play

Morphy’s king is rather exposed – something which is rarely seen in his games – but this is not a serious problem here, since White doesn’t have any pieces in threatening positions. It will soon become dramatically clear that it is in fact the white king that is in more danger. 10 ... Re8! The Morphy formula: rapid mobilization of the forces. This move also enables the black king to retreat without imprisoning the rook on h8. Now White must decide what to do about his attacked pawn.

11 Qb3+? A very strange decision; moving his only developed piece makes a poor impression. If Morphy had had the white pieces he would probably have chosen 11 0-0!, so that after 11 ... Rxe4 12 Qd5+ Re6 13 Bg5 Qf8 14 c3, he would have gained some compensation. 11 ... d5 The strange thing about White’s last move is that even the natural 11 ... Kg7 would have left Black with a great advantage, highlighting the futility of the check. But Morphy wants more and increases the pressure in the centre. 12 f3?! 016

This weakens White’s position, especially the e1-h4 diagonal, and allows lines to be opened in Black’s favour. Again 12 0-0 was preferable, even though after 12 ... Rxe4 13 Nd2 Re8 14 Nf3, Black stands well and has several good continuations such as 14 ... Qd6, followed by ... Bd7. Morphy might also have considered returning the pawn with 14 ... Bg4, intending 15 Qxb7 Qd7, when Black threatens to capture on f3, while after 16 Qb3 Rab8, all his pieces are active. 12 ... Na5 The alternative was 12 ... Qh4+ 13 g3 Qh3. Morphy’s choice is more unpleasant for White from the practical viewpoint: where should he move his queen? 13 Qd3? White fails to withstand the tension. This is the most obvious move, reinforcing e4, but Black’s initiative very soon becomes a whirlwind. Instead, it was better to play 13 Qb5 c6 14 Qe2, when the threat of an eventual b2-b4 provides White with some defensive resources; e.g. after 14 ... dxe4?! 15 00!. This time Black should prefer 14 ... Qh4+ 15 g3 Qh3 with the advantage. 13 ... dxe4 14 fxe4 Qh4+ 15 g3 Rxe4+? Black has an extra pawn and an attack, which can be conducted in various ways, but priority will be concentrate forces against the enemy king. Nevertheless, and surprisingly, Morphy’s move is a mistake; 15 ... Qxe4+ was better, as Black doesn’t need the queen anymore.

Exercise: Why was 15 ... Rxe4+? incorrect?

16 Kf2? Answer: It is easy to overlook the defensive move 16 Be3!, when 16 ... Rxe3+ can be met by 17 Qxe3, while after 16 ... Qg4 17 0-0+ Bf5, Black is better but it is still a fight.

16 ... Qe7 017

It was not necessary to defend the rook with the queen. It was more precise to play 16 ... Qf6+!, when 17 Bf4 Bf5 is strong, while after 17 Kg2 Black has 17 ... Bh3+! (in addition to 17 ... Bf5) 18 Kxh3 Qe6+ 19 Kg2 Re2+ 20 Kf1 and now, for instance, 20 ... Re1+ 21 Kg2 Qd5+. 17 Nd2? White misses the opportunity to complicate the struggle and improve his development at the same time with 17 Bg5!, when the position would no longer be clear. 17 ... Re3! 18 Qb5 White is trying to keep the e2-square defended. If 18 Qxd4, Black wins with 18 ... Re2+ 19 Kg1 Bh3 etc.

18 ... c6! Access to e2 is worth much more than the inactive knight on a5. 19 Qf1 After 19 Qxa5 Re2+, it is mate in two moves. 19 ... Bh3! The quickest way to bring the a8-rook decisively into the attack. 20 Qd1 After 20 Qxh3 Re2+ 21 Kf3 (or 21 Kf1 Re1+) 21 ... Qe3+ 22 Kg4 h5+ and mates. 20 ... Rf8 21 Nf3 Ke8 0-1 P.Morphy-J.Schulten Blindfold game, New York 1857

White to play

Black’s king has lost the right to castle and to do so artificially would cost several moves. For now he is two pawns up, although White can regain one by taking on c6. 018

16 Bd3 Objectively it was probably better to regain a pawn with 16 dxc6 bxc6 17 Bxc6, but the exchanges would alleviate Black’s mobility problems after 17 ... Nxc6 18 Qxc6 g6, or first 18 ... Rb8. Morphy prefers keep the tension by sacrificing a third pawn. This seems akin to a game of “chicken”, don’t you think? When he gave simultaneous displays, played blindfold, or gave material odds, this was not an unusual scenario in Morphy’s games. 16 ... Nf6 Two pawns up is enough, Black seems to be thinking, and he’s not wrong. It was possible to accept the third pawn: 16 ... Nxd5 17 Nxd5 Bxb2 18 Rab1 Bf6, but then White could play, for example, 19 Nxf6 Qxf6 20 Rxb7 and even in the endgame after 20 ... Qxf3 21 gxf3, the difference in activity between the two sides would be huge. White would likely regain another pawn and Black’s winning prospects would be problematic. 17 Rae1 Of course 17 dxc6 bxc6 did not feature in White’s plans, which were to mobilize all his forces. Is this the best move? Possibly not, but what Morphy was seeking here was to create tension, by giving Black various options and thus the chance of making a mistake. 17 ... Bxc3 Black finally takes the offered pawn, which was a correct decision, as was 17 ... Nexd5. 18 Bxc3 Nfxd5 19 Ba1

Black is now three pawns up, but the situation is not as clear as it might appear. Although there are no immediate threats, the white bishops are powerful, Black’s pieces rooks are not in play and his other pieces are not well coordinated either. White plans to double rooks on the e-file, increasing the pressure. 19 ... f6 This blunts White’s dark-squared bishop, at least for the moment, and plans to play ... Kf7 and ... Re8 as soon as possible. 20 Re6 Qd7 21 Rfe1 White has managed to activate his pieces, but ... “three pawns are three pawns”. 21 ... b5? Once again we see Morphy’s opponent unable to withstand the tension. He doesn’t have time to prevent Bc4 in this manner. It was preferable to continue with his original idea of 21 ... Kf7, overprotecting f6, or else play 21 ... Ng6, as the complications after 22 Bxf6! don’t appear to be fatal for Black; e.g. 22 ... gxf6 23 Bc4 (now 21 ... b5 becomes more comprehensible) 23 ... Kg7 24 Bxd5 Rf8.

019

22 Rxe7! Now the black king’s defences are demolished. 22 ... Nxe7 23 Bxf6! Nd5 24 Be7+ Kg8 If 24 ... Ke8 then 25 Qh5+ g6 26 Bxg6+ or 26 Qxg6+ and mates. 25 Bf5 1-0 T.Barnes-P.Morphy London 1858

Black to play

In this position Morphy’s king is the one that has remained in the centre, while his opponent’s king is castled. The pin on the black knight might therefore seem serious at first sight; Morphy demonstrates that this is not the case. 12 ... Ne5! Forced; 12 ... Bd7? 13 Nxd4 Nxd4 14 Bxa8 is no use. 13 Bxa8? “Too greedy ... ” (Karsten Müller). The position would be balanced after 13 Bg5! Qxg5 14 Nxg5 Bxd1 15 Bxa8 Be2 16 cxd4 (16 Rfe1? d3 is worse for White) 16 ... Bxf1 17 Kxf1 Ke7. 13 ... Nxf3+ 14 gxf3 Bxf3 15 Qd2 dxc3

020

16 Qg5 If 16 Bc6+, then not 16 ... Ke7? on account of 17 e5! cxd2 (17 ... dxe5 allows mate after 18 Qd7+, while if 17 ... Qg6+ or 17 ... Qe6 then 18 Qg5+ is decisive) 18 exf6+ gxf6 19 Bxf3 and wins. However, Black can instead play 16 ... Kf8! 17 e5 (or 17 Qg5 cxb2 18 Qxf6 gxf6 as in the game) 17 ... cxd2 18 exf6 Bxc6 19 b4 gxf6!, opening the g-file, when White has no time for 20 Bxd2 due to the threat of 20 ... Rg8 mate. 16 ... cxb2 17 Bc6+ Ke7 18 Qxf6+ Now 18 e5 dxe5 is futile. 18 ... gxf6 19 Bf4 Rg8+ 20 Bg3 bxa1Q 21 Rxa1 f5!

As well as a decisive advantage in material, Black has the initiative. 22 a4 Or 22 Kf1 Bc3 23 Rc1 b4 with an easy win. The rest is straightforward. 22 ... Bb6 23 axb5 f4 24 bxa6 fxg3 25 hxg3 Rxg3+ 26 Kh2 Rg6 27 Rf1? Allowing a quick mate, but White would not survive after 27 Ra3 Rg2+ 28 Kh3 Rxf2 29 a7 Bxa7 30 Rxa7 Rc2 either, and if 31 Rxc7+ then 31 ... Kf6 and 32 ... Bxe4. 27 ... Bd4 28 Kh3 Be5 29 Kh4 Bf4 30 a7 Rh6 mate J.Owen-P.Morphy London 1858

021

Black to play

The material is equal, Black’s rooks are more active than White’s, but the white queen is more active and the black king is more exposed. Which is the most important factor here? With his following move Morphy provides the answer. 24 ... Rfxf2! Tactics “help” Morphy, but this help does not fall from a clear sky. Let us recall what Lasker said: “the brilliant move of a master is essentially conditional not on a sudden and inexplicable realization, but on the placing of the pieces on the board.”

25 Rxc2 Rxc2 26 Qxa5 The structure has changed greatly since the previous diagram. With his next move Black takes a great step towards changing the evaluation of the position in his favour. 26 ... h6! Now there can be no doubts as to which king is the weaker. Morphy threatens to win with 27 ... Qg5.

27 Rf2 Rc1+ 28 Rf1 Qc5! With Black’s king safe it is now possible to involve the queen in the attack. 29 Qd2 Rc3! 29 ... Rxf1+ 30 Kxf1 Qxb5+ was advantageous, but on the principle of “when you see a good move, look for a better one”, Morphy seeks more than just a queen ending with an extra pawn – he wants to win a pawn and at the same time keep the rooks on. 30 Qd4 Qxd4! Here 30 ... Qxa3 was strong too, but Morphy sees a simpler way to win. 31 exd4 Rxa3 32 Rc1

022

32 ... Kf7! This reminds me of Rubinstein’s 25 Kf2! in his famous game with Lasker, St Petersburg 1909. Activating the king is more practical and perhaps more accurate than 32 ... Rb3 33 Rc8+ Kh7 34 Re8 Re3 35 Kf2 Re4 36 b6 Kg6, which would also be advantageous. 33 Rc7+ Kf6 34 Rxb7 Rd3 35 Kf2 Rxd4 The two connected central pawns supported by their king and rook are decisive. 36 Ke3 e5 37 b6 Rb4 38 Rb8 Ke7!

The black king heads towards the dangerous white pawn. The loss of both Black’s kingside pawns is unimportant: his central passed pawns are further forward and will advance with the help of their king and the rook. It is an easy win for Black. 39 b7 Kd7 40 Rg8 Rxb7 41 Rxg7+ Kc6 42 Rg6+ Kc5 43 Rxh6 Rb3+ 44 Ke2 e4 45 Rh8 Kd4 46 Rg8 Rb2+ 47 Kd1 Kd3 0-1 P.Morphy-H.Bird Simultaneous game, London 1859

023

White to play

White is currently a pawn down, but the black pawn on h2 appears doomed and then the strong white centre would start to be important. But capturing the intruder on h2 is not so simple. After 12 Rf2 Kg8, the immediate 13 Rxh2? loses to 13 ... Ng4, while 13 Qf3 Qd7 14 Rxh2 can be met by 14 ... Bg4 15 Qg2 Be6! (threatening to win material with 16 ... Ng4) 16 Qg3 Nc6, when Black is ahead on development and the white king has no pawn protection. Further preparatory moves such as 14 Nc3 are possible, but after 14 ... Nc6 15 Rxh2 Be6 16 Rg2 Ng4, followed by ... Rf8, it is Black who has the initiative. Morphy opts for another solution. 12 Qh5+ Kg8 13 Rxf6! This is the idea, completely transforming the situation. In return for the exchange White is able to weaken the black kingside, open the g-file, create a weakness on f6, etc. It is not clear whether White will be objectively better, but his game is easier, more pleasant. 13 ... gxf6 14 Nc3 Re5?! Black benefits little from driving the white queen back because, as we shall see, the rook is badly placed here and the back rank is left weak. It was better to redeploy his precariously placed piece with 14 ... Be6!, and at the same time control the square d5. After, for instance, 15 Bf4 Nc6 16 Kxh2 Re7!, anticipating Rg1+, the position would offer chances to both sides. 15 Qf3 Qd7 The clumsy position of the rook on e5 means that 15 ... Be6 is not so good now. White can play 16 d4!, since 16 ... Qxd4 17 Qg3+ Kh8 18 Bh6 forces Black to return the exchange: 18 ... Rg5 19 Bxg5 fxg5 20 Qxg5, when White would have the initiative; then if 20 ... Nc6, White could play 21 Ng6+! hxg6 22 Qh6+ Kg8 23 Qxg6+ and the e6-bishop is lost. 16 Bf4

024

16 ... Nc6 Returning the exchange is again almost forced now. If 16 ... Re8 17 Bxh2, the pawn on f6 is threatened and the g-file available for White’s remaining rook with a raging attack; for example, 17 ... Rf8 18 Rg1+ Kh8 19 Qh5 Rg8 20 Rxg8+ Kxg8 21 Nd5 and there is no defence. It is no better to play 16 ... Bg4 17 Qg2 Rh5 (controlling d5 and f5; if 17 ... Re8 instead, then 18 Nd5), since after 18 Nf5 Rxf5 19 exf5 Qxf5 20 Rf1 Nd7 21 Nd5 (threatening forks on e7 and e3) 21 ... Qh5 22 Ne3 f5 23 Bxc7 (threatening to take on f5) 23 ... Rf8 24 Qxb7, material is now equal, while the black position is very weak. 17 Kxh2 This keeps all options open and enables the rook to use the g1-square, but in this case it appears that it was better for White to “content himself” with 17 Bxe5 Nxe5 (it is worse to play 17 ... fxe5, as after 18 Kxh2 Bg4 19 Rg1 h5 20 Nd5, all the white pieces are in the game) 18 Qxf6 Ng4 19 Qf3 Qd4 and now 20 Nf5, when material is once again equal and White’s advantage is clear. 17 ... Bg4 18 Rg1

18 ... h5? As often happened with Morphy’s opponents, Bird fails to withstand the tension of the struggle. 18 ... Rg5! was much more tenacious: after 19 Qg3 h5 20 Bxg5 fxg5 21 Nf5, White is slightly better, but the struggle can continue with 21 ... Rf8 or 21 ... Nd4. In contrast, Black’s position now goes rapidly downhill. 19 Bxe5 fxe5 20 Nd5 Nd4? 21 Nf6+ Kh8 Or 21 ... Kf7 22 Qf2. 22 Qe3 Another way was to launch an immediate assault with 22 Ng6+! Kg7 23 Nxh5+, when 23 ... Kxg6 24 Rxg4 just wins, and 23 ... Kh6 24 Qf6 Qe6 25 Rxg4 is no better. 22 ... Qg7 23 Nxh5 025

The last shred of protection is stripped away from the black king. White now has the attack and a material advantage as well, so the rest is very simple for Morphy. 23 ... Qh7 24 Rxg4 Qxh5 25 Qh3 Kh7 26 c3 Ne6 27 Rg6 Also good was 27 Ng6 Qxh3+ 28 Kxh3 and the e5-pawn falls. 27 ... Re8 Here 27 ... Qe2+ would change nothing: after 28 Rg2 Qh5, White can play 29 Rg3. 28 Rxe6 Rxe6 29 Qxe6 Qxh4+ 30 Qh3 Qxh3+ 31 Kxh3 The pawn endgame with an extra pawn is easily won. 31 ... c5 32 Kg4 Kg6 33 Kf3 Kf6 34 Ke3 Ke6 35 d4 exd4+ 36 cxd4 cxd4+ 37 Kxd4 Kd6 38 e5+ Ke6 39 Ke4 Ke7 40 Kd5 Kd7 41 e6+ Ke7 42 Ke5 a6 43 a3 Ke8 44 Kd6 1-0 In the final two examples in this section, the characteristic of Morphy’s style that we have already identified, his tendency to complicate and/or keep the tension, stands out even more clearly.

P.Morphy-T.Barnes London 1858

White to play

White is a piece down for a pawn. He could win a second pawn with 19 Qxh7, but 19 ... c6 leaves Black’s position very solid and White’s compensation would be insufficient; for instance, after 20 f4 e4 21 f5 Bf6 22 Qh6 Be5. The main defect of Black’s game is that his king is still not secure. At present this is not obvious, as the white pieces are not active enough. However, Morphy highlights this factor with his next two moves. 19 f4! First of all White needs to open lines. 026

19 ... e4 Exchanging a piece before closing the centre with 19 ... Bc5 20 Bxc5 Qxc5+ 21 Kh1 e4 fails to improve the situation, as White can play 22 Bxd5! Bxd5 23 Qh5+ and 24 Rxd5. 20 Rxd5! Riskier and more ambitious than 20 Bxd5 Bxd5 21 Qh5+ Nf7 22 Rxd5, when the engines are not afraid to distance the queen from the action with 22 ... Qxa2. This continuation is “inhuman”, and while it might be objectively correct, Black seems to have a simpler defence with 22 ... Qa6 and if 23 Qxh7 then 23 ... Qg6, while after 23 f5 Qf6 24 Qxh7 Rh8 25 Qg6 Qxg6 26 fxg6 Nd6, Black seems to be okay. 20 ... Bxd5 21 Qh5+

21 ... Kf8 The choice between this move, which loses the right to castle, and the self-pinning 21 ... Rg6 is a difficult one. Objectively, they seem to be of similar strength, but from the practical viewpoint the rook move is more complex and thus less advisable. A possible continuation might be 22 Bxd5 Bc5 23 Kh1 and Black is faced with another difficult decision. After the natural 23 ... Qa6, White has 24 Re1 Bxe3 25 f5! and his initiative is very dangerous; for instance, 25 ... Nf7 26 fxg6 Qxg6 (or 26 ... hxg6 27 Qh4! Kf8 28 h3 Bc5 29 Qf4) 27 Qxg6 hxg6 28 Rxe3 0-0-0 29 Be6+ Kb8 30 Kg1, and in all cases White comes out the winner. Instead, Black can sacrifice his queen with 23 ... Bxe3!, and after 24 Bf7+ Nxf7 25 Qxa5 0-0-0 26 Qf5+ Rd7 27 Qxe4 Bb6, chances are approximately equal. 22 Bxd5 Rg7 23 b4 White must dislodge the black queen so that he can advance his f-pawn. 23 ... Qa6 24 f5

24 ... Nf7? Once again the opponent cracks under the pressure. The knight move looks like a good idea, reinforcing the defences, but it fails tactically. It was essential to play 24 ... Qf6! 25 Bd4 Qg5, when 26 027

Bxg7+ Kxg7 27 Qe8! Bf6 28 Kh1 leads to a complex position. 25 f6! Bxf6 26 b5 Qd6 27 Bxf7 The refutation, threatening 28 Bc5. 27 ... b6 28 Bh6

White has a material advantage and more active pieces, so the game is won. 28 ... Ke7 29 Bxg7 Bxg7 30 Bb3 Rf8 31 Rf7+ Rxf7 32 Qxf7+ Kd8 33 Qxg7 Qd1+ 34 Kf2 Qd2+ 35 Kg3 e3 36 Qf6+ Kc8 37 Be6+ Kb7 38 Qf3+ 1-0 P.Morphy-C.Maurian Odds game, New Orleans 1866

White to play In this game Morphy played without his queen’s knight but, as we can see, he managed to regain the material and now his three remaining pieces are attacking the black king. There are possible mating patterns here; for instance, after 34 Rg7+ Kh8, White could play 35 Nf8, threatening 36 Ng6 mate, and the knight is taboo in view of mate in two with the rooks; but Black has 35 ... Re1+ 36 Kf2 Re2+ 37 Kg1 (37 Kf3?? allows the entry of the other rook with lethal effect: 37 ... Ra3+ 38 Kf4 Rf2 mate) 37 ... Re1+ and draws, since the attempt to flee with 38 Kg2?? loses to 38 ... Rxa2+ 39 Kh3 Bf1+ 40 Kh4 Rxh2 mate. How to keep up the pressure?

34 a4!! With this surprising move, which presents Black with two passed pawns, White prevents the deadly rook invasion on a1 and puts Black under pressure again. 34 ... bxa3? 028

We already know why this loses. Black can seek another invasion route with 34 ... Rae8, which is good enough to draw if he manages to find the right moves: 35 Rg7+ Kh8 36 Nf8! (threatening 37 Ng6 mate, or 37 Rh7+ Kg8 38 Rcg7 mate if the knight is taken; but Black has a narrow path to survival by harassing the white king) 37 Kg2 Bf1+! 38 Kf3 (not 38 Kf2?? R8e2+ 39 Kf3 Bg2+ 40 Kf4 Rf1 mate) 38 ... Be2+! 39 Kf2 Bh5! (threatening 40 ... R8e2 mate) 40 Rh7+ Kg8 41 Rxh5 R1e2+ and now 42 Kf3 R8e3+ (42 ... R2e3+? leads to an inferior rook ending after 43 Kf4 Re2 44 Kxf5 R2e5+ 45 Kg6 Rxh5 46 Kxh5 Rxf8 47 Rxc6 Rb8 48 Kg6 b3 49 Rc1) 43 Kf4 Re5 (threatening 44 ... Rf2 mate) 44 Rh8+ Kxh8 45 Ng6+ Kg8 46 Nxe5 fxe5+ 47 Kxf5 Rxh2 48 Rxc6 Ra2 should be a draw.

However, the simplest way to defend was to let the bishop cover the g6-square from d3 by playing 34 ... f4!; e.g. after 35 Rg7+ Kh8 36 Nf8 Re1+ 37 Kf2 Rf1+ 38 Kg2 f3+ 39 Kh3 Bd3. If instead 35 gxf4, then 35 ... b3 (35 ... Rae8? fails to 36 f5!) 36 Rb7 b2! (not 36 ... Kh8? due to 37 Nf8! again) 37 Rxb2 Kh8 and Black is okay; e.g. 38 Rb4 Rg8+ 39 Kf2 Bd5 (intending ... Rge8) 40 Rbb7 Rg2+ 41 Kf1 Ree2 42 Rd8+ Bg8 43 Nxf6 Rgf2+ 44 Kg1 Rg2+ with perpetual check. Let’s see one more example: 35 Rd4 b3 (35 ... Bd5 is also fine) 36 Rxc4 b2 37 Nxf6+ Kf8! (better than 37 ... Rxf6?! 38 Rb7) 38 Rb4 Re1+ 39 Kf2 b1Q 40 Rxb1 Rxb1 41 Rd7!, followed by Nh7+, Nf6+ etc with another draw. 35 Rg7+ Kh8 36 Nf8! 1-0 Morphy on the Defensive Finally, let’s look at three examples of another important facet of Morphy’s play, complementing his recognized mastery of the attack – this is his great resourcefulness in defence.

A.Anderssen-P.Morphy 6th matchgame, Paris 1858

Black to play Black’s position is in danger. He is the exchange up but is faced with the threat of 40 Rg3. The pin on the g7-rook is unpleasant and the black queen’s mobility is restricted both by the need to watch out for Ne5 and hinder the advance of the passed d-pawn. All this means that Black’s defence is far from easy. The natural 39 ... Rc8? loses to 40 Qf6!, with the threat of 41 Ne5 Rxc3 42 Qf8+ Rg8 43 Nf7 mate, while 40 ... Kg8 loses to 41 Qe6+, and 40 ... Qe8 fails to 41 d7 (or 41 Ne5 Rxc3 42 d7) 41 ... Qxd7 42 Ne5 Qe8 43 Nf7+ Kg8 44 Nh6+ Kh8 45 Nxf5 etc. Black has no time to unpin the rook with 39 ... h6? and ... Kh7, as after 40 Rg3 Qd7, White can regroup his forces and win with 41 Ne5 Qb7 42 d7. Finally, 39 ... Kg8? is refuted by 40 Rg3! Rxg3 41 d7. So how can Black defend? 39 ... e3! An elegant way to prevent 40 Rg3. Now White needs to realize that he must forget about winning and instead seek a way to draw. 40 Rxe3? Anderssen fails to adapt to the altered situation. With 40 Qf6! he could have made a draw; for instance, 40 ... Rxc4 41 Qf8+ Rg8 42 Qf6+ etc. Black loses if he tries 40 ... Qd1+? 41 Kh2 Qd5, since 42 Nxe3 defends g2, and then 42 ... Qf7 43 Qd4! is029very strong; e.g. 43 ... Rxc3 44 Qxc3 h6 (44 ... Qd7

meets the same reply) 45 Qe5, followed by 46 Nxf5 or 46 Nd5 with a decisive initiative. Surprisingly, 40 Qxe3! is also sufficient, thanks to the strength of the passed pawn. For example, 40 ... Rxc4 41 Rxc4 Qxc4 42 Qe8+ Qg8 (or 42 ... Rg8 43 Qe5+) 43 d7 ends in a draw by perpetual check after 43 ... Rxg2 44 d8Q Rg1+ etc. 40 ... Rxc4 As simple as that – the queen controls e8 and there is no time to create counterplay. 41 Qf6 Rc1+ 42 Kh2 Qxf4+ 0-1 P.Morphy-J.Löwenthal London 1859

White to play The endgame looks unpleasant for White. His own pawns present no danger, whereas the black pawns look very menacing. To discover the most secure defence it is essential to possess a solid knowledge of the endgame, which Morphy now exhibits. 40 Bxa3! Calculating that he cannot be prevented from reaching a ‘fortress’ position. 40 ... bxa3 41 Kd4 Bxf4 42 Kxc4 Bxh2 43 Kb3 Bd6 44 Kc2 ½-½ Reaching a theoretical draw; the white king cannot be dislodged from b1 and c2. P.Morphy-W.Thomas Odds game, Philadelphia 1859

White to play 030

This game began with Morphy giving odds of a knight. His position now looks desperate, since Black threatens mate in two with 45 ... Be1+. Note that 45 Rxe4+? Kxe4 46 Kg5 fails to draw, since Black manages to hang on to a vital pawn after 46 ... Be7+ 47 Kxg6 h4 48 Kh5 Kf4 and will win by capturing the pawns on g2 and h3 and queening his h-pawn. The point is that, even though Black has the ‘wrong bishop’ in respect of the h1-square, the white king will be unable to reach g1. Black also wins after 45 Rf6+? Bf5 46 g3+ Ke5 47 Rf7 Ke6, followed by 48 ... Be7+. Morphy finds the only way to save the game. 45 g3+! This is the correct move order, improving the first variation mentioned above. 45 ... Kf5 46 Rxe4! Kxe4 47 Kg5 This is the difference; Black is unable to save his pawn by ... h5-h4, so the exchange of all the remaining pawns is inevitable. 47 ... Kf3 48 Kxg6 Kxg3 49 Kxh5 Kxh3 ½-½

031

Chapter Two Games Against Family Members Paul Charles Morphy was born in New Orleans on the 22nd of June 1837. According to his uncle Ernest, Paul learned to play without anyone teaching him how to move the pieces, just by watching how his uncle and his father were playing. Curiously, this beginning is very similar to the one that legend attributes to José Raúl Capablanca. Chess was played by all the family; and Morphy’s father Alonzo Michael, his brother Edward, and his uncle Ernest – who was one of the best chess players in the city – were his first opponents. Let us look at some games from that period, in which the stamp of Paul Morphy is already clear. Game 1 P.Morphy-A.Morphy New Orleans 1849 Evans Gambit [C51] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 At 12 years old Morphy already played the Evans Gambit regularly; it is a pawn sacrifice which allows White to gain time to develop his pieces and occupy the centre, which, as Euwe stated, were two of the three aims that Morphy was striving for. Throughout his short career Morphy played this position many times and with both colours. He also made use of it when giving odds of his queen’s knight. It was tried occasionally by Fischer (albeit mostly in simuls) and in more recent times by Kasparov and it keeps appearing from time to time at the top level. At the time of writing it is not going through its best period. Fabiano Caruana, for instance, does not have a good opinion of the opening because, according to him: “The Evans Gambit may have a romantic reputation, but modern opening analysis has entirely defanged it.” 4 ... Bxb4 5 c3 Bc5 The alternative is 5 ... Ba5, which makes no difference if Black plays ... Bb6 as in the game, but it gives Black more options after 6 d4 as the bishop is not attacked. We shall see this in later games. 6 d4 exd4 7 cxd4 Bb6 8 0-0

Question: There are no immediate threats. What has White got for his pawn? Answer: White has achieved his objective of occupying the centre and he has the freer game, while Black still has his king in the centre.

8 ... Na5 032

This move is successful in dislodging the bishop from its active position, pressing against the weak point f7, but at the cost of giving up control of the centre – and Black runs the risk of the knight remaining out of play, whereas the white bishop will still be active on d3. 8 ... d6 is more popular, to develop the c8-bishop and control the e5-square. Against this Morphy played all three main lines: 9 d5, 9 Nc3 (see Games 13-14) and, less frequently, 9 Bb2. 9 Bd3 d5? This is already excessively ambitious. Black wants to develop his queenside and castle long, but opening lines when the king is still in the centre is something that should be assessed very carefully. In another game Alonzo Morphy, Paul’s father, continued more cautiously with 9 ... Ne7 10 Nc3 0-0, when his king was much better placed. The game continued 11 Ba3 d6 12 e5 Bf5 13 exd6 cxd6 14 Ne4 d5?! (14 ... Nc4 is better, or else 14 ... Bxe4 15 Bxe4 Re8) 15 Nf6+! gxf6 16 Bxe7 Qxe7 17 Bxf5 Nc4? (a serious error; 17 ... Rfe8 is better, intending to defend with 18 ... Qe2 with a serious struggle ahead) 18 Re1 (it is very tempting to activate a piece with gain of time, as we shall see in many of Morphy’s games, but in this case 18 Nh4!, giving the queen access to the weakened black kingside, was much stronger) 18 ... Qd6

Exercise: What should White play here? Answer: With the black queen on d6, the move 19 Nh4 is less strong, though still advantageous: 19 ... Qf4 20 Qh5 h6 21 Nf3! (threatening 22 g3) 21 ... Nd6 22 Bh3 (renewing the threat), when Black has nothing better than 23 ... Ne8 24 g3 Ng7, sacrificing the d5pawn.

It is even better to eliminate a potential defender with 19 Nd2!, with the threat of 20 Qg4+ Kh8 21 Qh5 mating or 20 Qh5 at once (which would be the response to 19 ... Nxd2). There is no good defence: 19 ... Qf4 20 Qh5 h6 now fails to 21 Nxc4 dxc4 22 Re4 Qd2 (or 22 ... Qg5 23 Rg4) 23 Rd1 and wins. Instead, Morphy chose the incorrect 19 Ne5?, which failed to justify itself after 19 ... fxe5 20 Qg4+ Kh8 21 Qh5 Kg7 22 Qg5+ Kh8 23 Qh5 h6, although White ended up winning an error-strewn game. 10 exd5 Qxd5

033

11 Ba3! It was difficult to decide between this typical Evans Gambit move, highlighting the absence of the b6bishop from its original diagonal and hindering Black’s kingside castling, and the equally strong alternatives 11 Nc3 or 11 Re1+. 11 ... Be6 12 Nc3 Qd7

Exercise: There are several tempting moves; what do you think was Morphy’s choice? Answer: Given a breathing space, Black will castle queenside, maintaining his extra pawn, so this is a critical moment. White must prevent Black from bringing his king to safety at all costs.

13 d5! It was equally strong to play 13 Bb5 c6 14 d5! cxb5 15 Ne5!, but there is no need to enter complications when there is a simpler continuation which is just as effective. 13 ... Bxd5 14 Nxd5 Qxd5

034

15 Bb5+! In many of Morphy’s games we clearly see the importance of the time element in chess. In certain positions, being able to play “two moves running” is more valuable than the material it costs. This is in accordance with Euwe’s explanation of Morphy’s style, based on: 1. rapid development of the pieces; 2. control of the centre; and 3. open lines. Here he sacrifices the bishop to keep the black king stranded in the centre, and does so in ideal circumstances, since he regains the piece straight away.

The immediate 15 Re1+ Kd8 16 Be4! was also winning, forcing the exchange of queens, which brings the a1-rook into play: it is mate after 16 ... Qxd1 17 Raxd1+ Kc8 18 Bf5+ Kb8 19 Re8. 15 ... Qxb5 16 Re1+ Ne7

17 Rb1! We shall see this scenario many times: it’s always nice to activate a piece with a gain of time. In this case Morphy wants to force the black queen to decentralize. There is more than one way to win here: 17 Bxe7 is decisive, as is 17 Rxe7+ Kf8 and now 18 Ne5, or 18 Rb1 as in the game, (18 Re5+ is less effective due to 18 ... c5, although White is still winning.) 17 ... Qa6 18 Rxe7+ Kf8 19 Qd5 All Black’s pieces are ineffectual, whereas White’s occupy ideal positions. Mate in a few moves is inevitable. 19 ... Qc4 Preventing 20 Qxf7 mate, but that was just one of the threats. 20 Rxf7+ Kg8 21 Rf8 mate Game 2 P.Morphy-E.Morphy New Orleans 1850 Evans Gambit [C52] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Ba5

Ernest Morphy, Paul’s uncle, retreats his bishop to a5 instead of c5, as played by Paul’s father in the 035

previous game. 6 d4

6 ... exd4 Fabiano Caruana has no faith in this move, because: “The old games, such as those by Anderssen, had Black capturing on d4 and coming under attack, often losing in under 30 moves (the “Evergreen Game” being the most famous example).” The most popular treatment in the 21st century is 6 ... d6 7 Qb3 Qd7 8 dxe5 Bb6, and if 9 Nbd2 dxe5, Black gains a good position and has scored well. In an attempt to revive the line White has tried 9 a4, but after 9 ... Na5 10 Qa2 Nxc4 11 Qxc4 Ne7 12 exd6 cxd6 13 0-0 0-0, as in H.Nakamura-V.Anand, London 2014, “Black already has a comfortable game and is the only player with chances of an advantage”, according to Caruana. In L.Nisipeanu-F.Caruana, Dortmund 2015, White tried 12 Ba3 (“?” – Caruana), when 12 ... 0-0 was good, but 12 ... d5 13 exd5 Qxd5 is even better and “White will most likely end up in a sad endgame with a wrecked structure, and Black has a bishop pair”, according to Caruana, describing the outcome of an Evans Gambit completely lacking in any trace of Romanticism. 7 0-0 Bxc3?! Black is now no less than three pawns up which, other things being equal, would constitute a decisive advantage. That’s not the case here, and neglecting development is really not advisable against the Evans Gambit. This is not exactly a losing move, but it requires very precise defence. The most popular move here is 7 ... Nge7, ahead of 7 ... d6 (as in Game 13). The third placed choice 7 ... bxc3 is again more risky, as we shall see in Game 15. 8 Nxc3 dxc3 9 Ba3 The alternative was another typical Evans Gambit move, 9 Qb3. It is difficult to say which is better. Löwenthal pointed out that Morphy deviated here from the line recommended in the German magazine Schachzeitung in 1851: 9 Ng5, which leads one to believe that this game was perhaps played after 1850. In any case the position after 9 Ng5 Nh6 does not appear to be better for White than Morphy’s choice. 9 ... d6

036

Exercise (easy): How did Morphy continue?

Answer: 10 Qb3 Of course, this is the other typical move. There was not really any other way to activate the queen, which now increases the pressure on f7. 10 ... Nh6 Exercise: Black’s response was forced. Now what for White?

Answer: 11 Qxc3 We need prose as well as poetry! This regains one of the pawns and attacks g7, with an advantage in development. 11 ... Qf6 Exercise: Why did Black reject 11 ... 0-0 - ? Answer: Because with 11 Qxc3 White also opened the long dark diagonal. He would continue 12 Bb2 forcing 12 ... Qf6, and in the ending arising after 13 Qxf6 gxf6 14 Bxf6 the position opens up to the benefit of the bishop pair; furthermore the white rooks stand ready to be activated on the b- and c- files. All this would give White good compensation for the pawn.

Exercise: How did Morphy reply to 11 ... Qf6 - ? Answer: With another pawn sacrifice, avoiding the exchange of queens and reopening the a3-f8 diagonal, after which the black king will 037

be exposed once more.

12 e5! dxe5 It is noteworthy that simplifying with 12 ... Nxe5 13 Nxe5 dxe5?! (naturally not 13 ... Qxe5? 14 Rfe1) 14 Rae1 would favour the better developed side, since he would have more open lines, as we shall see, compared with the game. In this variation it would be better to play 13 ... 0-0!, when 14 Rac1 Qxe5 15 Qxe5 dxe5 16 Bxf8 Kxf8 leaves Black with three pawns for the exchange. However, this would be offset by him having a knight on the rim and an undeveloped queenside, while the three white pieces would be active and ready to invade after, for instance, 17 Rfd1. Admittedly, the position would not be totally clear. 13 Rfe1 Bd7 Exercise: The black monarch cannot castle on the kingside, so seeks refuge on the other wing. What did Morphy play here?

Answer: 14 Rab1! Not 14 Nxe5?, as this would be answered by 14 ... 0-0-0, when Black’s king is safer and he remains two pawns up, albeit still in a uncomfortable position. Morphy opted for making the queenside the new battlefront. 14 ... 0-0-0

Exercise: Black’s response came as no surprise. What had Morphy planned?

Answer: 15 Ba6! An accurate shot that has the devastating effect of a knockout blow, even though objectively it is not winning. 15 ... Na5? This loses fairly easily; it is Black’s only clearly bad move in the whole game. The critical line was 15 ... bxa6! 16 Qb3, threatening mate. Although Black’s position is tottering, the game is still not at all clear. Now 16 ... Be6? fails to 17 Qb7+ Kd7 18 Rbd1+ Nd4 (and the king cannot retreat because the c6knight falls) 19 Nxe5+ Ke8 20 Rxd4 Rxd4 21 Qa8+ (a preparatory check to ensure that d8 is occupied; 21 Qc6+ Kd8 22 Qa8+ Bc8 23 Nc6+ wins more slowly) 21 ... Qd8 22 Qc6+ Rd7 23 Nxf7! and the black position collapses. Instead, 16 ... Bg4! is better, so that the knight on c6 remains defended by the queen. After 17 Qb7+ Kd7 18 Rbd1+ Ke8 19 Rxd8+ Kxd8!, the white attack is dangerous, but it is not easy find a decisive continuation. (Not 19 ... Nxd8? 20 Qxc7 Bxf3 21 Rxe5+ Ne6 22 Qc8+ Qd8 23 Rxe6+ and mates.)

038

Exercise: How did Morphy punish his uncle’s mistake?

Answer: 16 Rec1 Bringing up fresh troops. Now there is no good way to defend c7. 16 ... Bc6 17 Qxa5 bxa6 18 Qxa6+ Kd7 White has succeeded in opening up the defences of the black king and all the white pieces are attacking.

Exercise: Now there are many good moves; Morphy chose the strongest – what is it?

Answer: 19 Rxc6! Of course; if 19 ... Qxc6 then 20 Nxe5+ wins. 19 ... Qf5 Leaving the rook on c6 leads to mate, but there was no defence. The rest is a massacre. 20 Rxc7+! Ke8 21 Qc6+ Qd7 22 Rb8 Qxc6 23 Re7+ Kf8 24 Rxd8+ Qe8 25 Rexe8 mate In 1849, at the age of twelve, Morphy played about fifty informal games against the master Eugene Rousseau, at that time considered to be the strongest chess player in the southern states of the USA. Morphy won comprehensively. We saw an extract from one of those games in Chapter One. In 1850 the Hungarian master Johann Löwenthal, a future opponent of Morphy’s in a ‘serious’ match, visited several cities in the south, amongst them New Orleans. Not very enthusiastically, since he thought it would be a waste of his time, he agreed to play against the young Paul Morphy. Löwenthal was 039

surprised by the strength of the young prodigy. They played three games, and Morphy won 2½-½ (other sources say that the score was 3-0). Let’s look at their second game. Game 3 P.Morphy-J.Löwenthal New Orleans 1850 Sicilian Defence [B21] 1 e4 c5 2 f4 e6 3 Nf3 d5

4 exd5 Question: Hmm, what an ugly move! It just seems to help Black’s game. Is it the best move? Answer: You’re right, it leads to an “ugly” position in which White would really rather have his pawn back on f2! We now know that the best move is 4 Bb5+, with the idea of 4 ... Bd7 5 Bxd7+ Nxd7 6 d3, maintaining his pawn formation. Another possibility is simply 4 e5.

The major beneficiary of the exchange on d5 is the c8-bishop. This decision by the then (very) young Morphy reminds us that against the French Defence (which was not very common at that time) 1 e4 e6 2 d4 d5, Morphy’s favourite move was 3 exd5. 4 ... exd5 5 d4 Bg4?! Question: Didn’t you just say that the major beneficiary of 4 exd5 was this bishop? In that case, why is this move dubious? Answer: The bishop’s increased freedom of movement doesn’t mean that it should be moved straight away, unless there is some benefit to be gained. Here the old advice of “knights before bishops” is valid. In other words, it’s best to start with the moves that you’re sure you’ll want to play, such as 5 ... Nc6.

6 Be2

040

6 ... Bxf3? There is no visible benefit from this unnecessary exchange, made without even waiting for White to play h2-h3. From f3 the white bishop puts pressure on the d5-pawn and Black gains nothing in return. There is no doubt that 6 ... Nc6 was better. 7 Bxf3 Nf6 8 0-0 Be7 If Black wants to play with an isolated queen’s pawn (IQP) it would be preferable to exchange on d4 at once; after 8 ... cxd4 9 Qxd4 Nc6, White is slightly better, but his coordination is not as good as in the game. 8 ... Nc6 also looks an improvement. 9 Be3 cxd4 Having dismissed 8 ... cxd4, Black now makes this exchange in rather inferior circumstances, with the white bishop already on e3. Changing the structure with 9 ... c4 was worth considering; then after 10 b3 cxb3 11 axb3 0-0 12 c4, it looks best to play 12 ... Bb4!, preventing 13 Nc3 from creating very unpleasant pressure on d5. 10 Bxd4 0-0 11 Nc3 Nc6

Exercise: What do you think Morphy played here?

Answer: 12 Bxf6 A very striking decision; instead of hanging on to his pair of bishops and keeping the isolated pawn on d5 with 12 Bf2, which would retain a static positional advantage, Morphy opts for dynamism, based on the strength of his knight (which will soon settle on d5) and the greater activity of his pieces. 12 ... Bxf6 13 Nxd5 Bxb2 14 Rb1 Bd4+ 15 Kh1 Rb8 White is slightly more active, but that’s all; there is nothing immediate. Morphy is planning to make gradual progress by forcing his opponent to take decisions. 041

16 c3 First of all he forces the bishop to decide which diagonal to remain on. 16 ... Bc5 The alternative was 16 ... Bf6, but that would suffer from the disadvantage of having to keep the queen tied to its defence, so as not to allow his pawns to be doubled. The two tempi required for a plan of ... g7-g6 and ... Bg7 would prove too much. After 17 Qa4, for instance, there is no time for 17 ... g6? 18 Rfd1 Bg7 due to 19 Nb6, followed by 20 Nd7, and White wins material.

Exercise: What was Morphy’s idea in forcing the black bishop to choose between the two diagonals?

Answer: 17 f5 Creating the latent threat of advancing to f6, and suddenly the ‘ugly duckling’ f4-pawn becomes ‘a thing of beauty’, a real asset to White. Nevertheless, the position still remains rather unclear. 17 ... Qh4?! A difficult choice. Years later Löwenthal commented that he should have played 17 ... f6, but White would still be better, thanks to the significant weakness on e6, and could continue 18 Qb3 Kh8 19 Nf4. Covering the f6-square with 17 ... Be7 looks better; even so, White has the more comfortable game. 18 g3 Qg5

Exercise: How did Morphy proceed?

Answer: 19 f6! Naturally, this was the idea. Now Black must choose between two evils: leaving the wedge on f6 042

after 19 ... g6 or allowing lines to be opened against his castled position. 19 ... Ne5?! In view of what happens in a few moves it is easy to recommend 19 ... g6, but it wasn’t a simple decision. 20 fxg7 Rfd8 Exercise: How did Morphy continue?

Answer: 21 Be4! Of course, preparing an assault along the f-file and the d1-h5 diagonal. 21 ... Qxg7

22 Qh5 A difficult decision. Sometimes the “problem” of having a favourable position is that there can be several attractive options and you have to calculate and then select the best one. It seems clear that White is going to bring his major pieces in to join the struggle. He can start with 22 Qh5, and then perhaps Rf5 and Rbf1. Or he can begin with 22 Rf5. Which move order is best? This sort of decision requires time to analyse. Furthermore there is even another possibility, a tactical one, namely the shot 22 Bxh7+. Let’s begin with this last option: 22 Bxh7+! is in fact a good move; White has some advantage after 22 ... Kf8 (or 22 ... Kxh7? 23 Qh5+ Kg8 24 Nf6+ Kf8 25 Qxe5) 23 Be4 Rd6 24 Qd2, but probably this did not satisfy Morphy and he wanted to obtain more from the position. It is understandable that Morphy didn’t play the strongest move, which was 22 Rf5!. As we said, the idea is to play Qh5 and Rbf1, and Black is surprisingly defenceless; e.g. 22 ... Kh8 23 Qh5 Ng6 (or 23 ... Bd6 24 Rbf1) 24 Rbf1 Rf8 25 Nf6, attacking the bishop and threatening 26 Nd7. Let’s examine another defence: 22 ... Ng6 23 Nf6+ Kh8 24 Qf3 (defending the knight and threatening 25 Rxc5) 24 ... Be7 25 Nxh7! and the knight is taboo, since Black would lose the queen. Question: Why do you say that it is understandable that Morphy didn’t choose 22 Rf5! - ? The move order, compared with 22 Qh5, doesn’t seem to make much difference. Answer: Agreed, it’s very difficult to discern the point, which is that after 22 Qh5 Black can reply 22 ... Rd6 (the only move to defend), whereas 22 Rf5 renders 22 ... Rd6 impossible, because the rook on d6 incurs a tactical weakness that can be exploited by 23 Rb5!! (a computer move which can easily escape attention, since White’s general idea is to concentrate his forces against the black king) and now 23 ... b6? loses to 24 Rxc5 bxc5 25 Ne7+, while 23 ... Rc6 is met similarly by 24 Rxc5 Rxc5 25 Ne7+ Kh8 26 Qd6 and White wins.

22 ... Rd6! 23 Bxh7+ Morphy seeks more than just an extra pawn in an ending after 23 Ne7+ Kh8 24 Nf5 Qg6 25 Qxg6 Rxg6 26 Rxb7. 23 ... Kf8 23 ... Qxh7 loses quickly to 24 Ne7+ Kh8 25 Qxe5+, or 24 Qxe5 at once. 24 Be4 Rh6 Now we see why it was important to prevent 22 043 ... Rd6. It wasn’t easy, was it?

25 Qf5 Qxg3 26 Rb2

Exercise: Löwenthal has defended very well so far. Now White threatens to win with 27 Rg2. What should Black play here?

26 ... Re8? Löwenthal cracks. This is a logical move, bringing an inactive piece into play and defending the insecure knight on e5, but Black has neglected to ask himself the all-important question: “What can my opponent do in reply?”. Answer: The correct continuation was 26 ... Qh3!, putting pressure on the f1-rook and gaining the huge relief of exchanging queens. After 27 Rg2 (27 Qf4 is answered by 27 ... Rh4) 27 ... Qxf5 28 Rxf5 Re8, Black would obtain the best position he has had all game.

27 Nf6 A simple move that wins material. 27 ... Re6 28 Rg2 White picks up the exchange and simplifies, which is more than enough to win. There were other ways, such as 28 Qxe6 fxe6 29 Nh5+ and 28 Nd7+ Ke8 29 Nxc5. 28 ... Qxg2+ If 28 ... Qh4, White wins with 29 Nd7+, among other things. 29 Bxg2 Capturing with the king was slightly better, gaining time on the game, but at this stage it’s not important. 29 ... Rhxf6 30 Qxf6 Rxf6 31 Rxf6 Ng4

Exercise: Here White has an accurate move which removes all doubt as to whether his position is winning. What should he play?

Answer: 044

32 Rf5! The pressure on the c5-bishop means that the check on f2 loses all its strength. 32 ... b6 Defending both the bishop and the b-pawn. After 32 ... Nf2+ 33 Kg1, none of the discovered checks are dangerous. 33 Bd5 Nh6 34 Rf6 Heading for the weakened c6-square. 34 ... Kg7 35 Rc6 a5 36 Rc7 Kg6 37 Kg2 f6 38 Kf3 Nf5

Exercise (easy): How to proceed? Answer: Exchanging pieces makes the technical task easier.

39 Be4 Kg5 40 Bxf5 Kxf5 41 h4 Advancing the passed pawn also helps to improve the situation of the white king. 41 ... Kg6 42 Rc6 Kh5 43 Kg3 f5 44 Rf6 f4+ 45 Kxf4 Bf2 46 Ke4 Bc5

Exercise (easy): White’s position is winning. What is the quickest way to conclude the game?

Answer: 47 Rf5+ After returning the exchange, the pawn endgame is an easy win for White, in view of the superior position of his king. 47 ... Kxh4 48 Rxc5 bxc5 49 Kd5 1-0 In December 1850 Paul Morphy immersed himself in his studies and played very little chess until 045

1857. Let’s look at one of his games from that period. Game 4 L.McConnell-P.Morphy New Orleans 1852 Evans Gambit [C52] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Ba5 6 0-0 The immediate 6 d4 is the alternative, as we saw in Game 2.

6 ... Nf6 7 d4 0-0 8 dxe5 Both H.Montgomery (New Orleans, blindfold simul 1857) and A.Mongredien (Paris 1859) played the erroneous 8 d5? against Morphy, closing up the game. After 8 ... Ne7, they tried respectively 9 Qc2 and 9 Qd3, but failed to gain sufficient compensation for the pawn. There are few modern games with this position, but judging by the examples from the 19th century, 8 Nxe5 appears to offer White more chances than his move in the game. 8 ... Nxe4

9 Ba3 Here 9 Bd5 has been suggested and it is true that, as with most gambits, great care is needed when capturing further material: if 9 ... Nxc3?! 10 Nxc3 Bxc3, White has 11 Ng5! and now Black’s best is to sacrifice his queen with 11 ... Bxa1 12 Qh5 Qxg5 13 Bxg5 Bxe5, hoping to survive the complications after 14 Bxc6 g6!. On the other hand, 9 ... Bxc3!? is fully playable. White’s compensation is more than doubtful after 10 Nxc3? Nxc3, followed by 11 ... Nxd5; while in the event of 10 Bxe4 Bxa1 11 Bxh7+ Kxh7 12 Ng5+ Kg8 13 Qh5, the queen sacrifice 13 ... Qxg5, followed by ... Bxe5, takes place under better conditions than in the previous line. 9 ... d6 10 exd6? Question: On move eight it was wrong to close the game, so why now is it a mistake to open things up? Answer: We are talking about two different evils here, but evils is what they both are.

With 8 d5, closing the game restricted the activity of White’s own pieces. In general, when you are a pawn down but have the more active pieces, it is appropriate to open the game, but you have to choose the best moment to do so. Right now taking on d6 facilitates Black’s development, since the e4-knight reaches an ideal square by recapturing on d6. It was better to maintain the tension with something like 10 Qc2, or else 10 Bd5, after which 10 ... Bxc3 11 Bxc6! bxc6 12 Nxc3 Nxc3 13 Qc2 Nb5 14 Bb2 is unclear. 10 ... Nxd6 11 Bb3 Bg4

046

The failure of White’s opening is clear; Black completes the development of his minor pieces and remains with an extra pawn, while White still has to develop his queen’s knight. 12 h3 Bh5 13 Qd5 Bg6 14 Ne5?! Question: From what we’ve seen, moving a developed piece for a second time is not something that Morphy would do, is it? Answer: White wants to obtain the bishop pair in order to have some compensation for the pawn. The problem is that it is not easy to improve his position. He could try 14 Re1, bringing an inactive piece into play, but after 14 ... Qf6, the situation is no better. Another possibility was 14 Nd4 (also moving a developed piece), but 14 ... Nxd4 15 cxd4 b6 (or the sharper 15 ... b5) still sees Black standing better, because White would have to lose time developing the b1-knight.

14 ... Nxe5 15 Qxa5 Gaining the bishop pair, but leaving his pieces rather uncoordinated. It would have been worse to play 15 Qxe5 Bb6, as the pressure on d6 could not be increased with 16 Rd1? due to 16 ... Re8 or 16 ... Qh4.

Exercise: How did Morphy improve the coordination of his pieces?

Answer: 15 ... Qg5! This defends the knight on e5, while increasing his activity on the kingside and gaining a dynamic advantage with the threat of 16 ... Nf3+. 16 Kh1 Exercise: What now?

Answer: 047

16 ... Be4! In only two moves Black has created serious threats against the white king, exploiting not only White’s lagging queenside development but also the absence of defenders on the kingside. 16 Qxc7 would have met with the same reply. 17 f3 17 Rg1 is no better, owing, among other things, to the natural ‘human’ move 17 ... Qh5, while the analysis engines indicate a quicker win with 17 ... b6, to push the white queen to a worse square and ensure that the e5-knight is not attacked; e.g. 18 Qb4 Qh6 19 Kh2 (with the white queen on a5 he could have played 19 Qxe5 Qxh3+ 20 Qh2) 19 ... Ng4+ 20 Kg3 Nf5+ 21 Kxg4 Qh4 mate.

Exercise: What did Morphy play in this position?

Answer: 17 ... Bxf3! Once again Morphy chooses the most ‘human’ continuation, which destroys the defences of the white kingside and gains a winning attack. It would again have been good to interpolate 17 ... b6!, forcing the white queen away from the knight on e5, so as, after 18 Qa4, to be able to play 18 ... Nf5!; e.g. 19 fxe4 Ng3+ 20 Kg1 Qe3+ 21 Rf2 Qe1+ and wins. 18 gxf3 Qg3 19 Nd2 White tries to bring defenders across to the kingside as fast as he can. Since it is impossible to defend h3, it would have made sense first to eliminate a potential attacker with 19 Bxd6, but after 19 ... Qxh3+ 20 Kg1 Qg3+ 21 Kh1 cxd6, it only remains for Black to bring a rook into the attack and in fact this cannot be prevented; e.g. 22 Qd5 Rae8 23 Qxd6 (or 23 Nd2 Re6; while if 23 Qe4, the e8-rook concludes matters after 23 ... Nd3! 24 Qxd3 Re5) 23 ... Qh3+ 24 Kg1 Nxf3+ 25 Kf2 and now 25 ... Ne1! 26 Rxe1 Qf5+ is the strongest continuation, with a forced mate according to the engines.

048

Question: White’s move looks more logical because, after all, the knight on d6 is pinned, isn’t it? Answer: I’ll reply to that with an exercise. Exercise: How did Morphy continue here?

Answer: 19 ... Nf5! The safety of the monarch is many times more important than material. After 20 Bxf8 Qxh3+ 21 Kg1 Qg3+ 22 Kh1 Ne3, White has no defence.

20 Rae1 Qxh3+ 21 Kg1 Rfe8 Logical and clear play – another piece joins in the attack, with devastating effect. However, thanks to the shaky defences of the white monarch and the tactical weakness constituted by the undefended white queen, Black could also have concluded the game without further assistance by playing 21 ... Qg3+ 22 Kh1 Qh4+ 23 Kg1 Qg5+ 24 Kf2 (or 24 Kh1 Ng3+) 24 ... Nd3+ 25 Ke2 Ng3+ and 26 ... Qxa5. There are other lines but no better defences.

22 Rf2 Anticipating the threat of 22 ... Nh4 23 Rf2 Qg3+ 24 Kf1 Nd3. Against the alternative 22 Re2, there are several ways to breach the defences: for instance, 22 ... Ng3 23 Rh2 Nxf3+! 24 Rxf3 Re1+, while 23 Rxe5 is refuted by 23 ... Ne2+! 24 Rxe2 (or 24 Kf2 Qh2+) 24 ... Rxe2 25 Bxf7+ Kh8!.

22 ... Qg3+ Morphy chooses to conclude the game in a similar fashion to what we have already seen, with all his pieces in the attack. Also decisive was 22 ... Ng3 23 Rh2 Nxf3+! 24 Nxf3 Rxe1+ 25 Nxe1 Qf1 mate, as pointed out by Maróczy, but you only need one way to win. 23 Kf1 Nd3 24 Rxe8+ Rxe8 25 Bxf7+ 049

A “spite check”, probably with only faint hope that Morphy would play 25 ... Kxf7?? allowing 26 Qxf5+. 25 ... Kh8 0-1

In 1855 Morphy returned to New Orleans and enrolled to study Law at the Law School of Louisiana. Morphy completed the course in a year and a half, and on the 7th of April 1857, at the age of 19, he qualified as a lawyer. As he wasn’t permitted to practise Law until he was 21, Morphy devoted himself to chess. When he received the invitation to the First American Congress, which would begin on the 5th of October 1857 in New York, he hesitated to accept, because he was affected by the recent death of his father, but he eventually agreed to participate.

050

Chapter Three The First American Chess Congress 1857 The first American Chess Congress took as its model the London tournament of 1851, with preliminary matches where the player who was first to three wins progressed (draws not counting), with the final being the first to five wins. The country’s best sixteen players took part. Morphy’s first opponent was James Thompson. Let’s look at the first game of this match. Game 5 J.Thompson-P.Morphy First American Congress (rd.1, game 1), New York 1857 Giuoco Piano [C50] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 d3 Nf6 5 Nc3 h6

It is more usual to play 5 ... d6, since 6 Bg5 h6 is not really to be feared. So 5 ... h6 is unnecessary and is basically a defensive move, but it has been played many times and, as we shall see, Morphy manages give it an aggressive character.

6 Ne2?! Question: Why is this dubious? It’s just a normal move, heading for g3, as in the Ruy Lopez, isn’t it? Answer: Yes, the idea is well known and it’s not bad, but that’s not enough; we need to see whether it’s appropriate to the position or if there’s some snag with it. The way to do this is to ask yourself: “what can my opponent do in response?”

It was preferable to choose either of the more popular moves, 6 0-0 or 6 Be3. 6 ... d6 Morphy is not seeking to exploit the disadvantage of 6 Ne2 immediately, which is the reduction of White’s control of the d5-square. Otherwise, he could have played 6 ... d5! straight away or after 6 ... 00 7 Ng3. 7 c3 With the idea of advancing d3-d4, or maybe b2-b4, at an opportune moment. 7 ... 0-0 8 h3 Question: Hmm, is this precaution better than 8 Ng3 or even 8 0-0 - ? Answer: Those moves are equally good or even superior. However, 8 h3 is not only a preventive move, it can also become an aggressive one. We should note that White has waited for Black to castle kingside and is now planning to advance with a timely g2-g4, before playing Ng3, perhaps with the intention of exploiting the “hook” on h6 with a further g4-g5.

8 ... Kh8

051

Question: This move looks rather slow, doesn’t it? Answer: You can rarely speed things up “just because you want to”. The position itself generally dictates what needs to be done. Besides, this is not really a passive move – bearing in mind White’s lagging development, Morphy is planning to open the game at the right moment with ... f7-f5, and this is the first step.

9 Ng3 To some extent admitting the failure of his previous aggressive ideas. Exercise: Why did White give up on the idea of advancing the g-pawn? Answer: Because, in response to 9 g4 (suggested by Maróczy), Morphy would have played 9 ... d5!, when White would soon have cause to regret the weaknesses created by his pawn thrust.

However, the preliminary 9 Bb3 was interesting, and only after 9 ... Nh7?! then 10 g4. It is useful for the bishop to be on b3, as we will see in the game, and its retreat anticipates an eventual ... d6-d5 attacking the bishop. 9 ... Nh7 Morphy continues with his idea, even at the cost of exchanging his c8-bishop. 10 Qc2 White tries to restrain 10 ... f5 by tactical means. There was nothing wrong with 10 0-0 f5 11 Nxf5, when it is not clear whether giving up his light-squared bishop is a good idea for Black. 10 ... f5 Morphy invites tactical complications, since he considers that the opening of the game will be in his favour. 11 exf5

052

Exercise: Black has several options now – what do you think was Morphy’s choice? Answer: Morphy elects to occupy the centre with gain of time.

11 ... d5 Exercise: Despite appearances (which induced White to play 10 Qc2), the capture 11 ... Bxf5 doesn’t lose. Can you see why not? Answer: After 12 Nxf5 Rxf5 13 d4, Black has 13 ... Rxf3! since 14 gxf3?! exd4 not only gives Black a pawn for the exchange, White’s pawn structure is damaged and his king is insecure. White would do better to play 14 dxc5, with a slight advantage after 14 ... Rf8 15 cxd6 cxd6 16 Be3, followed by Rd1 and 0-0. Exercise: 11 ... Ne7 was also playable. How should Black then respond to 12 Nh4, defending the f5-pawn?

Answer: With the rather surprising 12 ... g5!, exploiting the situation of the white king, which is still in the centre. After 13 fxg6 Bxf2+ 14 Kd1 Bxg3, the position becomes complex; in fact 15 Bxh6! is White’s only move to stay in the game. Then 15 ... d5 16 g7+ Kg8 17 gxf8Q+ Nxf8 leads to an unusual position where chances are balanced.

12 Bb3 Exercise: How did Morphy proceed here?

Answer: 12 ... e4 Morphy continues to advance. However, this move is not clearly better than 12 ... Bxf5, as after 13 Nxf5 Rxf5 14 d4, Black has the now familiar resource 14 ... Rxf3! 15 dxc5 (better than 15 gxf3?! exd4) 15 ... Rf8, with equal chances. 13 dxe4 dxe4

Exercise: We are in a critical position – how should White defend? 053

14 Ng1? This retreat is equivalent to surrender. White “believes” his opponent and now his position collapses. 14 Nxe4? was bad too, since the pin after 14 ... Bxf5 is decisive. Answer: The only way to resist was with 14 Qxe4!, when 14 ... Re8 15 Be6 Nf8 18 0-0 leaves Black with compensation for the pawn, but the struggle continues. Exercise: What is Black’s most effective move now?

Answer: 14 ... Ne5 And the game is over – the invasion on d3 is deadly. 14 ... Bxf5 was also good. 15 Be3 The black attack is equally devastating after 15 Kf1 Nd3 16 Nxe4 Bxf5 17 Nxc5 Nxc1 18 Qd1 Bd3+ 19 Ne2 Ne4. 15 ... Nd3+ 16 Ke2 Bxe3 17 fxe3 Qh4

This wins a piece at least. White could well resign here. 18 Nxe4 Qxe4 19 Qxd3 Qxg2+ 20 Kd1 Bxf5 Naturally, 20 ... Qxh1 wins as well. 21 Bd5 Bg4+ 0-1 The second game ended in another win for Morphy, this time in a rook and pawn endgame. Let’s look at the third game of the match. Game 6 J.Thompson-P.Morphy First American Congress (rd.1, game 3), New York 1857 Giuoco Piano [C54] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 c3

This time Thompson prefers to play c2-c3 immediately, instead of prefacing it with Nc3-e2, as in the first game. 4 ... Nf6 5 d3 d6

054

6 h3 A rather passive move; 6 ... Bg4 wasn’t a threat to be feared. More usual are 6 Bb3 and 6 b4 (frequently reached via 5 b4 Bb6 6 d3), although every reasonable move in the position has been tried here. Question: After 6 Bb3, is the pin 6 ... Bg4 not annoying? Please explain briefly how White should respond. Answer: It is worth noting that Kasparov faced 6 ... Bg4 numerous times in simultaneous displays; let’s take a look at one of these games. G.Kasparov-Da.Johnson, London (simul) 1998, followed a course similar to several others, involving a standard manoeuvre familiar from the Ruy Lopez: first of all White puts the question to the bishop, to force it to choose which diagonal to operate on. After 7 h3 Bh5, White is in no hurry to castle on the kingside but first sets out to bring his queen’s knight to g3, starting with 8 Nbd2, followed by Nf1, and then decides whether to play g2-g4 followed by Ng3, or Ng3 immediately, depending on how Black responds.

In the game Black opted for 8 ... Qd7, intending to castle queenside. Kasparov played 9 Qe2, a useful waiting move, and one that he was considering anyway in response to ... d6-d5 by Black. The game continued 9 ... 0-0-0 10 Nf1 d5 11 Ng3 Bxf3 12 Qxf3 d4 13 Bc2 Ne8 14 0-0 Nd6 15 Bd2 Rdf8? 16 b4 Bb6 17 c4 and White’s methodical attack on the queenside proved decisive. 6 ... Be6 7 Bb3 Question: Isn’t it better to double Black’s pawns with 7 Bxe6 rather than lose a tempo? Answer: It is a reasonable alternative, but it is not clearly better. The weakness of the doubled pawns cannot be exploited immediately. After 7 Bxe6 fxe6 8 Qb3 Qc8 9 Ng5 Black has 9 ... d5, showing the positive side of the exchange on e6 – Black strengthens his centre and, in addition, has the use of the half-open f-file, which could soon become an important factor, once he has completed his development.

Exercise: There are several playable moves here. What do you think was Morphy’s choice? Answer: Occupying the centre without delay, encouraged by the tempo White has devoted to playing 6 h3. 055

7 ... d5 It can’t be stated that this is definitely the best move, but it’s certainly the most ambitious. 8 exd5 This is playable, since the weakness created on d3 should not be significant. White could support e4 with our familiar 8 Qe2 or 8 Nbd2, but the most testing move here was 8 Ng5!, aiming to obtain the bishop pair to offset Black’s central superiority. This would demonstrate a good side to 6 h3, which avoids the resource ... Bg4. 8 ... Bxd5 9 0-0 0-0 10 Bg5 Bxb3 Morphy wants to neutralize the annoying pin. The immediate 10 ... h6?! allows 11 Bxd5 Qxd5 12 Bxf6 gxf6 13 Nh4, when the weakness of the black kingside is obvious.

11 axb3 This is the standard reply, opening the a-file; 11 Qxb3 is playable, but Black has few difficulties after the simple 11 ... Bb6 (if 11 ... Qxb3 then 12 Bxf6 gxf6 13 Qxb7) 12 Rd1 h6 13 Bh4 Qd6. 11 ... h6 12 Bh4

Exercise: Again, there are many playable moves here. What do you think Morphy chose?

Answer: 12 ... g5! Morphy could also have played the quiet 12 ... Bb6, in anticipation of b3-b4. Instead, he opted to seek activity, at the cost of weakening his kingside, reckoning that it is a risk worth taking. 13 Bg3 Exercise: How do you think Morphy intended to continue?

Answer: 13 ... e4! Of course; now that the knight is unpinned it would be awkward for White to play 14 dxe4 Nxe4, as Black then has strong pressure on f2 and g3. 14 Ne5! Best – the game is approaching its crisis. 14 ... Nxe5 15 Bxe5 exd3

056

Exercise: How should White proceed?

16 Bxf6? A weak move, regaining the pawn but allowing Black dangerous activity. Answer: It was better to play 16 Qf3!, fighting for the initiative and reminding Black that 12 ... g5 weakened his kingside. Then chances would be approximately equal after the natural 16 ... Nd5 17 Nd2 Re8 18 Bg3, planning to round up the pawn later; or if 16 ... Nd7 (heading for c5 or e5 at an opportune moment) then 17 Bg3 Re8 18 b4 (note that 18 Qxd3?! would be inaccurate, in view of 18 ... Nc5 19 Qc2 Qd3!).

16 ... Qxf6 17 Qxd3 Rad8 The black rooks quickly centralize, while the white knight remains on b1.

18 Qc2 Rfe8 Simple and strong; Black now has all his pieces in play. 18 ... Qg6! might have been been better, since after 19 Qxg6+?! fxg6, the pressure on f2 is very strong, with threats of 20 ... Rxf2 (or 20 ... Bxf2+) 21 Rxf2 Rd1+, or if 20 Na3 then 20 ... Rd2, so White must look for alternatives.

19 b4 Bb6 20 Na3

Exercise: All Black’s pieces are active and he now needs to find a way to utilize this advantage – how did Morphy do this?

Answer: 20 ... Qf4! With the obvious threat of 21 ... Rd2. 21 Rad1 Exercise: What was the other idea behind 20 ... Qf4 - ? Answer: It was to create threats along the h2-b8 diagonal. 057

21 ... c6! 22 Rd3? A tactical error that turns the game greatly in Black’s favour. Nevertheless, White’s position was inferior in any case. He could have tried 22 b3, preparing Nc4, but after 22 ... Bc7 23 g3 Qf3 24 Nc4 (not 24 Rfe1? Bb6) 24 ... b5 (simply 24 ... Re2 25 Rxd8+ Bxd8 26 Qd1 Bf6 is good too) 25 Rxd8 Rxd8 26 Nb2 (26 Ne3 Bxg3 and 26 Nd2 Qe2 are worse) 26 ... Bb6 27 Kh2, White is discoordinated and almost defenceless. One possible plan for Black would be to set his pawns rolling with ... h6-h5, intending ... h5-h4 or ... g5-g4, with ... Kg7 coming at some point, clearing the back rank for the rook.

Exercise (easy): What’s wrong with 22 Rd3 - ?

Answer: 22 ... Bxf2+ Winning a pawn, since 23 Rxf2 loses to 23 ... Re1+. 23 Kh1 Rxd3 24 Qxd3 Re3 Now White has to deal with the threat of ... Rxh3+ and mates. 25 Qd8+ Kg7 26 Qd4+ If 26 Qd7, defending h3, Black could play 26 ... Re2 or the more elegant 26 ... Bg3 27 Qd1 Qe4 with total domination. 26 ... Qxd4 27 cxd4 Re2 28 Nc4

28 ... Re1 A decision characteristic of Morphy: he chooses the simplest way. Other continuations were possible, such as 28 ... Bxd4 or 28 ... f5 29 Nd6 f4 30 Nxb7 Bxd4, both winning; but Morphy prefers to simplify the position, after which the bishop will be far superior to the knight, added to which his king will be very active, enabling Black to win without any complications. 29 Rxe1 Bxe1 30 Na5 Bxb4 31 Nxb7 058

Exercise: There are several possibilities now – how did Morphy proceed?

Answer: 31 ... Kf6 Of course, activating the king, something that White cannot do quickly. 32 Nd8 c5 33 Nc6 Ke6 34 dxc5 Bxc5 35 g4 Kd5 36 Nd8 f6 37 Kg2 a5

The target is the pawn on b2. The difference in strength between the bishop and the knight here is obvious, but no less important is the difference in activity between the kings. 38 Kf3 This allows Black to immobilize the b-pawn, but 38 b3 would not have saved White either: 38 ... Bb6 39 Nf7 Kc5 40 Nxh6 Kb4 41 Ng8 Bd8! and the black king captures the pawn on b3, winning easily. 38 ... a4 39 Ke2 Bd4 40 Kd3 Bxb2 The rest is very simple. 41 Nf7 Be5 42 Kc2 Kc4 43 Nd8 a3 44 Nb7 a2 45 Na5+ Kb4 46 Nb3 Ka3 0-1 So Morphy won his first match 3-0. His next opponent was the judge Alexander Meek, who had been a frequent opponent of Morphy’s in earlier days. We saw an extract from one of their games in Chapter One; the result of this match was also 3-0 in Morphy’s favour. His opponent in the semi-final was Theodor Lichtenhein. Morphy won the first game with Black in only 19 moves. Perhaps that explains the rather carefree spirit with which he played the second game. Let’s take a look at it. Game 7 059

P.Morphy-T.Lichtenhein First American Congress (semi-final, game 2), New York 1857 Bishop’s Opening [C42] 1 e4 e5 2 Bc4 Nf6 3 Nf3 Nxe4 4 Nc3

4 ... d5 Instead of trying to hold the extra pawn after 4 ... Nxc3 5 dxc3 f6, entering the Boden-Kieseritzky Gambit, Lichtenhein returns the pawn to seek a less complicated position.

Morphy himself, in a casual game in London in 1858 against Samuel Boden, one of the inventors of this gambit, varied with 5 ... c6, likewise returning the pawn, and emerged well from the opening after 6 Nxe5 d5 7 0-0 Bd6! (note that 7 ... dxc4? 8 Qxd8+ Kxd8 9 Nxf7+ Ke8 10 Nxh8 Be7 11 Re1 Kf8 would be unwise, since the white pieces are quickly activated, for instance with Re2, Be3-d4, Rae1, and it is not even clear whether Black can collect the knight on h8) 8 Re1 Be6?! 9 Bd3?! Nd7 10 f4 Nxe5 11 fxe5 Nc5+ 12 Kh1 Qh4, although the game ended in a draw. As it happens, White might have gained the advantage with 9 Qh5!, threatening 10 Nxf7! among other things, so the most sensible continuation for Black was simply 8 ... 0-0 9 Bd3 Nd7 with equality. We shall take a brief look at the retreat 4 ... Nf6 in Supplementary Game 7.1. 5 Bxd5 The theory of the time, namely the Handbuch, recommended 5 Nxd5, but this seems no stronger than Morphy’s move. 5 ... Nf6 6 Bb3 Bd6 7 d3 0-0 Morphy’s opponent also liked to play this line with White. Two games of the match T.LichtenheinH.Montgomery, Philadelphia 1861, continued 7 ... Nc6 8 Bg5 h6 9 Bh4 g5 10 Bg3 Bg4 11 h3 Bh5 12 Ba4 Qd7 13 Qe2 0-0 14 0-0-0 and here they diverged, with White winning one game and Black the other. 8 h3 This is to prevent the pin with 8 ... Bg4 (as might follow 8 0-0) which, with the queen’s knight already on c3 (therefore unable to support its colleague from d2) and the king’s bishop outside the pawn chain (and unable to disarm the pin with Be2) could be awkward. But that wasn’t White only idea, as we shall see. 8 ... h6 It is not clear that copying White and preventing Bg5 is better than playing 8 ... Nc6, since after 9 Bg5 h6 10 Bh4, Black can easily develop his pieces (for example, with 10 ... Bf5, perhaps followed by ... Re8 and/or ... Nd4), and even a timely ... g7-g5 doesn’t look dangerous for him.

060

Exercise: What idea did 8 h3 contain, in addition to ruling out ... Bg4 - ?

Answer: 9 Be3 This was the secondary idea. Of course 8 h3 prevents ... Ng4 as well, so the bishop can safely develop to e3, perhaps preparing to castle queenside which 8 ... h6 actually encourages, as we shall see. 9 ... Nc6 10 Qd2 Na5! A good decision – the white bishop could be very unpleasant when the approaching storm breaks.

Let’s put ourselves in Morphy’s shoes here. We know that, when he was facing considerably weaker opponents, he often took risks, relying on his superior calculating ability in complex positions – he had also won the first game of this match with Black in only 19 moves. Giving knight odds was an expression of the same approach; Morphy played many games at those odds. In fact, exactly two years later, he played six such games against Lichtenhein and defeated him 3½-2½. Exercise: Following the same logic, what do you think was Morphy’s choice here?

Answer: 11 g4 Encouraged by 8 ... h6, Morphy prepares to break with g4-g5. It is not clear whether this is objectively best, but it fits the risk-taking approach that Morphy opted for in this game, and to which he remains faithful even in the endgame. 11 ... Nxb3 12 axb3 Bd7 061

Exercise: What preparatory move did Morphy choose now?

Answer: 13 Rg1 Building up for the advance g4-g5, which Black prevents with his next move. 13 ... Nh7 14 Ne4 Exploiting the fact that the knight has retreated from f6, Morphy insists on preparing g4-g5 and deploys another piece to the kingside, rejecting the very reasonable alternative 14 0-0-0. 14 ... Kh8

15 g5 The moment has come to advance. Morphy prefers not to wait, again rejecting the less committal, but equally (or more) dangerous move, 15 0-0-0. 15 ... h5 Naturally, Black cannot allow the g-file to be opened. 16 Nh4 The pawn on h3 can be safely sacrificed, given that the open h-file would provide more than sufficient compensation. 16 ... g6

Black seals up the kingside – at a cost, since this move also weakens the long dark diagonal. It is not immediately apparent how this might be dangerous, but it should be kept in mind. 17 Qe2 This is a rather mysterious move. It doesn’t appear to be better than 17 0-0-0, when 17 ... Bxh3 is strongly met by 18 d4; while after 17 ... Bc6, White can play 18 f4! with the better prospects. 17 ... Bc6 This isn’t a bad move, even though Morphy eventually exploits the bishop’s absence from the c8-h3 diagonal. Instead, 17 ... Bxh3 was inadvisable in view of 18 0-0-0, followed by 19 d4; but it would have been interesting to try to open a second front with 17 ... a5 and ... a5-a4, emphasizing the fact that the white king is still in the centre, while anticipating his future queenside castling. 18 f4?! Steinitz called this “hazardous and unsound”, and he was very probably right. Nevertheless, the practical element should not be ignored; Black is under constant pressure and keeps having to take important decisions. At the risk of repeating myself, once again 18 0-0-0! was good.

062

Exercise: How can we try to demonstrate that Steinitz was right?

18 ... exf4 This move unleashes tremendous complications – the outcome is uncertain, but the black king comes under definite pressure. Answer: Max Lange suggested opening the game and exposing the white king with 18 ... Bxe4! 19 dxe4 Bb4+ 20 c3 exf4! (in this way Bd4+ is prevented) 21 Bxf4 Bc5 and now after 22 Be5+ Kg8, compared with the game, there are many open lines that endanger White’s king as much as Black’s, so the success of White’s attack is not guaranteed.

19 Bd4+ Kg8 Exercise: How did Morphy continue the attack?

Answer: 20 Nf5 Perhaps making Black regret having played 17 ... Bc6, although objectively that move wasn’t an error, as we’ve already noted. Now 20 ... gxf5 loses to 21 Nf6+, followed by 22 Qxh5. Since White is also threatening 21 Nh6 mate, Black’s reply is forced. 20 ... Re8 21 Nh6+ Kf8 Exercise: What was Morphy’s next step?

Answer: 22 0-0-0 Finally! At this point there was nothing better: Black was threatening 22 ... Bxe4 23 dxe4 Bb4+ and White’s position collapses after, for example, 24 Bc3 Bxc3+ 25 bxc3 Nxg5.

063

Exercise (difficult): How should Black defend?

Answer: 22 ... Bxe4? Black rejected the complicated 22 ... Nxg5!, against which Maróczy suggested 23 Nxf7, intending 23 ... Nxf7? 24 Rxg6. After, for instance, 24 ... Be5 25 Bxe5 Rxe5 26 Qf2! (not yet 26 Rdg1 on account of 26 ... Qd4!) 26 ... Qd7 27 Rdg1, White definitely has a strong attack.

But the merciless analysis engines immediately find 23 ... Kxf7!, and after 24 Nxg5+ (not 24 Rxg5? due to 24 ... Rxe4) 24 ... Qxg5 25 Qxe8+ Rxe8 26 Rxg5 f3 (threatening ... Bf4+) 27 Kb1 Re2, Black has the advantage, since the passed f-pawn is worth more than the lost exchange. 23 dxe4 Qe7

Exercise (difficult): What is White’s best continuation now?

Answer: 24 e5 This is the beginning of an attractive and complicated line, but is not the best. The “inhuman” move 24 Qc4! was stronger, threatening to win with 25 e5 because of the attack on f7. After 24 ... Qe6 25 Qc3!, threatening Bg7+ followed by e4-e5, White has a strong initiative. 24 ... Bxe5 25 Bxe5 Qxe5! Better than 25 ... Nxg5 26 Rge1!, when the complications favour White; e.g. 26 ... Qxe5 27 Qxe5 Rxe5 28 Rxe5 f6 29 Rxg5 fxg5 30 Rd7, with a clear advantage. 26 Rd7!

Exercise: This fine move is the key to White’s idea – but despite appearances, it’s not decisive. How must Black respond? 064

Answer: 26 ... Qg7? Justifying Morphy’s play. 26 ... Kg7? 27 Rxf7+ Kh8 also loses; i.e. 28 Rf8+! Nxf8 29 Nf7+ etc. The correct defence is 26 ... Nxg5! 27 Rxg5 Qf6!, when the position is still very complicated but seems to be objectively equal. Black has three pawns for the piece and the white knight is trapped, while the white king is not completely secure either. Let’s look at a few possibilities: After 28 Qd2 (the spectacular 28 Qxh5? fails to 28 ... Rad8! 29 Rxd8 Qxd8 and wins) 28 ... Rad8 29 Qb4+ c5! 30 Rxc5 Rxd7 31 Rf5+ Kg7 32 Rxf6 Kxf6 33 Qxf4+ Kg7, surprisingly, the analysis engines confirm that Black is fine despite the material disadvantage, since the knight is still trapped and White may have to resort to Nf5+ and try to give perpetual check before Black coordinates his rooks; but obviously this is a variation more suited to a computer. The simpler and more ‘human’ line is 28 ... Re7 29 Rd8+ Re8 30 Qb4+ Kg7 31 Nf5+ Kh7 32 Rxa8 Rxa8, which also offers equal chances. One possible continuation is 33 Qxf4 Re8 34 Qh4 Qe6 35 Rxh5+ gxh5 36 Qxh5+ Kg8 37 Qg5+ Qg6 38 Ne7+ Rxe7 39 Qxe7, reaching a level endgame.

27 Qc4 Threatening 28 Qc5+ as well a capture on f7. There is no defence. 27 ... Re7 28 Rxe7 Kxe7 29 Re1+ 1-0 Lichtenhein at least managed to salvage a draw (a pawn down in a rook ending) in the third game, but Morphy won the fourth to take the match 3½-½. Supplementary Game 7.1 P.Morphy-W.Potier Blindfold simultaneous, Paris 1858 1 e4 e5 2 Nf3 Nf6 3 Bc4 Nxe4 4 Nc3 Nf6 5 Nxe5 d5 6 Bb3 Be7 6 ... Bd6 7 d4 c6 was possible, but the text move is fine. Black can be satisfied that he has shut the bishop out of play on b3. 7 d4 c6 8 0-0 Nbd7 9 f4 Nb6 If Black wants to develop his queen’s bishop along the c8-h3 diagonal he needs to move the knight at some point, but it is better to postpone that decision. 9 ... 0-0 looks better. 10 Qf3 h5? Seemingly, Black was reluctant to play 10 ... 0-0 on account of 11 g4. That looks scary but is not particularly dangerous, as the knight can go to e8 after, for example, 11 ... Nc4 12 g5.

Exercise: How did Morphy respond?

Answer: 11 f5 Shutting the c8-bishop’s diagonal, while opening that of his own c1-bishop. 065

11 ... Qc7?! Black doesn’t want to castle short now, since it would be in worse circumstances than before, but this move helps White develop. 12 Bf4 Bd6 13 Rae1 Kf8?! It was better to play 13 ... 0-0 anyway and “hope for the best”.

Exercise: How did Morphy proceed?

Answer: 14 Qg3! With the threat of 15 Ng6+. 14 ... h4 Exercise: What was White’s strongest continuation after 14 ... Kg8 - ? Answer: There are several good moves, but the line that best exploits the awkward situation of the opposing pieces is 15 Ng4! hxg4 16 Bxd6 Qd7 17 Be7 and Black’s position is on its last legs.

15 Ng6+ Kg8 16 Bxd6 hxg3 17 Bxc7 fxg6 18 fxg6 gxh2+ Black maintains material equality but the difference in activity between the two sides is huge. 19 Kh1 Bg4 20 Re7 Nbd7 21 Be5 Kf8? This is a fatal waste of time; instead, in view of the threat of 22 Nxd5! and wins, 21 ... Rh5! was essential, to give the king the h8-square. 22 Rf7+ Kg8

Exercise: How did Morphy conclude the struggle in his favour? 066

Answer: 23 Nxd5! The bishop exiled to b3 takes its revenge. This is not the only move to win; he could also take on f6 first. 23 ... cxd5 24 Bxd5 Nb6 24 ... Nxd5 allows mate in one, while 24 ... Nxe5 25 R7xf6+ only delays the end. 25 Bb3 1-0 There is no good defence against 26 R7xf6+. Morphy’s opponent in the final was the German master Louis Paulsen, who was living in the USA and who was a very famous player of the time. Today a popular line of the Sicilian Defence bears his name. Let’s look at the first game. Game 8 P.Morphy-L.Paulsen First American Congress (final, game 1), New York 1857 Sicilian Defence [B40] 1 e4 c5 2 d4 cxd4 3 Nf3 e6 4 Nxd4 Bc5 Paulsen is planning to develop his king’s knight on e7. This line has been practically abandoned, displaced by 4 ... a6, 4 ... Nc6 and 4 ... Nf6. 5 Nb3 In the fifth game of the match Morphy switched to the stronger move 5 Be3 (see Game 9). 5 ... Bb6 6 Nc3 Ne7 7 Bf4

7 ... 0-0?! Exercise (easy): Demonstrate that Black’s last move is at least dubious.

Answer: 8 Bd6! Of course – the bishop is a very unpleasant intruder, hindering the natural development of the black pieces. Despite the strange appearance of Black’s set-up, he could have obtained a reasonable position after 7 ... d5 8 exd5 exd5, when the structure resembles that of a Tarrasch French arising from 1 e4 e6 2 d4 d5 3 Nd2 c5 4 Ngf3 Nc6 5 exd5 exd5 6 Bb5 Bd6 7 dxc5 Bxc5 8 0-0 Ne7, a line in which the recently departed but never to be forgotten Viktor Korchnoi was almost invincible as Black. 8 ... f5 9 e5 a6 This is a bad sign, another loss of tempo. Black wants to prevent White from playing Nb5 at some point, perhaps with the idea of replacing the bishop on d6. Here one of the suggestions of the analysis engines is 9 ... Ng6. 067

Exercise: How would you respond in that case? Answer: There are several interesting moves, such as 10 g3 or 10 Qd2, preparing the defence of the e5-pawn with 11 f4. On the other hand, it would not be the most practical (and possibly not objectively best either) to accept the exchange sacrifice straight away with 10 Bxf8 Qxf8, since there is no way to defend the e5-pawn and Black would have good compensation with a pawn for the exchange, the possibility of building a strong centre, and the strength of his dark-squared bishop.

10 Be2 Morphy is looking to castle quickly on the kingside and then see how to exploit the advantage provided by his strong bishop. This is the most sensible and simple plan, but it’s not the most energetic. There are more ambitious options, such as 10 Qd2 with the plan of castling queenside, or 10 f4, bolstering the e5-pawn and preventing the game continuation.

10 ... Nbc6 11 0-0 Rf7 Unpinning the rook, so as to be able play ... Ng6. 12 Kh1 Another idea was 12 Bh5 and if 12 ... Ng6 then 13 Qe2, supporting the vital e5-pawn, and the pin is very uncomfortable for Black, while 12 ... g6 13 Bf3, followed by Qd2, would fail to offer Black much relief either. Now 12 ... Ng6 can be answered by 13 f4 and Black can’t free his position, so Paulsen seeks another way to gain some activity. 12 ... f4

Question: This move creates weaknesses, while the bishop on c8 is still dormant. It looks like an emergency solution, almost smacking of desperation. Is that so? Answer: There is a saying that we can all be brave when there is no alternative. This move concedes the e4-square and weakens the b1-h7 diagonal. On the other hand, if White is allowed to play f2-f4, there will be no way for Black to complicate the play. So for good or ill, 068

f2-f4 must be prevented. Now ... Nf5 becomes possible, in addition to ... Ng6. 13 Ne4 Nf5

13 ... Ng6 is answered by 14 Bh5, when the exchange sacrifice would be under rather worse conditions than before, as after 14 ... Ncxe5 15 Bxe5 Nxe5 16 Bxf7+ Nxf7, White can restrain the black centre with 17 c4!. 14 Bh5 This seeks to weaken the black kingside, and succeeds, but without achieving much. Instead, 14 Nbd2 was interesting, and if 14 ... Nxd6 15 Nxd6 Nxe5 (after 15 ... Rf8 16 Nf3, the black forces remain bottled up) 16 Nf3! Nxf3 17 Nxf7 Kxf7 18 Bxf3 d5, then once again 19 c4!, and White’s greater activity is obvious. 14 ... g6 15 Bg4

15 ... Ng7? Unnecessarily passive. Löwenthal’s recommendation 15 ... Nxd6 16 exd6 (after 16 Nxd6 Rf8, nothing special is apparent) 16 ... Ba7 was preferable, followed by 17 ... b5, or 17 ... b6 if necessary, releasing the c8-bishop to b7, followed by ... Rc8. The snag is that the other bishop might become temporarily passive, but Black’s problems would be much diminished compared to the game, because there would be no target or serious weakness in his position.

16 Qf3 Question: What is the idea behind Morphy’s move? Is it the best? Answer: It is difficult to say what the best move was here. There were other ideas, such as simply 16 Qe2. Morphy, as was his custom, was seeking to bring his pieces into play quickly and he chose a flexible move that not only (as with 16 Qe2) releases the inactive a1-rook, but also envisages defending his centre with Qc3, if required.

16 ... h5?! Given Black’s lagging development, this move is simply too optimistic. Here such an attack with pawns does not inspire confidence. It looks more appropriate to play 16 ... Bc7 17 Qc3 Ne8. 17 Bh3 Qh4 White could have answered 17 ... g5 with 18 g4!, and if Black wants to prevent the opening of the g-file, he would have had to seal up the kingside with 18 ... h4, which leaves him paralysed and allows White a free hand in the centre and on the queenside after, say, 19 Bg2.

069

Exercise: Black’s set-up has a defect, which White can exploit with a precise series of moves. How should he begin?

Answer: 18 Nf6+ The first step – the knight settles on an ideal post, since 18 ... Rxf6 19 exf6 Qxf6 20 Qxf4 isn’t playable for Black. 18 ... Kh8 Exercise: What is Morphy’s next step?

Answer: 19 Qe4! Making Black rue his decision to play 16 ... h5?!, which weakened the g6-pawn. 19 ... Qg5 If 19 ... Nf5 then 20 Bxf5 gxf5 21 Qf3. 20 g3 This is another important move, seeking to open up the game. 20 ... f3 This move, which tries to keep the game closed, will be quickly punished. Naturally, opening the game is not attractive for Black, because White´s lead in development is overwhelming. Nevertheless, the engines suggest 20 ... fxg3 21 fxg3 Nf5 as a more tenacious defence, even if White is still clearly better after 22 Bxf5 gxf5 (not 22 ... exf5? 23 Qd5) 23 Qe2.

Exercise: What had Morphy thought up here?

Answer:

070

21 Nd2! The most elegant method of bringing another piece into the kingside campaign. There is no good way to defend the f3-pawn, as 21 ... Qxd2 22 Qxg6 loses straight away.

21 ... Bd8 22 Nxf3 Qh6 23 Rg1 Question: What a strange move! Answer: White’s position is very advantageous; while 23 Rg1 is not necessarily the best move, it has a definite idea behind it and does not worsen White’s position at all, despite its unusual appearance. Note that if Black elects to regain the pawn with ... Bxf6 etc, it will be at the cost of remaining with very weak dark squares. Meanwhile White needs to bring more forces into play. Morphy is setting up the possibility of a timely g3-g4, opening the g-file; e.g. 23 ... b6 24 g4! hxg4 25 Ng5! Qxg5 26 Rxg4 and wins.

23 ... Bxf6 24 exf6

24 ... Ne8 This loses; but capturing immediately with 24 ... Rxf6 would offer no hope of salvation after 25 Bf4 Qh7 (or 25 ... d5 26 Qe2 Qh7 27 Ne5) 26 Rad1 (or just 26 c4) 26 ... d5 27 Qd3, intending 28 c4. 25 Bf4 Nxf6 Neither 25 ... Qh7 26 Ng5 nor 25 ... Qf8 26 Qxg6 was any better. 26 Qxc6 Winning a piece and the game. 26 ... Qxf4 If 26 ... bxc6 27 Bxh6 Ne4 (or 27 ... Ng8 28 Bf4), then 28 Ne5 Nxf2+ 29 Kg2 Rf6 30 Rgf1 wins. 27 Qxc8+ Rxc8 28 gxf4 Rxc2

29 Rac1 Activity always! Instead of, say, 29 Ne5, followed by Nd3 defending everything (which of course wins), Morphy seeks to activate his pieces and bring the game to a rapid conclusion. 29 ... Rxf2 30 Rc8+ Ng8 31 Ne5 071

Resignation by Black would be far from premature. 31 ... Rg7 32 Nxg6+ Kh7 33 Nf8+ Kh6 34 Nxd7 Rxd7 35 Rcxg8 Rxf4 36 Bxe6 Re7 It’s now mate in four moves. 37 R8g6+ Kh7 38 Bg8+ Kh8 39 Rh6+ Rh7 40 Rxh7 mate In those days there were no chess clocks. Morphy played quickly; his rivals did not. Hence the games could last for more than ten hours. The second game in fact lasted fifteen hours, twelve of which were used up by Paulsen. It ended in a draw, which can be considered a surprise. Let’s look at part of this game. L.Paulsen- P.Morphy First American Congress (final, game 2), New York 1857

Black to play

It is easy to see that the opening has not gone well for White: four black pieces are in very menacing positions, pointed in the direction of the white king. It would be no surprise if the game were to be over in a few moves. 21 ... Rxg2+! The prettiest way to win; the less spectacular 21 ... f5 22 Ng3 Qh4 is very strong too. 22 Kxg2 f5 23 f3

23 ... Qg6+? A typical case of mixing up the move order. Löwenthal commented: “As soon as the second player had touched the queen, he remarked that, had he taken the knight, the contest could not have been prolonged a dozen moves. Black’s error consisted in reversing what should have been his 23rd and 24th 072

moves.” After the obvious 23 ... fxe4 24 Rxe4 (24 fxe4 Qg6+ 25 Kh1 Rf2 leads to a quicker mate) 24 ... Qg6+ 25 Kh1 Rxf3 or 25 ... Qg3, Black mates in a few moves. 24 Ng5 Now White will emerge the exchange up, although Morphy’s tenacious defence enabled him to draw after 53 moves. The rules stated that, in the event of a draw, the game should be replayed with the same colours. Morphy lost the third game. His bad patch continued into the fourth game, in which he wasted a winning advantage. Let’s now look at the fifth. Game 9 P.Morphy-L.Paulsen First American Congress (final, game 5), New York 1857 Sicilian Defence [B40] 1 e4 c5 2 Nf3 e6 3 d4 cxd4 4 Nxd4 Bc5 5 Be3 This time, instead of 5 Nb3 (as he played in the first and fourth games), Morphy opts for a developing move. 5 ... Qb6

6 Nb5

In the seventh game Morphy switched to 6 Nc3, which is the most popular move here. In recent years this position has been reached many times, albeit mostly via the order 4 ... Qb6 5 Nc3 Bc5 6 Be3, when the usual reply is 6 ... Nc6. Curiously, in a game with the white pieces the same year, Paulsen chose 6 b3 here, while towards the end of the 20th century, 6 Na3 and 6 c3 were often tried as well. After 6 Nc3, accepting the pawn with 6 ... Qxb2?? (as Paulsen played) loses to 7 Ndb5! and White wins at least a piece: if 7 ... Qb4 then 8 Rb1, or 7 ... Bb4 8 Bd2 and wins. Paulsen gave up his queen for a rook and a piece with 7 ... Bxe3 8 Rb1 Qxb1, but after 9 Nxb1 (inserting 9 Nd6+! is even stronger) 9 ... Bf4 10 g3 a6 11 gxf4 axb5 12 Nc3, Morphy went on to win quite easily. Question: 6 Nc3 is a typical Morphy move, developing; whereas the text move, sacrificing a pawn or allowing doubled pawns, looks untypical of him, doesn’t it? Answer: It’s an interesting decision. Morphy wants to exchange the defender of the dark squares, in particular the d6-square. The doubling of the e-pawns, while “ugly”, is not serious. After 6 ... Bxe3 7 fxe3, White gains control of d6 and the use of the half-open f-file, and also has the possibility of playing Qd4 at some point, with the probable restitution of his pawn centre. If Black accepts the sacrifice with 7 ... Qxe3+, White has clear compensation after 8 Be2 and threatens 9 Nc7+; while if 8 ... Na6, he can also continue development with 9 N1c3, keeping Nd6+ in reserve.

6 ... Nf6 Despite the comments in the previous note, Black’s best course may still be to play 6 ... Bxe3 7 fxe3 and then 7 ... Nc6. If White continues 8 Qd2, planning N1c3 and 0-0-0, Black can reply with 8 ... Nf6. 073

7 Bxc5 Qxc5

8 Nd6+? A very surprising decision, since losing the right to castle is no great handicap to Black in this position – he can easily develop his pieces, especially after the following exchange on c8. One would normally have expected Morphy to play the obvious developing move 8 N1c3!.

8 ... Ke7 9 Nxc8+ Rxc8 10 Bd3 Black’s lead in development means that 10 Nc3 is inadvisable on account of 10 ... Qb4!, with a double attack. In contrast to Paulsen’s 6 ... Qxb2?? in the seventh game (quoted above), the capture 11 ... Qxb2 would now leave the knight on c3 unprotected. 10 ... Nc6 Question: After what you’ve said, why didn’t Paulsen try 10 ... Qb4+ 11 Nd2 Qxb2 - ? Answer: This was playable, but it is not clear that it would be advantageous to Black, and it would certainly be more dangerous. The circumstances are rather different here: on d2 the knight is better placed than on c3, and after 12 Rb1 Qc3, White can play either 13 Rxb7 or 13 0-0, followed by f2-f4; or if 12 ... Qxa2, White has the further option of 13 e5, followed by Qg4, with great complications.

11 0-0 h5!? An ambitious move, Black is not content just to have a sound position (e.g. with 11 ... Ne5) and starts a bayonet attack, encouraged by the paucity of white defenders on the kingside, even though Black himself does not have numerical superiority on that wing. Logically, since “pawns can’t move backwards”, if Black’s offensive fails he will need to be careful that his pawn advances don’t create, or become, weaknesses.

12 Nd2 h4 13 h3 g5 Now that White has created a “hook” on h3, Black follows up consistently. 14 a3 Preparing to evict the black queen from the g1-a7 diagonal, so as to play f2-f3. 074

Question: Wasn’t it simpler just to play 14 Kh1 first? Answer: The problem is that the king possibly stands worse on h1. Black might continue with 14 ... Rg8 and if 15 f3 then 15 ... Nh5! is unpleasant.

Black now has several attractive options. 14 ... Rg8 This is a logical move, intending to support the advance 15 ... g4, but as White is able to prevent this, it was perhaps better to prepare it with 14 ... Ne5 which, as we shall see, would prevent the queen from being forced off the g1-a7 diagonal. It was even worth considering the immediate 14 ... g4, sacrificing the pawn, although if we go down the line suggested by the analysis engines, 15 hxg4 Rg8 16 b4 Qg5 17 f3 h3, Black is not better after 18 g3!.

15 b4 Qb6 16 Nc4 This could have been prevented by 14 ... Ne5. 16 ... Qc7 17 f3 Radically preventing ... g5-g4, at the cost of closing the kingside. 17 Ne3 was no better, on account of 17 ... Ne5! 18 f3 and now, for instance, 18 ... Nh5, followed by ... Nf4.

Entering complications with 17 Re1 g4 18 e5 Nxe5 19 Nxe5 gxh3 would not be the most practical course of action – the outcome is not exactly clear, but Black has obvious compensation for the piece.

17 ... Ne5 Black exchanges knights in order to remain with the superior minor piece, because it’s not easy for White to alter the pawn structure to improve the mobility of his bishop. This decision seems a correct one, although it’s surprising that Paulsen never gets round to playing ... Nh5, heading for the very attractive square f4, for which this was a favourable moment. Alternatively, 17 ... d5 has been suggested, but the opening of the game doesn’t appear to favour Black after, for example, 18 Ne3 dxe4 19 Bxe4 Nxe4 20 fxe4 Ne5, when there are several moves: 21 075

Qh5, 21 Qe2, or even 21 c4, since 21 ... Nxc4?? 22 Nxc4 Qxc4 loses to 23 Rc1. 18 Nxe5 Qxe5 19 Qd2 Rg7?! A strange move, the point of which is unclear. Perhaps he wanted to have f7 defended in case of a future f3-f4 by White. Paulsen is not satisfied with the equal endgame arising after 19 ... Qd4+ 20 Qf2, but can’t find the way to increase his minimal advantage. Once again 19 ... Nh5!, blockading the kingside, looks a better idea. 20 Rad1 Rd8 Preparing the possibility of playing 21 ... d5. 21 Qf2 b6

With his previous move Morphy forced Black to create a weakness on the queenside. Now the light squares are slightly weaker, even if the position remains balanced. Exercise: How do you think Morphy continued?

Answer: 22 f4! Making Paulsen regret that he never got round to playing ... Nh5. 22 ... gxf4 23 Qxf4 Qg5 24 Rf2 Qxf4 25 Rxf4 Rdg8 If Black has decided against the advance 25 ... d5, which leads to a drawn ending (e.g. 26 exd5 Nxd5 27 Rxh4 Ne3 28 Rd2 Nxc2 29 Rxc2 Rxd3 or 27 ... Rxg2+ 28 Kxg2 Ne3+ etc), it was better to play 25 ... Rh8 immediately. 26 Rd2 Rh8

Exercise: Granted that there is no great imbalance in this position, what do you think Morphy played now? 076

Answer: 27 e5! Hindering Black from playing 27 ... d6, followed by ... Nd7-e5, which would improve the coordination of his pieces considerably. Admittedly, the black knight now becomes active, but so does the white bishop. 27 ... Nd5 28 Rd4 There is no good way to prevent Black’s next move; for instance, after 28 Re4 (so that ... f7-f6 and ... f6xe5 does not attack the rook) 28 ... f6 29 Bc4, instead of 29 ... Nc7 30 Red4, Black can improve with the counter-attack 29 ... Rg3!, intending 30 Bxd5 f5! 31 Red4 exd5 32 Rxd5 Rxa3 33 Rxd7+ Ke6, which should be drawn. 28 ... f6 29 exf6+ Nxf6 30 Rc4 Kd8 31 a4

31 ... Nd5?! This isn’t a serious error, but why allow the white bishop greater scope? He could have played 31 ... Rg5, preventing a4-a5 and preparing ... e6-e5, or else prioritized the advance of his central majority with 31 ... e5 32 Be4 Ke7 33 Rc7 Ke6.

32 Be4 Nc7 Now everything is ready for Black to play ... d7-d5. It is no surprise that opening the game with, for example, 32 ... Nf4 33 Bb7 Rf8 34 Rc8+ Ke7 35 Rxd7+! Kxd7 36 Rxf8 Nxh3+ 37 Kh2 Ng5 38 Rf4 favours White’s bishop, leaving him with the more pleasant position.

33 Bf3 d5 34 Rc6 Morphy concentrates all his forces on the e6-pawn to tie the black pieces down and gain greater freedom of action for himself. 34 ... Rd7 Question: Why not 34 ... Kd7, which defends the pawn with gain of time? Answer: Because, after 35 b5, the threat of 36 c4 is very unpleasant.

35 Bg4 Rh6 36 Re2 Ke7

077

White has activated his pieces as much as possible, while Black’s optimistic ideas have not been fulfilled. Nevertheless, White has still not achieved anything significant. Exercise: How did Morphy continue pressing for an advantage?

Answer: 37 Kh2! Preparing g2-g3 to create a passed pawn on the h-file. 37 ... Kf7 38 g3 hxg3+ 39 Kxg3 Re7 40 h4 “Mission accomplished”, although Black’s defensive resources should certainly be sufficient. 40 ... Ne8 41 h5 “Passed pawns must be pushed”, it’s true, but the pawn is weaker on this square, Instead White could continue to manoeuvre with 41 Re5. 41 ... Nf6 42 Rcxe6 Rxe6? Another inaccuracy. With the zwischenzug 42 ... Rc7!, threatening 43 ... Nxg4, Black could regain the pawn and equalize more easily; e.g. after 43 R6e5 Rc3+!. 43 Bxe6+

Exercise: What is the best flight square for the king?

43 ... Kg7? Paulsen finally cracks under the pressure. Answer: With 43 ... Kf8! he could have reached a rook ending a pawn down but with considerable drawing chances after, for instance, 44 Re5 Nxh5+ 45 Kf3 Nf6 46 Bxd5 Nxd5 47 Rxd5 Rc6, as pointed out by Karsten Müller.

44 Bg4 Nxh5+ 45 Bxh5 Rxh5 46 Re7+ 078

This is what Black could have avoided with 43 ... Kf8!. 46 ... Kf6 47 Rxa7

Another rook ending with three pawns against two, but this one is winning for White, who quickly gains two passed pawns. 47 ... Ke5 This loses another pawn. No better was 47 ... Re5 48 Ra6! Re3+ (or 48 ... Re6 49 a5) 49 Kf2 Rc3 50 Rxb6+ Ke5 51 a5 and the connected white pawns triumph; e.g. 51 ... Rxc2+ 52 Ke1 d4 53 a6 Ra2 54 b5 d3 55 Rb7 Kd4 56 a7 Ra1+ 57 Kd2 Ra2+ 58 Kc1 d2+ 59 Kd1 Kd3 60 Rd7+! Kc3 61 b6 etc. 48 Ra6 Rg5+ 49 Kf3 Rf5+ 50 Ke2 b5 51 axb5 Rf4 Or 51 ... Rf8 52 Rc6!, followed by 53 b6. 52 c3 d4 53 c4 Rh4 54 c5 Rh2+ If 54 ... d3+ 54 Kxd3 Rxb4, the simplest is to give up a pawn in order to cut off the black king on the fifth rank with 55 Rd6 Rxb5 56 Kc4 Rb1 57 Rh6, followed by c5-c6, with an easy win. 55 Kd3 Rh3+ 56 Kc2 White has few technical problems here, as long as he avoid traps such as 56 Kc4?? Rc3 mate. Although the three white pawns should win easily, one has to remain constantly alert. 56 ... Rh2+ 57 Kb3 Rh3+ 58 Ka4 Kd5 59 Rd6+ Kc4 60 c6 Rh1 Threatening mate. 61 Rxd4+ 61 Ka5 also wins, albeit more prosaically. 61 ... Kxd4

Exercise: The final hurdle – how does White win?

Answer: 079

62 c7! The only way. Not 62 b6? Kc4 63 Ka3 Kc3 64 Ka2 Rh2+ 65 Ka3 Rh1 66 Ka4 Kc4 and draws; nor 62 Ka5? Kd5 63 c7 Rh8 64 Ka6 (definitely not 64 b6?? Kc6 and Black wins) 64 ... Kd6 65 b6 Kc6 66 b5+ Kd7 67 Kb7 Rg8, again drawing, as pointed out by Karsten Müller. 62 ... Rh8 63 b6 Kc4 64 b7 1-0 And so we arrive at the sixth game, which is one of the most famous and most beautiful of his career. Game 10 L.Paulsen-P.Morphy First American Congress (final, game 6), New York 1857 Four Knights Game [C48] 1 e4 e5 2 Nf3 Nc6 3 Nc3 Nf6 4 Bb5 Bc5

This is the so-called Marshall Variation, although as we can see it was played well before Frank Marshall employed it. Both 4 ... Bb4 and 4 ... Nd4 gained popularity several decades later. 5 0-0

Exercise: The opening theory of the time would have been running out around here. How do you think Morphy responded to the threat of 6 Nxe5 - ?

Answer: 5 ... 0-0 Sacrificing the pawn for rapid development was surely the first idea that sprang to mind. The black pieces quickly become active and it is quite possible that the sacrifice is only temporary anyway. The alternatives 5 ... Qe7 and 5 ... d6 6 d4 exd4 7 Nxd4 Bd7 may be playable, but Black will not gain such activity as in the game. 6 Nxe5 Re8 “For the pioneers it is always difficult,” commented Kasparov, who considered it dubious not to regain the pawn straight away. The main line nowadays is 6 ... Nxe5 7 d4 Bd6 8 f4 and now either 8 ... Neg4 or 8 ... Nc6 9 e5 Be7 (or 9 ... Bb4).

080

Exercise: What should White do about the attacked knight?

7 Nxc6?! Simplifying, and at the same time doubling the opponent’s pawns, doesn’t look bad, but White will lose further tempi and Black will achieve what he wants, which is to activate his pieces. Answer: It’s been known for a long time that 7 Nf3! offers some advantage, since after 7 ... Nxe4 8 Nxe4 Rxe4 9 d3 Re8 10 d4, White is better developed.

The alternative 8 d4 is perhaps more questionable. After 8 ... Nxc3 9 bxc3 Bf8 (better than 9 ... Be7?! 10 d5 Nb8 11 Bf4, G.Maróczy-H.Pillsbury, Nuremberg 1896) 10 d5, then instead of 10 ... Ne5, which helps White’s mobilization by 11 Nxe5 Rxe5 12 Bf4 Re8 13 Qf3 c6 14 Bd3, the more modest 10 ... Ne7 is better, when White’s advantage looks minor. 7 ... dxc6 8 Bc4 b5?! Before regaining the pawn, Black forces the bishop to decide which diagonal it wants to remain on. Exercise: Why not 8 ... Nxe4 immediately?

Answer: Owing to the weakness of f7. White could play 9 Bxf7+! Kxf7 10 Nxe4, since 10 ... Rxe4 loses to 11 Qf3+.

But Black does have an alternative here, which would definitely suit Morphy’s style and is even slightly better for Black: 8 ... Ng4!, as in H.Erskine-C.Tattersall, correspondence 1909, eyeing both h2 and f2 and threatening 9 ... Qh4. After 9 Be2? Qh4 10 Bxg4 Bxg4 11 Qe1 Bf3!, Black gains a decisive attack. No better is 9 h3? because of 9 ... Nxf2! and if 10 Rxf2 Bxf2+ 11 Kxf2, Black wins with 11 ... Qd4+. Exercise: How should Black reply to 10 Bxf7+ in the second line? 081

Answer: Not with 10 ... Kxf7? due to 11 Qh5+ and 12 Qxc5, nor with 10 ... Kf8? 11 Bxe8 and the knight is pinned. The correct continuation is 10 ... Kh8! 11 Rxf2 Bxf2+ 12 Kxf2 Rf8, when Black gains a material advantage.

9 Be2 The bishop would remain out of play after 9 Bb3 Bg4 10 Qe1 b4 11 Nd1 Rxe4 or 11 ... Nxe4. 9 ... Nxe4 10 Nxe4 Exercise: Show your grasp of the tactical themes present in the position. How would you punish 10 Bf3? - ?

Answer: 10 ... Nxf2! 11 Rxf2 Qd4 is decisive, since 12 Qf1 Qxf2+! 13 Qxf2 Re1 is mate, while 12 Ne4 loses to 12 ... Rxe4!.

10 ... Rxe4 11 Bf3 This is a good choice. So too was 11 c3, as suggested by Steinitz, intending d2-d4 and keeping Bf3, followed by g2-g3, as a defensive resource. 11 ... Re6 12 c3?!

A surprising error, and a clear example of not using “prophylactic thinking”; i.e. not asking oneself “What can my opponent do in reply?”. The natural 12 d3 was appropriate, as was the pawn sacrifice 12 d4!, when 12 ... Qxd4 13 Be3 Qxd1 14 Rfxd1 gives White a slightly freer game with sufficient compensation. Exercise: How can Black punish White’s mistake?

Answer: 12 ... Qd3! Morphy must have asked himself “What is my opponent planning?”, to which the answer is 13 d4, and so he replied with a move that hinders it.082The bishop on c1 is the immediate victim, and

subsequently the rook on a1. 13 b4?! It was better to play 13 Re1, aiming to exchange a pair of rooks before Black’s build-up of major pieces in the centre becomes too much to bear. White would follow up the rook exchange with Qf1. 13 ... Bb6 14 a4?! Once again 14 Re1 deserved consideration. 14 ... bxa4 15 Qxa4

Exercise: How can Black continue to activate his pieces?

Answer: 15 ... Bd7? A very natural move, preparing 16 ... Rae8. Unfortunately, this time it is Morphy who fails to think prophylactically. Exercise: How could Paulsen have punished Morphy for his inaccuracy?

16 Ra2? “A fatal error” – Kasparov. Both sides have made the mistake of playing with only their own plans in mind, not looking for ways to restrict the enemy forces. Answer: White needed to dislodge the annoying intruder from d3 with 16 Qa6!, when the advantage would have switched to his side. Once the black queen leaves the blockading square White is able to play d2-d4, shutting the b6-bishop out of the game and highlighting the hitherto unimportant weakness of Black’s queenside. If then 16 ... Qxa6 17 Rxa6 Rae8, White can play 18 Bg4! (not yet 18 d4? due to the surprising response 18 ... c5! 19 bxc5 Bb5) 18 ... Bc8 19 Ra1 Rf6 20 Bxc8 Rxc8 21 d4 with an obvious advantage. Black does no better with 16 ... Qf5 17 d4 Rae8 18 Be3 c5 19 bxc5 Bxc5 20 Qb7! (or 20 Qa2!) 20 ... Bb6 21 c4 and White again has the advantage.

Going back to move 15, it is now clear that the correct move was the ‘ugly’ 15 ... Bb7!, which seems to sideline the bishop well away from the focus of the struggle, but in fact it fulfils a more important function, which is to keep control of the a6-square. White does not have time to play 16 Ra2, followed by Qc2, because of 16 ... Rae8 (threatening 17 ... Qxf1+! 18 Kxf1 Re1 mate) 17 Qd1 Ba6! 18 Rxa6 Qxa6 19 d4 Qc4 20 Bd2 a5, solving the problem of the inactive bishop on b6, with a decisive advantage to Black.

083

16 ... Rae8 Once again threatening 17 ... Qxf1+ etc. 17 Qa6 “Paulsen found the correct idea after all, but for some reason a move later. At that time tempo play was still unusual!” – Kasparov. If instead 17 Qd1, Black is able to activate his d7-bishop with 17 ... c5! 18 bxc5 Bxc5 19 Ba3 (19 Bg4 f5 only postpones ... Bb5) 19 ... Bxa3 20 Rxa3 Bb5 and wins.

Exercise: How did Morphy demonstrate that in chess “a tempo is an eternity”? Answer: All the black pieces are ready to join in a mating attack, exploiting the isolation of the a2-rook and the c1-bishop from the defence, as well as the absence of the white queen.

17 ... Qxf3!! 18 gxf3 Rg6+ 19 Kh1 Bh3 20 Rd1 The threat was 20 ... Bg2+ 21 Kg1 Bxf3 mate. It wasn’t possible defend against this with 20 Rg1 owing to 20 ... Rxg1+ 21 Kxg1 Re1+ etc. Exercise: What is the best answer to 20 Qd3 - ? Answer: Black wins with 20 ... f5!, preventing the counter-sacrifice Qxg6. Now if 21 Qc4+ then 21 ... Kf8! wins (not 21 ... Kh8? on account of 22 Qf7!), while after 21 Rd1 Bg2+ 22 Kg1 Bxf3+ 23 Kf1 Bxd1, the attack is irresistible. The white pieces still lack any coordination and are unable to arrive in time to defend with 24 Qc4+ Kh8 25 d4 in view of 25 ... Bf3, threatening 26 ... Bg2+ again.

20 ... Bg2+ 21 Kg1 Bxf3+ 22 Kf1

22 ... Bg2+ As Zukertort pointed out, there was a quicker win with 22 ... Rg2!; e.g. 23 Qd3 (attacking the bishop to prevent 23 ... Rxh2, followed by mate on h1) 23 ... Rxf2+ 24 Kg1 Rg2+ 25 Kh1 Rg1 mate. 23 Kg1 Bh3+ 084

Although it scarcely matters at this stage, it was quicker to play 23 ... Be4+ 24 Kf1 and now the switchback 24 ... Bf5! 25 Qe2 Bh3+ 26 Ke1 Rg1 mate. 24 Kh1 Bxf2 25 Qf1 Bxf1 26 Rxf1 Re2! In addition to being two pawns down, White’s king remains weak. 27 Ra1 Rh6 28 d4 Be3 0-1

The seventh game ended in another victory for Morphy, following a serious error by Paulsen in the opening (as we saw in the notes to Game 9); and Morphy won the eighth as well, so that the match finished 6-2 (or 5-1 discounting the drawn games, according to the tournament rules). Paul Morphy thus became the first champion of the USA. 100 years later another player destined to reach the pinnacle of world chess, Robert Fischer, also won the US championship at the start of his career. After this first great success, Morphy settled in New York, where he played 161 games at odds (+107, -36, =18) and 100 on even terms (+87, -5, =8). His opponents were the best chess players in the country: Paulsen, Stanley, Lichtenhein and Schulten. Let’s now view some games from that period, before Morphy’s first visit to Europe. Possibly one of the best known games is the following: Game 11 J.Schulten-P.Morphy Blindfold game, New York 1857 King’s Gambit [C32] 1 e4 e5 2 f4 d5 3 exd5 e4

It should come as no surprise that instead of accepting the gambit, which is the most popular continuation at present, Morphy opts to sacrifice a pawn himself by playing the Falkbeer CounterGambit, seeking rapid development. 4 Nc3 Opening theory was still in its early stages of development in those days. Subsequently, from the late 1960s onwards, it was considered better to play 4 d3 Nf6 5 dxe4 (Keres also tried 5 Nd2 a few times) 5 ... Nxe4 6 Nf3 (alternatively, 6 Be3, preventing ... Bc5, was employed by, among others, the most significant exponent of the King’s Gambit in the second half of the twentieth century, Boris Spassky) 6 ... Bc5 7 Qe2 Bf5 8 Nc3 Qe7 9 Be3, as suggested by Tartakower; the most famous game featuring this variation is D.Bronstein-M.Tal, USSR Team Championship, Riga 1968. 4 ... Nf6 5 d3 Exercise (easy): What move did Morphy play now?

Answer: 5 ... Bb4 085

Of course – now there is no doubt about the best square for this bishop. That’s why Keres used to played 5 Nd2, avoiding this pin. 6 Bd2 Neutralizing the pin. The line 6 dxe4 Nxe4 7 Qd4 Qe7 is harmless for Black. After 8 Be2 0-0 9 Bd2 Nxd2 10 Qxd2, as well as 10 ... Bg4 and 10 ... c6 (as played in C.Von Bardeleben-J.Blackburne, 3rd matchgame, London 1895), 10 ... Bc5 is attractive, with the threat of 11 ... Be3.

Exercise: How did Morphy respond to the threat of 7 Nxe4 - ?

Answer: 6 ... e3!? “Entirely in Morphy’s style! We have here a splendid example of a positional sacrifice,” commented Euwe. Black secures the superior development and the opening of the e-file in his favour, albeit now at the cost of two pawns. Nevertheless, it is not clear that this is the best move. Subsequently, the safer line 6 ... exd3 7 Bxd3 0-0 was played many times. Alternatively, in B.Spassky-D.Bronstein, Moscow 1971, Black opted for 6 ... 0-0, allowing 7 Nxe4, and after 7 ... Re8 8 Bxb4 Nxe4 9 dxe4 Rxe4+ 10 Be2 Rxb4, he had regained the sacrificed material with no problems – as Kasparov pointed out, this is “a variation by Falkbeer!”; the game continued 11 Nf3 Rxf4 and was drawn shortly afterwards. More recently, in P.Ponkratov-T.L.Petrosian, Martuni 2010, Black preferred 11 ... Qf6, which is also adequate, and obtained reasonable play after 12 0-0 Rxb2 13 Bc4 Bg4 14 Qe1 Nd7 15 Ne5 Bf5. 7 Bxe3 0-0 8 Bd2 Question: Hmm, moving the bishop once again? Is there nothing better? Answer: This isn’t a bad move; the bishop is exposed on e3 and would probably have to shift in any case after ... Nxd5 or ... Re8, so playing a move which will soon be ‘forced’ allows White more options on the following moves.

Schulten later tried to improve White’s play with 8 Ne2 in J.Schulten-I.Kolisch, Paris 1860, but his position deteriorated after 8 ... Re8 9 Bd2 Bxc3 10 Bxc3 Nxd5 11 Qd2, and now, instead of Kolisch’s 11 ... Ne3, Black could have played 11 ... Nxc3, leaving White’s dark squares weak; Black could follow up with ... Nd7-f6, intending either ... Nd5 or ... Ng4. Finally, if White tries 8 Be2, Black can reply in similar fashion to what we’ve already seen: 8 ... Bxc3+ 9 bxc3 Nxd5 10 Bd2 and here, among other things, Black has 10 ... Qf6 with a double attack.

086

8 ... Bxc3 Question: I don’t understand this; it doesn’t look a very ‘Morphy-like’ move. Why exchange the developed bishop? Answer: You’re right, this is not a move that Morphy would normally choose voluntarily. Here he makes it for tactical reasons, to prevent the knight from blocking the e-file in lines such as 8 ... Re8+ 9 Be2 Bg4 and now 10 Ne4!, when after 10 ... Bxd2+ 11 Qxd2 Nxe4 12 dxe4 Rxe4 13 0-0-0, White would gain the advantage. 8 ... Nxd5 9 Nxd5 Re8+ 10 Ne2 Bxd2+ 11 Qxd2 Qxd5 12 c4, followed by 0-0-0, doesn’t appear to grant Black enough compensation for the pawn either.

9 bxc3 Re8+ Morphy opts to bring another piece into play. Possible was 9 ... Nxd5, regaining one of the pawns and allowing a quick ... Nc6, while if 10 Nf3, Black has the familiar idea 10 ... Qf6!. However, White can pose more problems with a move that would be difficult to choose over the board (albeit much less difficult, once you realize that it’s almost forced): 10 c4! appears to weaken the position, but in his book Kasparov describes it as “more active”. After 10 ... Re8+ 11 Be2, the direct 11 ... Ne3 is answered by 12 Bxe3 Rxe3 13 Kf2! Re8 (not 13 ... Qd4? 14 Nf3 Qxf4 15 Qc1, winning the exchange, A.Petrov-I.Schumov, St Petersburg 1862) 14 Nf3 and, according to Kasparov, “Black still has to find real compensation for the pawn (for example, 14 ... Bg4 15 Qd2 Bxf3 16 Bxf3 Qd4+ 17 Kg3! etc).” The attempt to improve Black’s play with 11 ... Nf6 12 Nf3 Qe7 13 Ne5 Nc6 is strongly met by Estrin’s 14 Bc3!, or similarly 12 ... Nc6 13 Bc3! (not 13 0-0 Qe7! and White has to return the pawn with 14 Ne5, since 14 Re1? loses to 14 ... Qc5+! followed by ... Ng4), as once again the invasion on e3 is not as promising as it looks after 13 ... Ng4 14 Qd2 Ne3 15 Kf2. 10 Be2 Bg4 Again 10 ... Nxd5 can be met by 11 c4!, as in the previous note. After 11 ... Ne3 12 Bxe3 Rxe3 13 Nf3 Qe7 14 Kf2, Black no longer has a lead in development and remains a pawn down.

11 c4?!

087

Question: But now this is dubious, is it? Why is that? What should he have played? Answer: Clearly, this version of the c3-c4 idea does not fit into the category of “more active” that Kasparov mentioned. It is instead an extremely optimistic move: White maintains his two pawns advantage and at the same time prevents the f6-knight from coming into play with ... Nxd5.

Despite the position not being very open yet, the pin on the e-file is something that a modern master would try to deal with as a priority, even though there is no immediate danger. There are two reasonable moves to try to resolve the problem of the pin, one is 11 h3, and the other is the king move that by now will come as no surprise, 11 Kf2. Analysis shows that both moves are playable, and in fact White achieves a reasonable position in both cases. After 11 h3, rather than taking on e2 at once, Estrin’s suggestion of 11 ... Qxd5! looks better; for example, 12 Kf2 (of course not 12 hxg4? Qxg2) 12 ... Bxe2 13 Nxe2 and now, instead of 13 ... Qc5+ 14 Kg3, it might be better to play the simple 13 ... Nc6, keeping the check on c5 in reserve, with such ideas as the manoeuvre ... Ne7-f5, doubling rooks on the e-file, playing ... h5-h4, etc. In the light of this, the immediate 11 Kf2 makes more sense. After 11 ... Bxe2 12 Nxe2 Qxd5, as well as 13 h3 (transposing to the previous line) White can choose between 13 Rf1 or 13 Re1 with a complex position, apparently with chances for both sides. White is a pawn up, but his king is rather insecure and he has several weaknesses.

Exercise: White is relying on his material advantage; we know that Morphy, both in general and in particular in this game, approached the position in a different way. How do you think he continued now?

Answer: 11 ... c6! Seeking open lines and greater activity, bringing his queenside pieces into play. 12 dxc6? Question: This move obviously helps Black’s development and looks bad to me. The extra pawn doesn’t look so important now. What’s the explanation? Answer: Yes, you’re right. Kasparov’s classic description was: “the move of roughly a third category player.” It seems clear that some of Morphy’s opponents did not have the same understanding as he did of the value of time and the importance of development, although as Fischer pointed out, “Morphy was not responsible for his opponent’s mistakes.” Quite simply, chess is more advanced now, thanks precisely to the lessons of Morphy and other great players.

As Kasparov pointed out, “Absolutely essential was 12 h3 Bxe2 13 Nxe2 cxd5 14 cxd5 Qxd5 15 0-0 with an extra pawn, for which Black has some compensation, but not more.” The text move loses, as will soon be demonstrated. 12 ... Nxc6

088

With the threat of 13 ... Nd4, to which there is no good defence. 13 Kf1 The jump of Black’s knight is decisive in many lines; for instance, 13 h3 Bxe2 14 Nxe2 Nd4, or 13 Bc3 Nd4 14 Bxd4 Qxd4 15 h3 Bxe2 16 Nxe2 Rxe2+! 17 Kxe2 Re8+ 18 Kf1 Nh5 and White’s position collapses. Exercise (easy): What is the clearest finish after 13 Kf2 - ? Answer: The most convincing line is 13 ... Qb6+ 14 Kg3 Rxe2 15 Nxe2 Nd4 or 14 ... Bxe2 15 Nxe2 Re3+! and wins. Note that 13 ... Rxe2+? 14 Nxe2 Nd4 would be a mistake, because White can defend with 15 Re1.

Exercise: How did Morphy conclude the game? Answer: “Disaster strikes on e2” – Kasparov.

13 ... Rxe2! This move maintains the pin and increases the power of the knight’s incursion at d4. 14 Nxe2 Nd4 15 Qb1 Bxe2+ Now Black has both a material advantage and the attack. 16 Kf2 Ng4+ 17 Kg1 After 17 Ke1, the quickest win is with 17 ... Qh4+! (forcing White to weaken f3) 18 g3 Qe7, while if 17 Kg3 then 17 ... Nf5+ 18 Kh3 Nf2 mate.

089

The position is winning for Black. It does not require great imagination to perceive the irremediable weakness of the white king. Exercise: How did Morphy begin his mating attack?

Answer: 17 ... Nf3+! And with an analysis engines at our elbow we can safely announce that it is mate in a further six moves. 18 gxf3 Qd4+ 19 Kg2 Qf2+ 20 Kh3 Qxf3+ 21 Kh4 0-1 As Morphy announced mate in three: 21 ... Ne3 22 Rg1 Nf5+ 23 Kg5 Qh5 mate. Game 12 P.Morphy-J.Schulten Blindfold game, New York 1857 Ruy Lopez [C64] 1 e4 e5 2 Nf3 Nc6 3 Bb5 Bc5 4 c3

4 ... Nge7 A fully playable developing move. Other options are 4 ... Nf6, 4 ... f5, and the rarer 4 ... Qf6. 5 0-0 0-0?! Question: Now you’ve got me confused. How can such a natural move be dubious? Explain please. Answer: You’re right, it is not clear that this move is objectively bad. But it does require Black to follow it up extremely accurately. 090

Retreating the c5-bishop with 5 ... Bb6 is more precise, in order to answer 6 d4 with 6 ... exd4 7 cxd4 d5!.

6 d4 exd4 7 cxd4 Bb6

Exercise (easy): How can White try to demonstrate that 5 ... 0-0 was inaccurate? Answer: By gumming up Black’s development and preventing the freeing move 8 ... d5.

8 d5 Nb8 Exercise: A sceptic might argue that 8 d5 has made the white centre immobile and increased the scope of the b6-bishop. How should White continue, in order to quell these doubts?

Answer: 9 d6! Preventing 9 ... d6 as well; now the c8-bishop will be very difficult to activate. 9 ... cxd6

Exercise: What did Morphy play here?

Answer: 10 Bf4 White develops a piece and attacks a pawn that cannot be defended, instead of capturing it with the queen at once. Morphy himself asserted that his move was “an improvement on that given in the leading treatises of the day”; i.e. on 10 Qxd6, as played in W.Hanstein-T.Von der Lasa, Berlin 1842, quoted in Staunton’s Handbook. In that case Black can improve his position with 10 ... Bc7 11 Qd3 d5. 10 ... Bc7? Black underestimates the risks of falling behind in development. He plans, over-optimistically, to 091

exchange the bishops after 11 Bxd6. It was necessary to return the pawn to bring the c8-bishop back to life with 10 ... d5! 11 exd5 d6. 11 Nc3 It should be no surprise to us that Morphy calmly continues with his development, rather than allowing exchanges by capturing a pawn which cannot run away. 11 ... a6 12 Bc4 b5 13 Bb3 Bb7

Exercise: Black has found a way to bring his queen’s bishop into play. What do you think Morphy played now?

Answer: 14 Bxd6 Having exhausted all his useful developing and positional moves, Morphy turns his attention to more mundane material matters. Now is the moment to regain the pawn. 14 ... Bxd6 This move looks dubious – why help the white queen to become active? It was better to develop with 14 ... Nbc6. 15 Qxd6 h6?! The contrast between Morphy and most of his opponents in understanding of the time element in chess is strikingly clear. Here 15 ... Nbc6? would have been answered by 16 Rad1, so it was best to preface it with 15 ... Nc8. 16 Rad1 Nc8 17 Qf4 Nb6?! It was more important for Black to bring his pieces into play than control central squares. He should have played 17 ... Nc6, with a not very enviable but still viable position.

Exercise: What is the defect to Black’s last move? 092

Answer: 18 Ne5 Now the f7-square is difficult to defend. 18 ... Qf6 Sad, but as a result of his 15th move, 18 ... Qe7 is not possible in view of 19 Ng6. Exercise: What did Morphy play in this position?

Answer: 19 Qxf6! Valery Beim, in Paul Morphy: A Modern Perspective, wrote that this move was typical of Morphy’s style, which then influenced Capablanca and Fischer. Presented with two options of similar strength, Morphy would choose the clearer of the two, without fear of simplification – or, putting it another way, to extract the maximum benefit from it. A player with a more aggressive and less positional style, such as Anderssen, would have opted to keep the queens on with 19 Qg3 and tried to exploit his advantage on the kingside – which, it has to be said, would have been very strong as well/ 19 ... gxf6 20 Ng4 Kg7

Exercise: Why did Morphy elect to exchange the queens?

Answer: 21 Nxf6! The decision is justified by tactics, naturally. 21 ... Bc6 After 21 ... Kxf6 22 Rd6+ Ke7 23 Rxb6 Bc6 24 Nd5+ Kd8, the white rook is trapped but in no danger. 22 e5 a5 23 Rd3 This prevents 23 ... a4 because, despite the absence of the queens, Black’s lack of development is still a significant factor. The response would be 24 Rg3+ Kh8 25 Bxf7! and wins. 23 ... Rh8

093

Exercise: White is a pawn up with a clearly winning position, but good technique is still required to convert it. How did Morphy make progress?

Answer: 24 Ncd5! Employing a method which would later become a characteristic of Capablanca’s play: “clearing all the dead wood from the position”; i.e. exchanging all the inessential pieces. 24 ... Nc4 This allows the further weakening of his pawn structure; but after, for instance, 24 ... Nxd5 25 Bxd5 Bxd5 26 Rxd5 b4 27 Rc1, the black position is indefensible in any case. 25 Bxc4 bxc4 26 Rg3+ Kf8 27 Nb6 Ra7 Exercise: Among many attractive options, what do you think was Morphy’s choice here?

Answer: 28 Rd1 The last piece comes into play, looking to attack. White doesn’t need to capture another pawn (c4) to win. 28 ... Bb5 29 Rd4 Rc7

Exercise: How did Morphy now compel resignation?

Answer: 30 Rdg4 1-0 094

“Winning by force,” wrote Morphy, and added: “This game has certainly no claims to brilliancy, but illustrates the difficulty of a correct defence to the Ruy Lopez game.” (A.C.M. 1858) In contrast, Maróczy considered this to be one of Morphy’s best games, in view of the “iron precision of the moves.” Game 13 P.Morphy-C.Stanley Blindfold game, New York 1857 Evans Gambit [C51] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Ba5 6 d4 exd4 8 0-0 d6 9 cxd4 Bb6 9 Nc3

The “Morphy Attack”, which was subsequently taken up by Chigorin, develops the knight without committing the c1-bishop or advancing in the centre, although Morphy also played 9 Bb2 and 9 d5, as we’ve already mentioned. 9 ... Nf6? This move, which looks so natural, is in fact a serious error, one which was made several times by Morphy’s opponents. In Game 14 (and its supplementary games) we shall see the correct replies, 9 ... Na5 and 9 ... Bg4. Question: How is it that Morphy’s opponents made the same mistake several times over? Answer: Information was scarce in those days. There were no magazines and there was much less opening study and research than in later years. Also, many of Morphy’s opponents were amateurs, strong ones, but amateurs at the end of the day.

In P.Morphy- J.Arnous de Rivière, Paris 1863, Black played 9 ... Qf6?, which not only looks bad, it is bad. Morphy responded with the natural 10 Nd5, and after 10 ... Qg6 11 Nf4 Qf6, you now have the pleasant task of playing with White.

095

Exercise (easy): What did Morphy play here? Answer: He tried to open the e-file with 12 e5! dxe5 13 dxe5 Qf5 (the pawn cannot be captured again due to Re1) and continued his idea with 14 e6. Black felt forced to play 14 ... f6 (14 ... fxe6 15 Bxe6 Bxe6 16 Nxe6 is even worse, since the opening of the e-file is decisive), but after 15 Nh4 Qc5 16 Be3!, the black queen continued to facilitate White’s development: 16 ... Qg5 (16 ... Qxc4 loses at once to 17 Qh5+ g6 18 Nhxg6) 17 Nf3 Qa5 18 Bxb6 Qxb6 19 Nd5 Qa5 20 Nd2! Nd4 21 Nb3 Nxb3 22 axb3 Qc5 and now finally 23 Qh5+ Kd8 (23 ... g6 24 Nxf6+ wins the queen) 24 Rad1 1-0. Exercise: Let’s return to the game. What exactly is wrong with 9 ... Nf6 - ?

Answer: 10 e5! Opening lines is White’s top priority. He can’t allow Black to bring his king to safety “for nothing”. 10 ... dxe5 It is noteworthy that, five years before the game noted above, Arnous de Rivière had already reached this position against Morphy. He chose 10 ... d5?, which is worse since it allows White both to open lines and destroy Black’s kingside: 11 exf6 dxc4 12 fxg7 Rg8 13 Re1+ Be6 14 d5 and White won quickly, P.Morphy-J.Arnous de Rivière, Paris 1858.

Exercise: How should White proceed?

Answer: 11 Ba3! A thematic move in the Evans Gambit, and one with which we are already familiar. As we’ve already mentioned, the priority, even at the cost of sacrificing several pawns, is to prevent the black king from 096

castling into safety. 11 ... Bxd4?! Trying to keep lines closed. Opening the e-file with 11 ... exd4 is clearly worse; e.g. 12 Re1+ Be6 13 Qb3 Qd7 (or 13 ... Na5 14 Bxe6 Nxb3 15 Bf5+) 14 Rxe6+ fxe6 15 Bxe6 and White gains a decisive material advantage. In P.Morphy-T.Hampton, London 1858, Black played 11 ... Bg4. Exercise (easy): How did Morphy respond? Answer: With our familiar 12 Qb3!, attacking f7. Black tried 12 ... Bh5 (if 12 ... Na5, one winning line is 13 Bxf7+ Kd7 14 Qc2 exd4 15 h3 Bxf3 16 Qf5+ Kc6 17 Qb5 mate) 13 dxe5 Ng4 (now if 13 ... Na5, then 14 Qa4+ Nd7 15 e6 or 14 ... Qd7 15 Bb5 c6 16 Rad1 cxb5 17 Nxb5 Qc6 18 exf6 is one way to win), but resigned after 14 Rad1 Qc8 15 e6 f6 16 Qb5 Bg6 17 Bd5 1-0. The most tenacious defence seems to be 11 ... Nxd4 12 Nxe5 Be6, but despite being two pawns up and with fewer open lines, Black is almost paralysed with his king stranded in the centre after 13 Qa4+ c6 14 Bxe6 fxe6. One curious move which emphasizes Black’s helplessness is 15 Rab1, with the idea of capturing on b6; e.g. 15 ... Qc7?! 16 Rxb6! Qxb6 17 Nc4, followed by 18 Nd6+ and wins.

Exercise: What did Morphy play after 11 ... Bxd4 - ?

Answer: 12 Qb3 Of course – once again White has f7 in his sights, and his position is already much better. Sometimes chess looks easy. 12 ... Be6 If 12 ... Qd7, the strongest reply is a typical Morphy move, 13 Rae1!, gaining dynamic superiority in the centre by involving an inactive piece. If 13 ... Na5 then 14 Nxe5! Nxb3 15 Nxf7+ forces Black to return the queen with 15 ... Qe6, after which White gains both a material advantage and the better position; e.g. 16 Bxe6 Bxe6 17 Nxh8 0-0-0 18 Nf7 Rd7 20 Nd6+! cxd6 21 axb3. Exercise: Instead of 13 Rae1, some sources have awarded 13 Ng5 an exclamation mark. To remind us that chess is rarely easy, can you find a way to call this assessment into question?

097

Answer: There is only one move that defends, but it looks good enough to equalize – the natural counter-attacking move in such positions: 13 ... Na5! 14 Bxf7+ Kd8 and Black remains two pawns up. Then 15 Qb4 gives Black a choice between 15 ... c5 16 Qxa5+ b6 (e.g. 17 Qa4 Bxc3 18 Rad1 Bd4) and 15 ... Nc6 16 Qc4 Na5, with a possible draw by repetition. If, in the second line, the white queen prefers to slip away with 17 Qe2, Black has 17 ... Qf5, defending against the check on e6, after which 18 Nb5 Bxa1 leads to tremendous complications, where the analysis engines consider chances to be fairly equal.

13 Bxe6 fxe6 14 Qxe6+ Ne7 15 Nxd4 exd4 16 Rfe1

After only sixteen moves Morphy has a winning position. Now if 16 ... dxc3 then 17 Rad1 leads to a quick mate.

16 ... Nfg8 Trying to hang on to the extra piece. In a blindfold simultaneous display, another game P.MorphyC.Golmayo Zupide, Havana 1864, continued 16 ... Qd7 17 Qxe7+ Qxe7 18 Rxe7+ Kd8 19 Rd1 and White’s material advantage was decisive. 17 Nd5 Here 17 Bxe7 Nxe7 (or 17 ... Qd7 18 Qc4 Nxe7 19 Nd5) 18 Nd5 was simpler, but the text move is equally strong. 17 ... Qd7

098

White’s position is winning, and it looks as if “everything wins”, but it’s important never to relax. Exercise: So, how does White win?

18 Bxe7? It is surprising that Morphy should choose one of the few obvious moves that doesn’t win almost straight away. Answer: There was a clear win with 18 Qxd7+ Kxd7 19 Nxe7, when White has a piece for two pawns that will soon be reduced to one, e.g. 19 ... b6 20 Nxg8 Rhxg8 (or 20 ... Raxg8 21 Re7+ Kc8 22 Bb2 c5 23 Rxa7) 21 Re7+ Kc8 22 Rae1, followed by the invasion of the seventh rank by both white rooks, or if 22 ... Kb7 then 23 Rc1 Rac8 24 Bd6.

Alternatively, 18 Nxe7, 18 Qe5 and 18 Qe4 were all good too; in each case White wins a piece for very little compensation, whereas Morphy’s move gives away some of his advantage. 18 ... Qxe6 19 Rxe6 Kd7! Thanks to the threat of ... c7-c6, Black regains the piece. 20 Rae1 It is worth mentioning that had White played 18 Nxe7 (instead of 18 Bxe7?), he could have answered 18 ... Qxe6 19 Rxe6 Kd7 with 20 Re4!, keeping the extra piece. 20 ... Re8!

And ... c7-c6 is coming next. Exercise: True, White has lost part of his advantage, but still has enough to win. What is his best move now?

21 R6e4 Morphy contents himself with a favourable endgame – he has active rooks and a strong bishop, which are not insignificant trumps. Answer: Yet 21 R6e5! was even stronger, threatening to retreat the bishop, while after 21 ... c6, White can play 22 Bd6!, remaining with 099

a piece for two pawns, since 22 ... Rxe5? 23 Bxe5 cxd5 fails to 24 Bxg7.

21 ... c6 22 Rxd4 cxd5 23 Rxd5+ Kc6 24 Rd6+ Kc7 25 Rc1+ Kb8

Black’s king has managed to find shelter on the queenside and he has restored material equality; nevertheless, White’s advantage remains unquestionable – all three of his pieces are superior to their opposite numbers. 26 Bh4 Nh6 27 Bg3 Activating the bishop with tempo due to the threat of 28 Rd8 mate.

27 ... Ka8 28 h3 Nf5 29 Rd7 g6 Damaging the enemy pawn structure with 29 ... Nxg3 30 fxg3 would bring no relief, since White threatens 31 Rcc7 as well as attacking the g7-pawn. Exchanging a rook with 29 ... Re7 30 Rxe7 Nxe7 doesn’t help either, as after 31 Rc7 Nf5 32 Be5, Black will soon lose a pawn. 30 Rcc7 Nxg3 31 fxg3

Again, the structural damage is not serious. Here White will soon have a majority of three to one on the kingside, which will be easy to mobilize.

31 ... Rb8 The situation would be equally difficult after 31 ... Re2 32 Rxh7 Rb8 33 a4; e.g. 33 ... Rb4 34 Rh4 Rxh4 35 gxh4 b5 36 axb5 Rxb5 37 Rc6 with a winning endgame. 32 Rxh7 Rxh7 33 Rxh7 a5 34 h4 Rg8?! More tenacious was 34 ... b5, seeking to speed up the advance of his majority, but White would have the lead in the pawn race after 35 Rh6 Rg8 36 g4 Kb7 37 h5. 35 g4 b5 36 h5 a4 After 36 ... gxh5 37 gxh5 b4 38 h6 Rg6 39 Rh8+ Kb7 40 h7 Rh6 41 g4!, the white pawns are very quick. 100

Exercise (easy): What is the quickest way to win now?

Answer: 37 h6! Intending 38 Rg7, followed by h6-h7 and Rg8 etc; White is going to get there first. 37 ... b4 38 Rg7 Rh8 39 h7 b3 40 Rg8+ Kb7 41 Rxh8 b2 42 Rb8+ Kxb8 43 h8Q+ 1-0 Game 14 P.Morphy-J.Schulten Blindfold game, New York 1857 Evans Gambit [C51] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Bc5 6 0-0 d6 7 d4 exd4 8 cxd4 Bb6 9 Nc3 Bg4

The main alternative is 9 ... Na5, which we will look at in Supplementary Game 14.1. 10 Bb5

10 ... Bxf3 Morphy himself employed 10 ... Bd7 here, as we shall see in Supplementary Game 14.2. The most common move is 10 ... Kf8, which we shall cover in Supplementary Game 14.3. Question: But the text move, doubling White’s f-pawns to stabilize the centre, looks the most logical, doesn’t it? And Black is still a pawn up. Answer: It isn’t clear. Although doubling the opponent’s pawns is good most of the time, in this case the pawns are not noticeably weak and we shouldn’t forget that White has gained the half-open g-file. However, the main counter-argument is that Black’s pieces won’t be coordinated for several moves, so he will be in no state to apply pressure to the weaknesses he’s created.

11 gxf3 Black must now attend to the threat of 12 d5.

101

11 ... Kf8

This is a concession and Morphy would certainly have been satisfied with his position. Question: Yes, Black has lost the right to castle, but his king is not so badly placed, is it? Does this fully compensate White for being a pawn down? Answer: Although the black king is in no immediate danger, he has problems with his lagging development and the lack of coordination among his pieces. It is not easy for the black rooks to come into play and will take time. All this is more than enough compensation for the pawn.

12 Be3 Nce7

Exercise: Black wants to stabilize the centre by advancing his pawn to d5. What did Morphy do next?

Answer: 13 Kh1! The most flexible move. Morphy doesn’t yet know how best to deploy the rest of his forces, so he decides to speed up the possibility of playing Rg1. 13 ... c6 14 Ba4 d5 Here 14 ... Nf6 has been suggested, but it doesn’t seem that isolated moves will change the evaluation significantly. Black still faces the problem of how to activate his pieces. In that event White naturally would not break up his formation with 15 e5?, since he does not gain anything after 15 ... dxe5 16 dxe5 Nfd5; instead, he could continue to improve his pieces, for instance with the obvious 15 Rg1 or maybe 15 Bb3. 15 Rb1 Question: A change of plan? Answer: Not really; Black’s bishop on b6 is “biting on granite”, as they say, and he plans to redeploy it to c7, so White prevents it for the moment, forcing Black to make a preparatory move.

15 ... Rb8 16 Qd3 Controlling the b1-h7 diagonal is advantageous to White, as we’ll see. 16 ... Bc7

102

Exercise (easy): How to continue now? Answer: With the simplest move.

17 Rg1 There is flexible “building” left to do, so now is the moment to play a move which is an essential component of any future white plan: occupying the g-file. 17 ... Ng6 Exercise: And now?

Answer: 18 e5! We know that Morphy didn’t take such decisions lightly, but right now this advance has many things to recommend it – one is that it blocks the action of the c7-bishop, another is that it prepares to attack with f3-f4-f5. 18 ... Qh4?! A provocative move, since the queen can be attacked in several ways. Objectively it is not the best move, but it is an interesting choice from a practical viewpoint and could have been successful, if Black had subsequently seized his opportunity. 19 Bg5 There are many tempting moves here, such as 19 Rg4, 19 Rg3, 19 Bc2, etc. It is difficult to know which is best, but ... you have to choose one. Meanwhile, it’s worth noting that this bishop move exposes it to a future ... f7-f6, so it will be necessary to figure out whether that pawn break favours White (by opening lines for his pieces) or Black (by facilitating his development). 19 ... Qh3

103

Exercise: What is the best way to keep harassing the intrepid black queen?

20 Ne2 Another tempting move; in addition to clearing the third line for a possible Qa3+, the currently inactive knight is ready either to exchange a defending piece with Nf4 or else help to attack the black queen. Answer: Nevertheless, it’s not clear whether Morphy’s knight move is better than the natural 20 Rg3 Qd7, followed by choosing the best moment to retreat the bishop from g5, with the plan of Rbg1, Bc2, etc. White would have the initiative in that case. Exercise: How can White’s last move be called into question?

Answer: 20 ... f6! This is the only good response – after other moves White could continue in the way outlined above. 21 Nf4?! This was one of the ideas behind 20 Ne2, but it’s not the best. 21 Bf4 was preferable

Exercise: How should play Black now?

Answer: 21 ... Nxf4? Now the game reverts to its natural course. Black could have achieved a good position with 21 ... Qd7! 22 Nxg6+ hxg6 23 Bf4 fxe5 24 dxe5 Qf5! or 24 Bxe5 Bxe5 25 dxe5 Ne7. In both cases the exchanges have greatly relieved Black’s position and he is now able to complete his development, while remaining a pawn up. 104

22 Bxf4 With the threat of 23 exf6 Bxf4 24 fxg7+ Kf7 25 gxh8N+ (or just 25 Rg2).

22 ... g5?! This advance is not a good sign. 22 ... Qd7 was more tenacious, although White’s advantage would not be in doubt. The black king is not as well protected as a few moves ago and White could continue with, for instance, 23 Rbe1, when all his pieces are active. Opening lines with 22 ... fxe5 is suspicious: White can play 23 Qa3+ Kf7 (not 23 ... Ne7? 24 Rxg7!) 24 dxe5 (threatening 25 e6+) 24 ... Qe6 25 Bc2 (now planning 26 Bf5 Qxf5 27 e6+, opening the position even more with a strong attack, while also attacking a7) 25 ... Ne7 26 Bg5! (threatening 27 Bxe7 Qxe7 28 Rxg7+ etc) 25 ... Rhe8 (or 25 ... Bxe5 26 Rbe1) 26 f4 with a strong initiative.

Exercise: How did Morphy respond?

Answer: 23 Qa3+! A strong move that forces Black to worsen the position of the king. He has no choice, since after 23 ... Kf7 24 e6+, or 23 ... Kg7 24 exf6+, or 23 ... Ne7 24 exf6, White wins in all cases.

23 ... Ke8 Exercise: There are several good moves now.

Morphy chose the most crushing – what is it? Answer: 24 Rxb7! The start of a king hunt. White gains a decisive attack and a material advantage into the bargain. The more prosaic 24 Rde1 would also have won easily. 24 ... Rxb7 25 Bxc6+ Kf7 26 Bxd5+ Kg6 27 Qf8 With various threats to mate in a few moves, such as 28 Rxg5+! or 28 Qxf6+! etc. Taking the rook on b7 was obviously good too.

105

27 ... Qd7 28 Bxb7 Bd8 29 exf6 Bxf6 30 Be4+ Kh5 31 Be3 h6 32 Rg3 With all four pieces attacking, the mate is not far away. 32 ... Bg7 33 Qf7+! 1-0 Since 33 ... Qxf7 34 Rh3 is mate. Supplementary Game 14.1 P.Morphy-NN Blindfold simultaneous, New Orleans 1858 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Bc5 6 0-0 d6 7 d4 exd4 8 cxd4 Bb6 9 Nc3 Na5

10 Ng5?! This is the only time on record that Morphy faced 9 ... Na5; it was a blindfold game against an amateur and he decided go on the attack, allowing the exchange of his powerful light-squared bishop in order to gain the initiative. Chigorin and Fischer both played 10 Bg5 here, which is still considered the strongest move; this also permits the c4-bishop to be exchanged following 10 ... f6, but at the cost of weakening Black on the light squares. 10 ... Nxc4 11 Qa4+ c6 12 Qxc4 Nh6 13 Kh1 It’s not a good sign that White feels he has to spend a tempo on this preparatory move, but after the aggressive 13 f4 0-0 14 f5 or the quieter 13 Be3 0-0 14 h3, White’s attack is stopped short, for the same reason that could have occurred in the game. 13 ... 0-0 14 f4 Kh8 15 f5

106

Exercise: Morphy is not hiding his intention to take the black position by storm – the advance f5-f6 is in the air. How should Black respond?

15 ... f6? This prevents 16 f6, but is not the most effective response. Answer: Black could counter White’s expansion with 15 ... d5!, since after 16 exd5 Bxf5, all the black pieces are developed and can easily contain White’s attacking ambitions; e.g. 17 dxc6 bxc6 18 Nxf7+ Nxf7 19 Rxf5 Qxd4. The same central counter-strike would also be effective at the end of the lines mentioned at move 11.

16 Ne6 Bxe6 17 fxe6 Qe7? Another weak move, after which the situation seriously deteriorates. Exercise (easy): What did Morphy play here?

Answer: 18 Bxh6 Of course; showing that 17 ... Ng8 would have been correct. 18 ... gxh6

Exercise: White just needs to bring his queen’s rook into the game. What did Morphy play next?

Answer: 19 Rf3 To deploy the a1-rook to its most effective square, f1. 19 ... Rg8 20 Raf1 Rg6 107

Exercise: White’s position now “plays itself”. How did Morphy proceed?

Answer: 21 Ne2 The knight joins in the offensive. 21 ... Rf8 22 Nf4 Rg5 23 d5 c5 24 Qc3 Bd8 25 Ne2! Heading for f5. 25 ... Qg7 26 Ng3 Qc7

27 Rxf6! The most elegant way; with this combination White is able to exploit his passed pawn on e6. 27 ... Bxf6 28 Rxf6 Rxf6 29 Qxf6+ Qg7 Or 29 ... Kg8 30 e7 Qc8 31 Qe6+ Qxe6 32 dxe6 and the front e-pawn queens. 30 Qd8+ Qg8 31 e7 Re5 32 Nh5 Rxe4 33 e8Q 1-0 It is mate in a few moves; e.g. 33 ... Rxe8 34 Qf6+ or 33 ... Re1+ 34 Qxe1 Qxd8 35 Qa1+ etc. Supplementary Game 14.2 J.Arnous de Rivière-P.Morphy Paris 1863 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Bc5 6 0-0 d6 7 d4 exd4 8 cxd4 Bb6 9 Nc3 Bg4 10 Bb5 Bd7

108

Morphy considered this to be the best move. Later, 10 ... Kf8 became more popular. 11 Bg5 The alternative is 11 e5, when opening the game leads to a difficult position after 11 ... dxe5? 12 d5!, so it is better to continue developing with 11 ... Nge7. A simultaneous game P.Morphy-NN, New York 1859, saw 11 Re1 Nge7 12 e5 and here, instead of 12 ... 0-0!, Black committed the error of opening lines: 12 ... dxe5?, which was again answered by 13 d5!, and after 13 ... Nd4? (13 ... Nb8 was almost forced), Black was quickly crushed: 14 Bxd7+ Qxd7 15 Nxe5 Qd6 16 Qa4+ c6 17 Ba3 Bc5 18 Ne4 Qxe5 19 Nxc5 Ne2+ 20 Kh1 Qd4 21 Qxd4 Nxd4 22 Nxb7 0-0 23 Bxe7 1-0.

11 ... Nce7 In theory, offering piece exchanges when a pawn up is not a bad idea for Black. Morphy was still reluctant to alter his structure with 11 ... f6; nor was he keen to place his knight in a pin with 11 ... Nge7, allowing White to increase the pressure with 12 Nd5. 12 Bc4 Be6 13 Qa4+ Here it would have been interesting to weaken the opposing structure with 13 Bxe6 fxe6. Admittedly, Black welcomes the exchange of pieces, but this comes at a price: after, for example, 14 a4 a5 15 Qb3 Qd7 16 Rac1, Black still needs to complete his development. 13 ... Qd7 14 Bb5 c6 15 Bd3 f6

16 Bh4 Retreating the other way with 16 Be3 seems better, where the bishop is more active and controls h6. Black should reply with 16 ... d5. 16 ... Ng6 Here 16 ... Nh6 was to be considered, but not now 16 ... d5?!, as 17 Rfe1 reveals why White chose 16 Bh4. 17 Bg3 N8e7 18 d5?! It was better to play 18 Nb5 first, after which 18 ... Nc8 is forced, and only then 19 d5. 18 ... Bf7 19 Nb5 Bc5 109

This possibility could have been avoided by playing 18 Nb5. Black is not forced into making such a passive move as ... Nc8. 20 Rac1 0-0 White’s fireworks have burnt out and he has to beat a retreat. 21 Nbd4 Kh8

A curious move – the point being that, with his king having left the a2-g8 diagonal, it is now possible for Black to take on e6 twice in the event of White playing Ne6. Instead, 21 ... Rfd8 was more natural. 22 Bc4 Strengthening d5 and preparing a future Ne6. 22 ... Ne5 23 Nxe5 fxe5 24 dxc6 Here 24 Ne6 Bxe6 25 dxe6 would have been more consistent with his previous moves. Nevertheless, even though White has an annoying passed pawn at e6, it is hard to see how he can assist its advance, whereas the alteration to the pawn structure has strengthened Black’s majority in the central/queenside zone. 24 ... bxc6 25 Nb3 Bb6 26 Rfd1 Applying pressure on Black’s centre. Max Lange suggested 26 Bxf7 Rxf7 27 Rcd1 Qe6 28 Nd2, but is not clear that this is a significant improvement. 26 ... Bh5 27 Rd2 a5 A move that looks like an inaccuracy, but in fact it has hidden depths. 28 Qa3 Rad8 29 Bxe5

Exercise: What had Morphy prepared?

Answer: 29 ... Rxf2!

110

The same shot would have followed 28 Bxe5. 30 Kh1? If 30 Bd4? Bxd4 31 Rxd4 then 31 ... Rxg2+! 32 Kxg2 Qg4+ and mates in a few moves. The result is similar after 31 Rxf2 Bxf2+ 32 Kxf2 Qa7+! (remember 27 ... a5!) and the queen invades with decisive effect. The only defence was 30 Rxf2! dxe5 31 Nc5 Qd4 32 Nd3, when 32 ... Ng6 33 Kh1 Qxe4 34 Rb2 leaves Black with two pawns for the exchange and the initiative, but White is still in the game. 30 ... Rxd2 Winning a piece; the rest is very simple for Morphy. 31 Bxg7+ Kxg7 32 Nxd2 Qa7 33 Qg3+ Bg6 34 Rf1 Bd4 35 Nf3 Qc5 36 Bb3 Rf8 37 h4 Be5 38 Qg4 Qe3 0-1 Supplementary Game 14.3 P.Morphy-J.Arnous de Rivière Paris 1863 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Bc5 6 0-0 d6 7 d4 exd4 8 cxd4 Bb6 9 Nc3 Bg4 10 Bb5 Kf8

Maróczy recommended interpolating 10 ... a6 11 Ba4 before 11 ... Kf8. 11 Be3

11 ... Nce7 An alternative was 11 ... Nge7, a move of which Tarrasch disapproved, commenting: “This position has occurred before, and my opinion is that White has a far superior game – complete freedom and opportunity for attack on all sides. It is difficult for Black to develop or even to find a counter-attack. True, Black has an extra pawn, but it is of very little use or importance; with equal players six times out of ten White would win.” 12 h3 Bh5 13 Bd3 f6 To provide a retreat for the bishop on h5, although of course it is a weakening move, and the black position begins to feel rather suspect. 14 a4 c6

111

One of the points of 11 ... Nce7 – Black prepares to retreat the bishop from b6 and controls the d5square, with the possibility of playing ... d6-d5 at an opportune moment, as we’ve seen before. 15 Qb3 Question: Hmm, knowing that one of Black’s ideas is to play ... Bf7, why did Morphy present him with a tempo? Answer: The tempo is not an especially useful one and, in any case, the white queen is heading for a3, to put pressure on the d6-pawn.

15 ... Bf7 16 Qa3 Ba5 17 Ne2 Nc8 Making way for the g8-knight; Black’s lack of space and mobility is becoming more and more evident. 18 Rab1 In contrast, White expands the scope of his pieces with almost every move. 18 ... b6 19 Ba6 Nge7

Exercise: There are several attractive moves here. What do you think was Morphy’s choice? Answer: He has decided that this is the moment to take concrete measures and try to open the centre and/or the kingside.

20 e5! Although this move concedes control of d5, “to get squares, you have to give squares” (Najdorf and Fischer). 20 ... Bd5 21 Nf4! fxe5 22 dxe5 dxe5 23 Nxd5 cxd5 He can’t play 23 ... Qxd5 due to 24 Rfd1. 24 Nxe5

112

The position has opened up, all White’s pieces are active, whereas Black’s are all uncoordinated. White’s missing pawn is irrelevant – his position is winning. The rest was very simple. 24 ... Qd6 25 Qb2 Nf5 26 Bf4 Qe7 27 Nc6 Qe4 28 Bb7 Qxf4 29 Bxa8 Qxa4 30 Rbd1 Qe4 31 Qa3+ Kf7 32 Nxa5 bxa5 33 Rfe1 Qb4 34 Bxd5+ Kg6 35 Re6+ Kg5 36 Qc1+ Qf4 37 h4+ Nxh4 38 Re5+ Nf5 39 Rxf5+ Kxf5 40 Be6+ 1-0 Game 15 P.Morphy-J.Bonford Blindfold simultaneous, New Orleans 1858 Evans Gambit [C52] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 Fischer commented that it was safer to play 4 ... Bb6, but added that the gambit could hardly be refuted like that. There are only two recorded casual games of Morphy’s featuring this retreat, both wins. In one of them Morphy played completely in gambit style: P.MorphyF.Lewis, London 1858, continued 5 0-0 Qe7 6 a4 Nxb4 7 a5 Bc5 8 c3 Nc6 9 d4 exd4 10 cxd4

White has reached a very favourable version of the Evans Gambit Accepted. The differences are all in his favour: the pawn on a5 hinders the normal retreat ... Bb6, while 10 ... Bb4 comes without check, since White has already castled; all Black has gained is the move ... Qe7. Now White has several options, of which the most attractive is 11 Bg5, provoking the weakening 11 ... f6. Instead, Morphy decided to sharpen the game even more with 11 Ne5!? Nxe5 12 dxe5 Qxe5 13 Qb3 and won in the complications. 5 c3 Ba5 6 d4 exd4 We shall look at 6 ... d6 in Supplementary Game 15.1. 7 0-0

113

7 ... dxc3 A risky line, but not clearly bad. Instead, 7 ... Bb6 8 cxd4 d6 leads to the so-called ‘normal’ position of the Evans (usually arising via 5 ... Bc5), which is solid enough in Fischer’s opinion. In the first game of the Morphy-Anderssen match, Paris 1858, Black played the unusual 7 ... Nf6 (7 ... Nge7 is more usual) and curiously Morphy replied with 8 e5? (8 Ba3! is critical, as Morphy himself played several times) 8 ... d5 9 Bb5 Ne4 10 cxd4 0-0!, when Black gained the advantage. The game continued 11 Bxc6 bxc6 12 Qa4 Bb6 13 Qxc6 Bg4 14 Bb2? Bxf3 15 gxf3 Ng5 and Black’s superiority had increased; Anderssen won in 72 moves.

8 Ba3 The main line is 8 Qb3, for which we can recommend looking at the game R.J.Fischer-R.Fine, New York 1963 – Game 44 in Fischer’s My 60 Memorable Games. 8 ... d6 9 Qb3 Nh6 10 Nxc3

10 ... Bxc3?! Question: Is this really dubious? I can understand that Black didn’t want to be left with his bishop inactive on the queenside, nor with the white knight threatening to settle on d5 at an opportune moment. Was that not reason enough to make this move playable? After all it simplifies as well, and Black is two pawns up! Answer: Naturally, if you’re allowed to win two pawns you expect to have to make some concession or other. White now has the advantage of the bishop pair in an open position. “It’s not the pieces that are exchanged but those that remain on the board that matter”, wrote Tarrasch. Yet the black bishop was not exactly inactive – control of the e1-a5 diagonal was very useful, to hinder the white pieces from manoeuvring easily (for instance, Qc3 is prevented, while if White plays Re1 then the knight on c3 would be pinned, hindering Nd5). So this exchange cedes control of important dark squares and makes White’s play easier.

11 Qxc3 0-0

114

Exercise: What did Morphy play in this position?

Answer: 12 Rad1 It should come as no surprise to us that Morphy brought a fresh piece into play, with the threat of 13 e5.

The analysis engines have few inhibitions about moving pieces that are already developed and they suggest the interesting move 12 Bb2, forcing 12 ... Qf6, when White can regain one pawn and maintain good compensation for the remaining pawn minus. Morphy, on the other hand, was seeking not only to play well but also to reach positions in which he could outplay his opponent in a complex struggle, so he preferred to keep the queens on the board for now. 12 ... Ng4 Black doesn’t waste the opportunity to activate his “knight on the rim”. There were other options available, such as 12 ... Bg4. In that case Morphy would perhaps have played 13 Bb2 Qf6 14 Qxf6 gxf6 15 Bxf6 after all, now that g4 is occupied by the bishop and 15 ... Ng4 therefore impossible. Black can’t exploit the pin with 15 ... Ne5? either, due to 16 Nxe5 Bxd1 17 Rxd1 dxe5 18 Rd3 and wins. 13 h3 Morphy can’t see how to improve his position and so acquiesces to an exchange of pieces, in order to start an attacking wave with the support of his f-pawn. 13 ... Nge5 Bonford continues with his plan. Alternatively, the zwischenzug 13 ... Qf6!? was interesting, to displace the white queen, banishing it from the long diagonal for the moment. 14 Nxe5 Nxe5

115

Exercise: What do you think played Morphy here?

Answer: 15 Be2! A difficult decision; with the text move he plans to follow up with f2-f4 and/or Bb2. Naturally, it was annoying to leave the a2-g8 diagonal, but after 15 Bb3 Black had a good continuation in 15 ... Be6 16 f4 Bxb3 17 axb3 Nc6, and if 18 Bb2 then 18 ... f6 closes the long diagonal. In this line 17 Qxb3 is better, with the idea of capturing on b7, regaining one of the pawns, at the cost of abandoning the attack. Another possibility was 15 Bb2 at once; here after 15 ... Qf6 16 Bd5 Be6 17 f4 Nd7, Black has all his pieces in play and doesn’t stand badly. It isn’t certain that retreating the bishop to e2 is much better than the two alternatives above, but it keeps more tension by preventing immediate exchanges. 15 ... f5? Seemingly Morphy’s opponent regarded the retreat 15 Be2 as very passive, and dropped his guard, becoming very optimistic, thinking that he could initiate complications without much risk. White play is now easy for the next few moves. It was much more cautious, and better, to retain the possibility of closing the long diagonal with ... f7f6. For instance, after 15 ... f6 16 f4 Nf7, the position is complex and offers chances to both sides. Instead, 16 ... Ng6 has been recommended but this is not convincing, since White can reply 17 e5! fxe5 18 fxe5 Rxf1+ (not 18 ... Nxe5? 19 Qxe5! dxe5 20 Bc4+ and wins) 19 Bxf1, when his greater activity is serious. There is also the typical ‘computer move’ 15 ... Qd7 (similar to 13 ... Qf6 above), planning to answer 16 f4 with 16 ... Qc6!, dislodging the white queen. 16 f4 Nc6

Exercise (easy): How did Morphy proceed?

Answer: 17 Bc4+ Of course; White forces the black king into the corner and prevents the developing move ... Be6 at the same time. 17 ... Kh8 18 Bb2 Here it was a question of taste, as 18 e5 was also crushing. Morphy wants to win in the shortest way. 18 ... Qe7 Exercise (easy): How did Morphy continue now?

Answer: 19 Rde1 Bringing a fresh piece into the attack.

116

19 ... Rf6?! Not the best defence, but there was no satisfactory way to parry the attack in any case; for example, 19 ... Bd7 20 Rf3 or 20 Qg3 (threatening 21 exf5) is difficult to answer, while after 19 ... Qf6 20 Rf3! Qxc3 21 Bxc3, the only way to answer the threat of Rg3 is by surrendering the exchange with ... Rf6. 20 exf5 Qf8

Exercise: There are several very good moves here. Morphy chose the most artistic (and strongest) one. What is it?

Answer: 21 Re8!! A beautiful way to eliminate the f6-rook, a vital defensive piece. 21 ... Qxe8 22 Qxf6! Qe7 Attempting to block the long diagonal with 22 ... Ne5 can be refuted by either the simple 23 Qg5 h6 24 Qg3 or, more artistically, 23 fxe5! gxf6 24 exf6 etc. 23 Qxg7+! Poetry right to the end. Morphy spurns the prosaic win with 23 Qxe7 Nxe7 24 Re1. 23 ... Qxg7 24 f6

24 ... Qxg2+ After 24 ... Qf8, the black king is forced to enter White’s camp: 25 f7+ Ne5 26 fxe5 h5 (or 26 ... Kg7 27 e6+ Kh6 28 e7 etc) 27 e6+ Kh7 28 Bd3+ Kh6 29 Rf6+ Kg5 30 Rg6+ Kf4, where mate awaits after 31 Kf2!. 25 Kxg2 Bxh3+ 26 Kxh3 h5 27 Rg1 1-0 Supplementary Game 15.1 P.Morphy-T.Ayers Mobile 1855 117

1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 b4 Bxb4 5 c3 Ba5 6 d4 d6 7 Qb3

Fischer considered Morphy’s choice to be superior to the more common continuation 7 0-0, when 7 ... Bb6 constitutes the Lasker Defence, “which put the Evans Gambit out of commission”, according to Fischer. 7 ... Qe7?!

Exercise (easy): The main continuation is 7 ... Qd7, as we already noted in Game 2. How can Black’s inaccuracy be exploited?

Answer: 8 d5 Showing that 7 ... Qd7 would have been better, leaving e7 free for the knight and controlling the a4e8 diagonal. 8 ... Nd4 Black’s reply was forced; any other retreat and he would lose a piece to 9 Qa4+. 9 Bb5+?! Morphy decides to enter complications. Here 9 Qa4+ fails to 9 ... Qd7 10 Qxa5 b6!, leading to a position in which Black can have no complaints; e.g. 11 Nxd4 bxa5 12 Bb5 exd4 13 Bxd7+ Kxd7 14 cxd4, and now he can choose between 14 ... Ba6 and 14 ... f5!. Taking the knight is critical: 9 Nxd4! exd4, but now in A.Kislova-G.Gresser, Subotica 1967, White again went for the piece prematurely with 10 Qa4+ Kd8 11 Qxa5? Qxe4+ 12 Kd2, when 12 ... Bf5! would have been very strong; e.g. 13 Na3 Qxg2 14 Re1 Nf6! and White’s position falls apart. Instead, simply 10 0-0 should offer him a slight advantage.

9 ... c6 10 Nxd4 exd4 11 dxc6 Qxe4+ 12 Kd1

Exercise: White threatens 13 Re1, while Black must also deal with the main threat of 13 cxb7+. How to defend? 118

12 ... Bg4+? Not a losing move, but not the best either. Answer: After 12 ... Kf8! 13 f3 Qg6, everything is defended and Black stands clearly better.

13 f3 Bxf3+ 14 gxf3 Qxf3+ 15 Kc2 Qe4+ 16 Kb2

Exercise: The last few moves were almost forced; now we are at a critical moment. What should Black play?

16 ... Bxc3+? This only helps White bring his pieces more quickly into play. Answer: It was essential to play 16 ... bxc6!, bringing about a very unusual position that probably offers equal chances; for example, after 17 Rd1 Ne7 18 Rxd4 Qg2+ 19 Rd2. The alternative is 17 Re1 Qxe1 18 Bxc6+ Ke7 19 Bxa8 Nf6 20 Qb7+ Ke6 21 Bd2 and “all three results are possible”, as they say.

17 Nxc3 dxc3+ 18 Qxc3 0-0-0 19 Re1 Qd5 20 cxb7+ Kxb7

21 Rb1! White has several strong moves here. As we have learned to expect of Morphy, he chooses the most elegant way to bring his inactive rook into the game. Soon all his pieces will be working together in attack. 21 ... Nf6 22 Bc6+ Both 22 Ka1 and 22 Re7+ win more quickly (both mate in nine according to the engines), but by now that doesn’t matter. 22 ... Qxc6 23 Ka1+ Kc7 24 Qa5+ Kc8 25 Qxa7 Nd7 26 Bd2 1-0

119

Chapter Four The Visit to London and the Match Against Löwenthal Morphy wanted to play a match against Howard Staunton, and his American supporters were equally keen to bring this about. The New Orleans Chess Club, together with the American Chess Federation and the New Yorkers, invited Staunton to travel to New Orleans to play Morphy, but Staunton rejected the offer, claiming that his many occupations prevented him from doing so, amongst these a work on Shakespeare that he was writing. In the press, Staunton was more incisive; in the Illustrated London News he wrote that Morphy should first measure himself against other masters. And so he did. On the 20th of June 1858, two days before his twenty-first birthday, Morphy arrived in Birmingham, and once more challenged Staunton, who was now forty-eight. Staunton once again put him off. The explanation was clear to Fischer, writing just over a century later: “Staunton appears to have been afraid to meet Morphy and I think his fears were well-founded. Morphy would have beaten him, but it wouldn’t have been the one-sided encounter that many writers now think it would. It would have been a great struggle.” Morphy and Thomas Barnes did play two consultation games against Staunton and John Owen, both of which ended in victory for Morphy and Barnes, although of course these can’t really be counted as games between Morphy and Staunton. During his stay in London Morphy played several games against the masters Samuel Boden, Thomas Barnes and Henry Bird, of which we saw a few fragments in Chapter One. Later he played against a familiar but far stronger opponent, Johann Löwenthal. Before looking at the games from that match, let’s examine three games played during his stay in London – first, one of his most spectacular victories. Game 16 H.Bird-P.Morphy London 1858 Philidor’s Defence [C41] 1 e4 e5 2 Nf3 d6 3 d4 f5

Question: Is this line any good? It looks very risky to open the game so much when playing Black. 120

Answer: Yes, you’re right. Kasparov considers the move to be not only extremely dubious, but perhaps also losing.

Morphy played this ambitious move of Philidor’s several times. It was quite popular in the nineteenth century but has not stood the test of time. 4 Nc3 Practice has shown that this and other logical moves, such as 4 dxe5 and 4 Bc4, are good for White; in fact the latter has been virtually worked out to a forced win. The only move that does not give White much is 4 exf5. 4 ... fxe4 Black wants to occupy the centre. According to some old analysis, 4 ... exd4 5 Qxd4 fxe4 6 Bg5 Nf6 7 Nxe4 Be7 8 Bc4 Nc6 9 Qe3 is not satisfactory: the weaknesses created by 3 ... f5 are obvious, and 9 ... Ng4 is well met by 10 Qc3!. Instead, 4 ... Nf6 5 dxe5 Nxe4 6 Nxe4 fxe4 7 Ng5 d5 8 e6 Bc5 is no good either.

Exercise: How can White refute this line? Answer: Not with 9 Nf7? on account of 9 ... Qf6, threatening mate on f2. Instead, 9 Nxe4! works, intending 9 ... fxe4 10 Qh5+ g6 11 Qxc5, while after 9 ... Be7 (seeking to recapture on e6 following the retreat of the knight), White can play 10 Qh5+! g6 11 Qe5 Rg8 12 Be3!, intending 12 ... dxe4 13 Rd1 Bd6 14 Bb5+ Nc6 15 Qxe4 and Black is lost; 11 ... Rf8 is sometimes recommended instead, though 12 Be3 is very good there too.

5 Nxe4 d5 Question: What is Black trying to achieve with all these pawn moves? Answer: In general, if Black manages to close the game and gain space, his position would more than playable – but not if the position opens up, as we have already seen. Zukertort commented that after the exchanges 5 ... Nf6 6 Nxf6+ gxf6 (or 6 ... Qxf6 7 Bg5, followed by 8 dxe5) 7 dxe5 fxe5?, the black position is full of weaknesses. White has several good continuations, such as 8 Ng5 or 8 Bc4, although today’s analysis engines discover straight away that White can play 8 Nxe5! dxe5 9 Qh5+ Kd7 and now the ‘silent killer’ 10 Be2!, when the black king cannot be defended against the imminent attack by White’s queen, two bishops and the a1-rook without losing material. So 7 ... dxe5 is necessary, when it is reasonable to exchange queens, though White should gain more advantage with developing moves such 8 Bd3, as Black’s kingside is clearly fragile.

121

6 Ng3 Playable; but 6 Neg5 was stronger, when 6 ... h6 can be met by the surprising 7 Nf7! Kxf7 8 Nxe5+ and the white queen joins in attack with devastating effect; e.g. 8 ... Ke7 9 Ng6+ Kf6 10 Qf3+ Bf5, as in P.Stepanov-E.Maljutin, Moscow 1992, and now simply 11 g4!. Zukertort pointed out another strong continuation: 6 Nxe5! dxe4 7 Qh5+ g6 8 Nxg6 Nf6 (no better is 8 ... fxg6 9 Qxg6+ Kd7 10 Qf5+ Ke8 11 Qe5+, followed by Qxh8, with a rook and three pawns for two minor pieces) 9 Qe5+ Kf7 10 Bc4+ Kg7 (not 10 ... Kxg6?? 11 Qg5 mate), when among several advantageous lines, such as 11 Nh4 and 11 Nf4, Keres gave 11 Bh6+ Kxh6 12 Nxh8 Bb4+ 13 c3 Qxh8 14 cxb4 and White’s advantage is almost decisive, according to Kasparov. 6 ... e4 7 Ne5 Nf6 8 Bg5 For years it was considered that 8 f3 was best. Kasparov gives this move an exclamation mark as “advantageously undermining the centre”, but perhaps that’s not so clear.

Exercise: How do you think that Morphy would respond? Answer: Rather than fall in with White’s intentions by playing 8 ... exf3? 9 Qxf3, Morphy would reach “his” type of position with 8 ... Bd6! 9 fxe4 0-0, after which, at the cost of a pawn, Black has more pieces in play and poses the question “which king is the weaker now?”.

8 ... Bd6! “A move typical of Morphy, striving as soon as possible to overcome his lack of development,” wrote Euwe. Morphy didn’t consider it necessary to unpin with 8 ... Be7 and characteristically chose a more active move. 9 Nh5?! Consistent, but White is moving another already developed piece, while the white king remains in the centre. It was obviously more prudent to play 9 Be2, followed by 10 0-0. 9 ... 0-0 10 Qd2?! 122

Another inaccuracy; White is now falling dangerously behind in development.

Exercise: How did Morphy respond?

Answer: 10 ... Qe8! Forcing White to come to a decision on the kingside. The counter-strike 10 ... c5 was interesting too, but Morphy prefers play with the pieces. 11 g4? “After a series of incomprehensible moves, Bird commits a decisive mistake,” commented Kasparov. This is certainly a bad move, although White’s position wasn’t an easy one to play. If 11 Bxf6, apart from the obvious 11 ... Qxh5, Black can try 11 ... gxf6, reaching a promising position after the semi-forced line 12 Nxf6+ Rxf6 13 Qg5+ Rg6 14 Nxg6 hxg6 15 Qxd5+ Kg7. Here White has a rook and two pawns for the two minor pieces, but the latter can be quickly activated. Maróczy evaluated the position as balanced, whereas Euwe disagreed: “In the resulting open position the strength of the two black bishops should quickly tell.” The alternative move order 11 Nxf6+ gxf6 12 Bxf6 has also been suggested, transposing above after 12 ... Rxf6. However, Black has a zwischenzug that enables him to seize the advantage.

Exercise: What should Black play? Answer: He should insert 12 ... e3! (“this Morphy would not have missed!” says Kasparov); for example, 13 Qxe3 Rxf6 14 0-0-0 (not now 14 Qg5+ Rg6 since the knight is pinned) 14 ... Kh8 15 Bd3, and now Black can play 15 ... Nc6! without any problems, since 16 Ng6+ fails to 16 ... Qxg6 17 Bxg6 Bf4.

11 ... Nxg4 12 Nxg4 Qxh5 13 Ne5 Nc6 After the exchange of the pride of his position, the knight on e5, White has zero compensation for the 123

pawn. 14 Be2 Qh3 15 Nxc6 bxc6 16 Be3 Exercise: There are several interesting moves here. What do you think Morphy chose?

Answer: 16 ... Rb8 This prepares an impressive combination, even if 16 ... Bg4! was objectively best, bringing about a favourable exchange of bishops. 17 0-0-0

Black’s position is very advantageous: he is a pawn up and has no problems. Once again the simple 17 ... Bg4! is attractive, but that wasn’t what Morphy played. He chose a move that is objectively inferior, unable to say no to the artist within him. Exercise: Can you spot the idea that Morphy conceived, beginning with a surprising move?

Answer: 17 ... Rxf2!?! Apart from 17 ... Bg4!, Euwe suggested 17 ... Bf5, followed by ... Bg6. Morphy’s move, on the other hand, is the start of a very beautiful combination. “We should be grateful to Paul Charles Morphy that he didn’t try to convert his extra pawn smoothly now, but tried to win by spectacular means,” commented Karsten Müller. Question: Hmm, I can respect these comments and I appreciate beauty, but other champions preferred accuracy over “playing to the gallery”. Isn’t it rather

irresponsible to choose an unclear line ahead of one which is clearly advantageous? Answer: Obviously Morphy saw that 17 ... Bg4 (as well as 16 ... Bg4) was better, as you say; but we mustn’t forget that this game, just like Anderssen’s “Immortal” and “Evergreen” games, was a friendly encounter, not part of a match or tournament, so for both artists it would have been unpardonable not to give free rein to their imagination.

18 Bxf2

124

Exercise: What was the even more spectacular idea that Morphy had in mind?

Answer: 18 ... Qa3!! I don’t know what impression this move makes on you, dear reader, but when as a teenager I saw this move, and the previous one, for the first time, I felt the same amazement as when seeing Fischer’s famous 11 ... Na4!! against Donald Byrne. In those days I wasn’t surprised by seeing pieces offered to the opponent (I did it myself, deliberately or not) – what astonished me was that this move, and Fischer’s too, was actually good ... It’s obvious that 19 bxa3?? leads to mate – as occurs after 18 ... Ba3? 19 bxa3?? as well, but in that case White would be able to exclude the black queen from the attack with 19 Qe3!. 19 c3! The only move. In the event of 19 Qg5, the quickest win is 19 ... Rxb2!, threatening 20 ... Rxa2+ 21 Kd2 Bb4 mate, against which White has no satisfactory defence. Similarly, 19 Qc3 can be met by 19 ... Qxa2 20 Rdg1 Rxb2! and wins, since 21 Qxb2 runs into 21 ... Bf4+. 19 ... Qxa2 Threatening mate, and guaranteeing a draw at least. Exercise: How does White defend against 19 ... e3?! 20 Bxe3 Bf5 - ?

Answer: The only move is 21 Qc2!, which defends adequately after 21 ... Qxa2 22 Bd3 Bxd3 23 Rxd3 Rxb2 24 Qxb2 Ba3 25 Qxa3!, reaching an ending of queen and three pawns vs. two rooks and a bishop which can arise in several lines. The black pawns can’t easily advance and the evaluation depends on how well the forces coordinate. In this case White is a bit better, not having any of the coordination problems that we shall see in another line. 125

The white pieces are even more active after 23 ... Ba3 24 Qb1! Bxb2+ 25 Kc2 Qa4+ 26 Kd2 Bxc3+ 27 Rxc3 Rxb1 28 Rxb1 Qa2+ 29 Kc1, “parrying the attack and retaining the extra material.” (Kasparov)

20 b4 After 20 Qc2?, the simplest way is 20 ... Bf4+ 21 Rd2 Bxd2+ 22 Kxd2 Rxb2 or 22 Qxd2 Qa1+ etc. 20 ... Qa1+ 21 Kc2 Qa4+

22 Kb2? “An unjustified attempt to win,” commented Euwe. This is the decisive error, as Maróczy had earlier pointed out. The correct move was 22 Kc1!. This position has been analysed literally for more than a century, and according to the latest analysis Black is still not forced to give perpetual check with 22 ... Qa1+, but of course this may well not be the last word on it. There are at least three interesting moves for Black here: a) First of all it’s necessary to examine 22 ... Bxb4. This gives no more than a draw, since if Black tries to win with 23 cxb4 Rxb4 24 Qg5 Qa3+ 25 Kd2 Rb2+ 26 Ke1 Rxe2+? 27 Kxe2 Qf3+, White emerges with a decisive material advantage after 28 Kd2! (28 Ke1 Bg4! is less clear) 28 ... Qxf2+ 29 Kc3. b) Another attractive move is to try to open lines with 22 ... a5, but then 23 Qc2 Qa3+ 24 Qb2 axb4 25 Kc2 Ba6 26 Bf1! “is unclear”, according to Kasparov. c) Kasparov also commented: “Now, with the help of a computer, it can be established that 22 ... Bf5! 23 Be1! Qa1+ 24 Kc2 e3+ 25 Kb3 exd2 26 Rxa1 Re8 27 Ba6 dxe1Q 28 Raxe1 Rxe1 29 Rxe1 Bxh2 30 Bb7 Be4 31 Bxc6 Kf7 would nevertheless have given Black a minimal advantage”. Another defence to consider is 23 Be3, when we might reach an ending similar to those we’ve already seen with 23 ... Bxb4 24 cxb4 Rxb4 25 Qc2 Qa3+ 26 Kd2 Rb2 27 Qxb2 Qxb2+ 28 Ke1. Here the e4-pawn is still on the board, but even so it’s not clear whether White is worse. Exercise: How did Morphy punish White’s mistake? Answer: Now the bishop can be sacrificed under better conditions, due to a key move.

22 ... Bxb4! 23 cxb4 Rxb4+ 24 Qxb4 Qxb4+ 25 Kc2 If 25 Ka2, the black bishop has a neat way to join in the attack: 25 ... c5! 26 dxc5 d4! (Euwe) or 26 ... e3! 27 Bxe3 d4 (Kasparov).

126

Exercise: Materially White is not badly off. If he is given time to redeploy his rooks there will be a complex struggle ahead. What had Morphy calculated to prevent this?

Answer: 25 ... e3! This is the coup de grâce – once the bishop comes into play the struggle is over; Black gains a decisive material advantage. Kasparov commented: “Morphy’s sound framework and his wide-ranging tactics already resemble the play of a modern grandmaster. To combat such a hurricane was simply impossible.” 26 Bxe3 Bf5+ 27 Rd3 Or 27 Bd3 Qc4+. 27 ... Qc4+ 28 Kd2 Qa2+ 29 Kd1 Qb1+ 0-1 Game 17 P.Morphy-S.Boden London 1858 King’s Gambit [C30] 1 e4 e5 2 f4 Bc5 3 Nf3 d6 4 c3

Only on a couple of occasions did Morphy vary from this move, which plans to occupy the centre with 5 d4. One was in a game against Thomas Barnes (London 1858) where he tried 4 b4!?; the other was against Alphons Perrin (New York 1859), where he chose 4 Bc4. The main alternative is 4 Nc3.

4 ... Bg4 This prevents 5 d4 in view of 5 ... exd4 6 cxd4 Bxf3, when White has to play 7 gxf3, which leaves him in an awkward position after 7 ... Qh4+ 8 Ke2 Bb6. In one of the earliest recorded games, Greco-NN, Europe 1620, Black tried to prevent d2-d4 with 4 127

... Qe7, but without success since Greco played 5 d4! anyway. After 5 ... exd4 6 cxd4 Qxe4+? 7 Kf2, the game ended quickly: 7 ... Bb4 8 a3 Ba5 9 b4 Bb6 10 Bb5+ Kf8 11 Re1 Qf5 12 Re8 mate. The main reply is 4 ... Nf6. 5 Be2 Morphy wanted to maintain the central tension and try for aggressive play with his pieces (and his pawns, as we shall see). In two casual games Capablanca preferred to stabilize the situation with 5 fxe5, as did many other twentieth-century masters, such as Réti, Euwe, Bronstein, Larsen, planning to answer 5 ... dxe5 with 6 Qa4+ (this manoeuvre is usually credited to Marshall) to misplace Black’s light-squared bishop. D.Bronstein-V.Panov, Moscow 1947, continued 6 ... Bd7 7 Qc2 Nc6 8 b4 (evicting the bishop from the g1-a7 diagonal, otherwise 9 b5 wins the e5-pawn) 8 ... Bd6 9 Bc4, followed by d2-d3, when White’s play is slightly more comfortable.

5 ... Nc6 6 b4 Bb6

7 b5 Exercise: It is a mistake to play 7 d4? first, intending to meet 7 ... exd4 with the zwischenzug 8 b5, followed by 9 cxd4. Why does this fail? Answer: Owing to 7 ... Bxf3 8 Bxf3 Nxd4!.

7 ... Na5 Instead, W.Steinitz-V.Green, London 1864, saw 7 ... Bxf3 8 Bxf3 Nce7 9 d4, and here Black committed an error with 9 ... exf4? 10 Bxf4 Ng6 11 Bg3 Nf6 12 Nd2, when White gained a powerful centre without even having to sacrifice a pawn. As it happens, Morphy had earlier reached the same position as White in his 1858 match with Löwenthal, who surprisingly made the same mistake (as we shall see in Game 19).

8 d4 Bxf3 9 Bxf3 exd4 10 cxd4 Qf6

Exercise: How did Morphy reply to this attack on his centre pawn?

Answer: 128

11 Be3 Maintaining the mobility of his centre even at the cost of losing a tempo and sacrificing a pawn. The alternative was 11 e5, but after 11 ... Qe6 12 d5 Qe7 or 12 Nc3 dxe5 13 dxe5 Rd8, White’s advanced pawns loses their mobility and holes start to appear, without helping the development of his pieces. 11 ... Nc4 12 Bf2 Qxf4?! Things would have been less clear after 12 ... Ba5+! 13 Ke2 (not 13 Kf1?! Qxf4, when 14 Qa4? fails to 14 ... Qc1+) 13 ... d5! (not now 13 ... Qxf4? 14 Qa4 and wins) 14 exd5 Nb6. 13 0-0

White’s compensation is clear – the opening of the f-file is not something he is sorry about and the pawn is not missed, though the sacrifice must still be justified tactically. 13 ... Nf6 Here 13 ... Ne3 was to be considered. Exercise (easy): In that case, how would 14 Qd2 be punished? Answer: With 14 ... Qxh2+!.

But despite being two pawns up after 14 Qc1 Bxd4, Black is playing only with his already developed pieces, and the tactics come out in White’s favour after a series of not very obvious moves: 15 Na3 Qh6 (to break the pin) 16 Rd1! Be5 17 Bxe3 Qxh2+ 18 Kf1 Bxa1

Now 19 Qxa1 Nf6, followed by castling, would give Black a satisfactory position, with a rook and three pawns for the two minor pieces, which are not very well coordinated. But instead of taking the bishop straight away, White has the zwischenzug 19 b6!!. If the bishop retreats with 19 ... Be5, say, then 20 Nb5 and there is no way to prevent White from 129

winning material; e.g. 20 ... c6 (or 20 ... Rd8? 21 Nxc7+, followed by 22 bxa7) 21 Nc7+ Ke7 22 Nxa8 Nf6 and Black is a piece down, as there is no way of trapping the white knight; while after 19 ... axb6 20 Nb5! Kf8 21 Qxa1, or 19 ... c6 20 Qxa1 Nf6 21 bxa7, White recaptures the bishop under far superior conditions.

Exercise: What is White’s best move?

14 Qd3 Morphy contents himself with the advantage of better development and he has a clear plan of attack in mind. First he wants to drive the black knight away, in order to play Nc3, since 14 Nc3?! immediately would be met by 14 ... Nd2 15 Re1 0-0. Answer: However, it was possible to attack at once, without waiting for all the pieces to be developed. After 14 Qe2! Na5, it’s clear that the queen is much more aggressively placed on e2, since White can now play 15 e5!, followed by e5-e6 to open the position, when his advantage is almost decisive; for example, 15 ... dxe5 16 dxe5 Nd7 17 e6 Nf6 18 Bxb6 axb6 19 Bc6+ wins the queen.

14 ... Na5 15 Nc3 0-0 It was more prudent to eliminate one of the white bishops with 15 ... Ng4. After, for example, 16 Bxg4 Qxg4 17 Nd5 0-0 18 Ne7+ Kh8 19 Nf5, with the idea of Rae1-e3, the offside black knight means that White is not worried about being a pawn down, but neither does Black have any immediate problems.

Exercise: All the white pieces are now in play, apart from the rook on a1. There are several possibilities here – which do you think Morphy chose?

Answer: 16 g3 With this and the following move Morphy tries to exploit the uncomfortable position of the black 130

queen. Question: This isn’t very threatening or impressive though, is it? Answer: Morphy plays with what he has. The position is approximately equal for now.

16 ... Qh6 17 Kg2 Morphy continues with his idea; now he’s controlling h3. 17 ... Rae8 18 Rae1 Kh8?! This is very passive; 18 ... Qg6!? would have placed more obstacles in White’s way. 19 Be3 Qg6

Exercise: How did White follow up the plan initiated with 16 g3 - ?

Answer: 20 Ne2 Of course; the threat is now 21 Nf4 Qh6 22 Ne6. The move is a simple one but must be justified tactically, since the e4-pawn is left hanging. 20 ... h6 This prepares a bolthole for the queen. 20 ... Nxe4? was impossible owing to 21 Nf4 Qf5 22 g4 and wins. The exchange sacrifice 20 ... Rxe4 21 Nf4 Rxf4 22 Qxg6 hxg6 23 Bxf4 Bxd4 was also unsatisfactory. Although Black has no less than three pawns for the exchange, his pieces are placed so awkwardly, and White’s so well, that his compensation is insufficient. Here’s a sample continuation: 24 Re7 Bb6 25 g4! Kg8 26 g5 Nh5 27 Bd2 Nc4 28 Bc1 a6 29 Bd5 Ne5 30 Bb2 and f7 falls with a big advantage to White. 21 Bd2!

131

Question: What’s so good about this? Wasn’t it better to play the obvious 21 Nf4 first? Answer: You’re right that 21 Nf4 was a good alternative, though after 21 ... Qh7 the best move is probably 22 Bd2 anyway.

Morphy’s choice strikes me as admirable – the idea is to lend more strength to the future advance e4e5 (which will come after the preparatory Nf4). Moving the bishop to d2 not only activates the rook on e1, it also puts pressure on the knight on a5, while the bishop may prove useful on the a3-f8 diagonal. 21 ... d5? A very weak reply; Black cracks under the pressure and makes White’s task easier – after a short sequence Black loses at least a pawn. However, his position was now very uncomfortable in any event; for instance, the strength of 21 Bd2 can be seen after 21 ... c6 22 Nf4 Qh7 23 Bb4. The exchange sacrifice 21 ... Rxe4 22 Nf4 Rxf4 23 Qxg6 fxg6 24 Bxf4 Bxd4 might again be considered, even if the evaluation is not very different than on the previous move: White has the advantage after 25 h4 or even 25 Re7 g5 26 Bd2, but this was preferable for Black than the text move. 22 Nf4 Qh7 23 e5 Qxd3 24 Nxd3

The d5-pawn will fall and with it Black’s whole position. 24 ... Nc4 Black decides to sacrifice the exchange instead. After 24 ... Nh7 25 Bxd5 Rd8 26 Bxf7 Rxd4 27 Bg6, Black’s position is hopeless; and 24 ... Ne4 25 Bb4 Rg8 26 Bxe4 dxe4 27 Rxe4 f6 28 e6 is little better. 25 Bb4 Ne4 26 Bxf8 Rxf8 27 Nf4!? Morphy prefers to play actively with his 31st move in mind. Objectively, the straightforward 27 Bxe4 dxe4 28 Nc5 was stronger. 27 ... Ned2 27 ... Ncd2? 28 Bxe4 Nxf1 29 Bxd5 is worse for Black. Exercise: What had Morphy planned here?

Answer: 28 Bxd5! He wants to return the exchange in return for capturing an important pawn, all based on concrete tactics, as we keep seeing. 28 ... Nxf1 29 Bxc4 Nd2 30 Bd5 Bxd4? Black had clearly overlooked Morphy’s reply, otherwise he would have played 30 ... g5!, when he is still in the game.

132

Exercise: What was that key move that Morphy had prepared?

Answer: 31 e6! He creates a passed pawn which will decide the game. The pawn is immune from capture owing to Ng6+. 31 ... g5 32 e7 Re8 33 Bxf7 gxf4 34 gxf4 Rxe7 35 Rxe7 1-0 Game 18 P.Morphy-S.Boden London 1858 Two Knights Defence [C55] 1 e4 e5 2 Nf3 Nc6 3 d4 exd4 4 Bc4 Bc5 5 0-0 d6 6 c3 Nf6?! Question: Why is this developing move dubious? Answer: Later developments in opening theory indicate that, instead of submissively allowing White to occupy the centre with 7 cxd4, it is more critical either to accept the gambit with 6 ... dxc3 or contest the centre with the pin 6 ... Bg4. Naturally, it was impossible to know that at the time this game was played.

7 cxd4 Bb6 8 Nc3 0-0

9 d5 Question: This decision by Morphy surprises me. White’s vaunted pawn centre loses much of its mobility, the diagonal of the c4-bishop is blocked ... How to explain this? Was there really nothing better? Answer: Yes, it’s true that this isn’t the most flexible move in the position. White is faced with the threat of 9 ... Bg4, putting pressure on 133

his centre. So ideally would like to play 9 h3, but this isn’t at all promising here owing to the ‘fork trick’ 9 ... Nxe4! 10 Nxe4 d5, which breaks up White’s centre and equalizes the game.

However, White could play the prophylactic retreat 9 Bb3, and answer 9 ... Bg4 by supporting the centre with 10 Be3. In P.Leonhardt-G.Maróczy, Karlsbad 1907, White retained his opening advantage after 10 ... Re8 11 Qd3 Bh5 12 Rae1. Note that White is not afraid of 11 ... Bxf3 12 gxf3 since, in return for the doubled pawns, White’s centre has been strengthened and he can hope to exploit the half-open g-file and his pair of bishops. Exercise: In this variation, can you indicate why the combination 11 ... Nxe4? 12 Nxe4 Bf5 is unsound?

Answer: First of all because White can play 13 Bxf7+! Kxf7 14 Nxd6+, winning a pawn directly, but 13 Nfg5! may be even stronger; e.g. 13 ... Bg6 14 Bxf7+! (or 14 Nxf7!) 14 ... Bxf7 15 Bxf7+ Kxf7 16 Ng5+ and wins, or if 13 ... d5 then 14 Nxf7! Kxf7 15 Nd6+ Qxd6 16 Qxf5+ Kg8 17 Qxd5+.

9 ... Na5? Deploying a knight on the edge of the board is something that should only be done after very careful evaluation of the consequences, since it runs the risk of requiring too many tempi to bring it back into play. Morphy will show that here it is an incorrect decision. It was better to play 9 ... Ne5 10 Nxe5 dxe5, when Black would have few grounds for complaint. 10 Bd3 With the threat of 11 b4. 10 ... c5

Exercise: Black’s last move was forced, in order to save his knight, but it fails to solve the other defects in his position. How did Morphy highlight this? 134

Answer: 11 Bg5! The pawn structure is similar to a Benoni, but with the a5-knight and the b6-bishop out of position on the queenside. These snags are underlined by this pin, which can only be shaken off at the cost of weakening the kingside. 11 ... h6 12 Bh4 Bg4 Black is also trying to restrict White’s mobility with a pin on the king’s knight, but it will soon be clear that these two pins are made under very different conditions. Exercise: What do you think Morphy played here?

Answer: 13 h3! First of all, White “puts the question” to the bishop, forcing a decision. 13 ... Bh5 A difficult choice – with this move Black maintains the pin, but risks this bishop being pushed out of play as well. The alternative 13 ... Bxf3 14 Qxf3 wasn’t pleasant either: as mentioned above the pawn structure resembles a Benoni, with obvious disadvantages for Black, since his bishop is over on b6 (instead of g7) and his knight is on a5 (instead of d7), with the result that his queenside pawn majority has been devalued. White, meanwhile, has a clear plus in the centre and on the kingside.

Exercise: How did Morphy deal with the pin?

Answer: 14 g4! There was no doubt that ... g7-g5 would weaken Black’s kingside, but the situation is very different with respect to White’s g2-g4 advance. Black is in no position to be able to exploit the weakening of the white kingside. Much more important factors are that the black bishop is shut out of play, while White gains space and, by unpinning, restores the mobility of his f3-knight. 14 ... Bg6 15 Qd2 Question: Given that we are, as you say, in a sort of Benoni, wasn’t it already possible to carry out the typical break 15 e5 - ? Answer: It was indeed possible and after, for example, 15 ... dxe5 16 Nxe5 Bxd3 17 Qxd3, White is slightly better. But Morphy prefers not to open the game just yet; he wants to have all his pieces activated first. With his last move he plans to bring his only inactive piece, the a1-rook, into play. This improves his position in way that Black is unable to imitate.

15 ... Re8 16 Rae1 Bc7 Question: This is hardly a very active move, is it? Answer: No, it isn’t, but there was scarcely anything better. The 135 positive part is it that reduces the impact of the e4-e5 advance. It’s also

the first step towards bringing the a5-knight back into the game by playing ... b7-b6 or, ideally, ... b7-b5.

17 Nb5 With the threat of 18 Bxf6 gxf6 19 Qxh6. 17 ... Kh7?

Exercise: Black defends the h6-pawn prophylactically, so that he can answer Bxf6 with ... g7xf6 without losing h6-pawn. How did Morphy show that this wasn’t a good decision?

Answer: 18 Bxf6! With this exchange Morphy starts a manoeuvre which gives him a clear advantage. In the first place the black kingside suffers damage. This was not the only good move – the thematic 18 e5! was also strong here. 18 ... gxf6 Exercise: How did Morphy proceed?

Answer: 19 Nxc7! Question: Exchanging the passive bishop – was this best? Answer: Yes. It will soon become clear that, instead of 17 ... Kh7, Black would have done better to play 17 ... Bb8, preventing this exchange of his “bad” one. The reason is that “it’s not the pieces that are exchanged but those that remain on the board that matter” (Tarrasch), so an important factor here is that the black knight still remains inactive on a5. And, of course, the weakness of Black’s kingside is important too.

19 ... Qxc7 20 Qc3 This attacks f6 and threatens to win a piece with 21 b4, since the c5-pawn is pinned. 20 ... Qd8

136

Exercise: What plan did Morphy have in mind to follow up the recent piece exchanges?

Answer: 21 Nh4 Starting an infantry attack; the knight makes way for the advance of the f-pawn. 21 ... b6 22 f4 Kg7 The alternative was 22 ... Kg8, evading the ‘X-ray’ pressure from the queen on c3, but it wasn’t really any better as White could continue as in the game anyway. Exercise: How did Morphy continue here?

Answer: 23 Nxg6 Yet another exchange that might appear surprising at first sight, since it repairs Black’s structure. But this is only the first step. Question: Why not shut the bishop completely out of play with 23 f5 - ? Answer: That is good too, yes, as White is virtually a piece up, but it would take longer to win the game. Morphy’s approach is a more concrete one.

23 ... fxg6

Exercise: How did Morphy complete his plan?

Answer: 137

24 e5! This pawn break was the idea, opening the game in order to attack Black’s castled position, involving all White’s active pieces on the kingside; in contrast, Black is playing without his a8-rook and a5-knight. Going back a move, 23 e5 would also have been strong. 24 ... Rc8 After 24 ... c4 25 Bb1, White demonstrates that g6 is a weakness; for example, 25 ... dxe5 26 fxe5 Qxd5 27 Rxf6 Re6, and now the quickest way is 28 b4! cxb3 29 Qc7+ Kh8 30 Bxg6, when there is no defence. Exercise: There are now several ways to try to breach Black’s position. What do you think was Morphy’s choice?

Answer: 25 Bb1! Consistent with 23 Nxg6, emphasizing the weakness of g6 again. In some lines White can line up a battery with Qc2 or Qd3.

25 ... Kf7 Trying to keep the game closed; opening things up with 25 ... dxe5 26 fxe5 Qxd5 27 exf6+ Kh7 would allow White several ways to win, the quickest being 28 Bxg6+! Kxg6 29 Qc2+ etc. 26 e6+ Kg7

27 Qd3 As planned – this destroys Black’s kingside and wins material. However, it was even stronger here to maintain the pressure on f6 with 27 g5!, threatening e6-e7+, or if 27 ... Qe7 then 28 Re2, followed by Rg2. 27 ... f5 Forced. 28 gxf5 Qf6 29 fxg6 Qxb2? Allowing White’s pawn majority to advance is fatal. The only way to resist was by 29 ... Re7!, praying that that the blockade on the dark squares holds. This would require White to open a second front, possibly by preparing the advance b2-b4. 30 f5 Qf6 If 30 ... c4, rather than 31 Qg3 followed by Qxd6, it is simpler to play 31 f6+ Qxf6 32 Rxf6 cxd3 33 Rf7+ Kg8 34 Bxd3 with an easy win.

138

Exercise: How did Morphy administer the coup de grâce?

Answer: 31 e7! Preventing ... Re7 and preparing to remove the blockading queen with 32 Re6. 31 ... c4 Of course 31 ... Rxe7? loses to 32 Rxe7+ Qxe7 33 f6+. 32 Qg3 c3 33 Re6 Qd4+ 34 Qf2 Qxd5 35 f6+ 1-0 It’s mate after 35 ... Kh8 36 g7+ Kg8 37 f7+ Kxg7 38 Qf6. The 17th of July 1858 saw the start of the first serious encounter of Morphy’s stay in Europe, a match with Johann Löwenthal. Although Morphy had beaten him on the other side of the Atlantic, Löwenthal was a professional player and shortly before had defeated Adolf Anderssen in Manchester. Furthermore, not long after the match against Morphy, in August 1858 Löwenthal achieved the greatest success of his career, winning the British Chess Association Congress knockout tournament in Birmingham. In the first game of their match, Morphy achieved a good position and, as usual, played to win, taking risks and objectively standing worse on a number of occasions, but it ended in a draw. The second game of the match is given in Supplementary Game 19.1; we shall see that, in spite of his victory, Morphy’s play was not yet at its highest level. In the third game Löwenthal gained the advantage; he could have agreed a draw but went on playing, apparently believing that his position was sound and that he could press without risk, but Morphy soon showed that this was not an easy task. Let’s now look briefly at an extract from this game. J.Löwenthal- P.Morphy 3rd matchgame, London 1858

139

White to play Here Löwenthal could still have forced a draw with 30 Re8+, but ... 30 Qe2? White is relying on his rook on the seventh and the passivity of the black bishop to be able to keep playing for a win, but in reality this move doesn’t threat 31 Rxa7 since 31 ... Re1 then wins. 30 ... Kf8! Taking the initiative by threatening 31 ... Rxf1+. In a few moves the situation is completely turned round. 31 Re5 f4 32 f3 Qc1 33 h4 h6 34 c5 Kg7!

Surprisingly, White no longer has any useful moves, and when his pawn moves run out he will find himself in zugzwang. 35 Re4 Qxc5+ 36 Kh2 Qc1 37 Kg1 Rd2 38 Qa6 Rxa2 39 Qd3 Rd2 40 Qa6 Rd1 41 g3 fxg3 42 Kg2 Qc5 43 Kxg3 Qg1+ 44 Bg2 Rd2 45 Qf1 Qxf1 Again playing like Capablanca in the following century. Morphy could keep the queens on and proceed with advantage, but heading for the endgame is the simplest way. 46 Bxf1 Kf6 47 Bc4 Bxc4 48 Rxc4 Rd6

140

The rook ending two pawns up is clearly winning. Morphy didn’t conduct it in entirely exemplary fashion, and in fact gave Löwenthal the chance of a theoretical draw at one point, but still ... 0-1 in 80 moves. Game 19 P.Morphy-J.Löwenthal 4th matchgame, London 1858 King’s Gambit [C30] 1 e4 e5 2 f4 Bc5 3 Nf3 d6 4 c3 Bg4 5 Be2

In the second game of the match Morphy chose 5 Bc4, and was very lucky to win, as we shall see in Supplementary Game 19.1. 5 ... Bxf3 Boden played 5 ... Nc6 first in Game 17. 6 Bxf3 Nc6 In M.Chigorin-G.Marco, Monte Carlo 1901, Black tried 6 ... exf4, in order to answer 7 d4 with 7 ... Qh4+, but after 8 g3! fxg3 9 hxg3 Qxg3+ 10 Ke2 Bb6 11 Rg1 Qh2+ 12 Rg2 Qh3 13 Rxg7 Nd7, the strong white centre provided more than enough compensation for the pawn. White could continue, for instance, with 14 a4 a5 15 Bg4 Qh2+ 16 Kd3 and Black has development problems.

7 b4 Morphy was very fond of this queenside expansion, which he also used in game two of the match. Another plan was 7 d3, followed by Qe2 and Be3. 7 ... Bb6 8 b5 Nce7 9 d4

This was the idea, occupying the centre and leaving Black with the problem of how to deal with the newly constructed pawn wall. Exercise: How should Black respond?

Answer: 9 ... exf4?

141

A bad decision, which strengthens the opposing centre, shuts the b6-bishop out of play, and opens the f-file in White’s favour, without gaining anything in return – White has not even had to sacrifice a pawn (as was the case in the Chigorin game mentioned above). Black had at least two satisfactory continuations. For instance, there was nothing wrong in maintaining the tension and continuing development with 9 ... Nf6, since taking twice on e5 isn’t advisable for White. Another idea was 9 ... exd4 10 cxd4 d5, and if 11 e5 Black can play 11 ... Nf5, followed by ... Nge7, as well as a timely ... a7-a6, opening the a-file in his own favour. 10 Bxf4 Ng6 11 Be3 Steinitz opted for 11 Bg3 here in the notes to Game 17, which may be a slight improvement. 11 ... Nf6 Question: Wasn’t it better at least to deprive White of the bishops with 11 ... Nh4 - ? Answer: This was playable, but wouldn’t change the evaluation. White’s advantage isn’t based on the bishop pair but on a solid centre and the possibility of using the half-open f-file, which the removal of the f3-bishop would do nothing to change.

12 Nd2 0-0 13 0-0

13 ... h6 A waiting move, preventing the pin with Bg5. Question: Okay, but after 13 ... Re8, say, the pin with 14 Bg5 could just be answered with 14 ... h6, couldn’t it? Answer: Perhaps 13 ... Re8?! would not be the best choice. Agreed, the pin could be broken, but 14 Bg5 h6 15 Bxf6 Qxf6 leaves the black queen ‘X-rayed’ by the rook on f1 which might prove dangerous for Black after a timely e4-e5. Not 16 e5 immediately, since Black can play 16 ... Qxe5, but 16 a4! creates serious threats and 16 ... a5?! 17 bxa6 Rxa6 fails to 18 Be2.

14 a4! Threatening a4-a5, as well as defending the b5-pawn prophylactically. 14 ... c6 The bishop on f3 can be pleased with this pawn contact. 15 Qe2 Morphy is still unbothered by the possible exchange of his f3-bishop with 15 ... Nh4. Otherwise he could have played 15 Qb3 – or 15 Qc2, with the same idea as in the game, when Black gains nothing by harassing the bishops with 15 ... Ne5 16 Be2 Neg4 17 Bf4 g5? since the weakness of his kingside is revealed after, for instance, 18 Qd1! Qd7 19 h3. It would be better to make a useful move, such as 15 ... Rc8. 15 ... Re8?! It’s unclear whether it would have been better to play 15 ... Nh4 here, but given Black’s cramped position, an exchange of pieces was well worth considering.

142

Exercise: How did Morphy proceed?

Answer: 16 Qd3! The queen evades the ‘X-ray’ pressure of the e8-rook and offers the f3-bishop a line of retreat in the event of ... Nh4. 16 ... d5?! Black wants to neutralize the f3-bishop, but the cost is too great as this grants White more space. Since e4-e5 isn’t a threat yet, something like 16 ... Rc8 was playable, but not 16 ... Qd7? due to 17 a5 Bc7 18 a6, weakening the black position. 17 e5 Nd7 Exercise: What is the best continuation for White?

18 Bh5? The idea is good, but it fails tactically. Answer: The preparatory move 18 Kh1! was much better, with a great advantage for White, because now Black completely lacks counterplay.

Exercise: How should Black exploit White’s error?

18 ... Re6? The inaccuracies from both players testify to the tension each felt in this third game. Answer: Black had the strong counter-strike 18 ... Ndxe5!. The knight cannot be taken, since 19 dxe5? Rxe5 20 Bd4 Rxh5 21 Qf3 Re5 22 Qxf7+ Kh7 gives Black an excellent position. Note that White can’t regain the pawn with 23 Qxb7? as after 23 ... Re7 24 Qa6 (24 143

Qxc6? loses the queen to 24 ... Rc8 or 24 ... Ne5) 24 ... c5! 25 Bf2 c4!, intending ... Nf4 and/or ... Re2, White’s position collapses. White’s best response in the circumstances would be 19 Qf5! Nh8 (forced) 20 a5! with an unclear position; for example, 20 ... g6 (20 ... Bxa5 21 dxe5 g6 is riskier; for example, 22 Qf2 gxh5 23 Qg3+ and Black’s king is in jeopardy, but he may yet survive after 23 ... Ng6 24 Rf6 Bb6!) 21 Qh3 Bxa5 22 bxc6 (22 dxe5 is met by 22 ... Rxe5, attacking both bishops; i.e. 23 ... Rxh5 and 23 ... Rxe3 24 Qxe3 Bb6) 22 ... bxc6, when Black is two pawns up but his king position is weak and the knight on h8 is nothing to boast about, so White isn’t worse. Exercise: How did Morphy continue here?

Answer: 19 a5! An important preparatory move, deflecting the bishop from the g1-a7 diagonal, in order to gain more freedom of action. 19 ... Bc7

Exercise: How did Morphy crown his idea?

Answer: 20 Rxf7! 19 ... Bxa5 would have been answered in the same way. Looking back, if White had played the immediate 19 Rxf7, Black could have countered with 19 ... Ndxe5! 20 Qf5 Nxf7 21 Qxe6 Qe7, and while White is better after 22 Qxe7 Nxe7 23 a5, followed by 24 a6, it’s not a decisive advantage. 20 ... Kxf7 Now that the black bishop has been driven from the a7-g1 diagonal, 20 ... Ndxe5 21 dxe5 Nxe5 would fail to 22 Qf5, since Black doesn’t have 22 ... Bxe3+. 21 Qf5+ Ke7 22 Bxg6 Qg8 After 22 ... Qf8 23 Qh5, White threatens both 24 Qh4+ and 24 Rf1; while returning the exchange with 23 ... Rxg6 24 Qxg6 Qf7 means acquiescing to an unpleasant ending a pawn down, although it would be even better for White to keep the queens on.

144

Exercise: There are several good moves here. What do you think was Morphy’s choice?

Answer: 23 Bf2 With the threat of 24 Bh4+. Interpolating 23 bxc6 or 23 a6 was also strong. 23 ... Nxe5?!

As insufficient as any another move; after 23 ... Nf8, for instance, White wins with 24 Bh4+ Kd7 25 Bf7. 24 dxe5 Rf8 24 ... Rxe5 also loses, most simply to 25 Bh4+ Kd6 26 Bg3. 25 Bc5+ Kd8 26 Bxf8 Rxe5 27 Qf2 Qe6 28 b6 axb6 29 axb6 Qxg6 30 bxc7+ Kxc7 31 Rb1 1-0 Supplementary Game 19.1 P.Morphy-J.Löwenthal 2nd matchgame, London 1858 1 e4 e5 2 f4 Bc5 3 Nf3 d6 4 c3 Bg4 5 Bc4 Bxf3

The less radical 5 ... Nc6 is more usual, as Bird played in London 1858; although Morphy won resoundingly, it was nothing to do with the opening. The game continued 6 b4 Bb6 7 a4 a6 8 h3 Bxf3 9 Qxf3 Nf6 10 d3; Morphy reached similar positions more than once.

Here Black could have played 10 ... 0-0! and if 11 f5?! then 11 ... d5! is a good reply: after 12 Bxd5 Nxd5 13 exd5 Ne7 14 c4 f6, in addition to the discoordination of the white pieces, the white king has no safe refuge, which will become significant as soon as the position opens up (e.g. with ... c7-c6), so Black’s compensation for the pawn will be more than enough. Instead, Bird played the inaccurate 10 ... Qe7?!, also with the idea of ... d6-d5, but after 11 f5! Rd8 12 Bg5, his position was inferior. In P.Morphy-S.Boden, London 1858, Black tried 5 ... Qe7, in similar fashion to Greco’s opponent in the notes to Game 17. 145

Exercise: What do you think was Morphy’s reply? Answer: Since this was a friendly game, Morphy played very aggressively – if perhaps not entirely soundly – putting constant pressure on the black position and compelling his opponent to take a series of difficult decisions. Instead of, for example, 6 fxe5 dxe5 7 h3, and if 7 ... Bd7?! 8 d4!, when 8 ... exd4 9 cxd4 Qxe4+? 10 Kf2 leads to a lost position for Black (as in Greco’s game), Morphy chose the less clear 6 d4 exd4 7 0-0!? and continued to sacrifice pawns with 7 ... Nc6 8 b4 Bb6 9 a4. Black defended reasonably well for a large part of the game, which continued 9 ... dxc3+ 10 Kh1 c2 11 Qxc2 Bxf3 12 gxf3 Nxb4 13 Qb3 a5 14 Nc3 Nf6 15 e5 dxe5 16 fxe5 Nh5 17 Ne4 0-0 18 Bb2?! Nf4 19 Rg1 Rad8?! (there was nothing wrong with 19 ... Bxg1, since 20 Rxg1 Kh8 21 e6 f6 is just good for Black) 20 e6 Bd4 21 exf7+ Kh8 22 Rg4 Bxb2 23 Qxb2 Rxf7 24 Bxf7 Qxf7 25 Ng5

but here Boden collapsed with 25 ... Qd5? (the exchange of knights favours White, since he gains space for his rooks to operate; 25 ... Qf5! still leaves chances for both sides) 26 Rxf4 Qxg5 27 Rg1 Qh6? (now the attack becomes decisive; 27 ... Qe7 was necessary) 28 Rf7 Rg8 29 Rxc7 Nd3 30 Qd4 10. This was a game in which the energy and ultra-aggressive manner (if not the accuracy) of White’s play was rewarded. 6 Qxf3 Nf6

146

7 b4 This move, which in reality is one of the ideas behind 4 c3 and which, as we have seen, Morphy was fond of playing, has come in for criticism, but it isn’t clear that the suggested alternative is really any better: if 7 d4 exd4 8 e5 dxe5 9 fxe5, Black can defend f7 with the zwischenzug 9 ... Qe7!. White can then gain material with 10 Qxb7, but after 10 ... 0-0! 11 Qxa8 Nbd7, followed by 12 ... Nxe5, White is so behind in development and his king so insecure that Black’s compensation for the rook (!) is more than sufficient, and in fact the analysis engines award Black the advantage; while after 10 Kd1 Nfd7 11 Qxb7 Nb6, Black is not worse either. 7 ... Bb6 8 d3 The point of the expansion with 7 b4 is not to open the game prematurely with 8 d4?! since 8 ... exd4 9 e5 dxe5 10 fxe5 can now be met by 10 ... 0-0! 11 exf6 Re8+ 12 Kf1 Nc6, when Black has more than enough compensation for the piece. Nor is 8 fxe5 dxe5 9 d4 any better, as apart from 9 ... exd4 10 e5 00! again, Black can also play 9 ... Bxd4!. 8 ... Nbd7 9 f5 Qe7 Löwenthal doesn’t want to commit his king and instead prepares to counter with ... d6-d5 at an opportune moment. 10 g4 h6 The complications arising from 10 ... d5 at once weren’t bad for Black; e.g. 11 Bxd5 Nxd5 12 exd5 a5! 13 b5 Qc5 14 a4 Nf6, but there is nothing wrong with his useful waiting move. 11 Ke2?! It is noteworthy how Morphy relies so much on his advantage in space and the closed (for now) character of the position. 11 ... c6

12 g5? Another excessively optimistic move; the opening of the game, and the h-file in particular, will be in 147

Black’s favour. 12 ... hxg5 13 Bxg5 d5! This move highlights the failure of White’s concept. Now if 14 exd5 then 14 ... e4! 15 Qxe4 (not 15 dxe4? Ne5!) 15 ... Nxe4 16 Bxe7 Nf2 etc (which is even better than 16 ... Ng3+ 17 hxg3 Rxh1).

14 Bb3 Qd6 15 Nd2 a5! More black pieces come into play; now it’s the turn of the rook on a8. 16 bxa5 Rxa5 17 h4 Nh5! Black plays this phase of the game very well, exploiting all the holes in White’s position. 18 Nf1 Nc5 19 Bc2

19 ... Rb5 Heading for b2; even better was 19 ... Ra3 20 Bd2 Ba5 and White’s position collapses. Over the next few moves Löwenthal’s play begins to deteriorate – he misses several stronger continuations and fails to cash in what was undoubtedly a clearly winning position. 20 Bc1 dxe4 It was better to play 20 ... Ba5! or 20 ... Nf4+ 21 Bxf4 exf4, followed by 22 ... Ba5, 22 ... Rb2, or even 22 ... 0-0, bringing the rook to the centre or the queenside. 21 dxe4 Rb2?? A serious error in calculation. Black’s advantage would still have a clear advantage after, for instance, 21 ... Nf4+ 22 Bxf4 exf4. 22 Bxb2 Nf4+ 23 Ke1 Nfd3+ 24 Bxd3 Nxd3+ 25 Kd2

Löwenthal later commented that, when he played 21 ... Rb2, he thought that this move was impossible. 25 ... Nxb2+ 26 Kc2 Qa3?! Here 26 ... Nc4 would have offered greater resistance. 27 Nd2! Bc7 28 Nb1 1-0 148

Question: Morphy was really lucky in this game! Answer: You’re right. As previously remarked. Morphy’s play at the start of each of the three most important matches of his career, this one and those against Harrwitz and Anderssen, was quite weak compared with his usual level. There was some talk of his having health problems. In the fifth game Löwenthal clearly outplayed Morphy – he gained a material advantage and won in 70 moves, which brought the score in the match to 3-1 in Morphy’s favour. However, Morphy easily won the sixth and seventh games after serious blunders by Löwenthal; in these games there was hardly any struggle at all. In Supplementary Game 26.1 we shall see the eighth game, which ended in victory for Löwenthal, basically because Morphy took too many risks. Let’s now look at the ninth game. Game 20 J.Löwenthal-P.Morphy 9th matchgame, London 1858 Ruy Lopez [C64] 1 e4 e5 2 Nf3 Nc6 3 Bb5 Bc5 4 c3 Qe7?!

For some strange reason Morphy repeats an original but dubious-looking idea that Boden had played against him shortly before this match, and which Morphy as White had not handled very well. 5 0-0 f6?!

As Löwenthal pointed out: “It seems to us to give a more cramped position than any other defence, and we cannot recommend it.” Question: Indeed, how do you explain why Morphy should choose this variation, which doesn’t look at all like the sort of line in which he would typically overcome his opponents? Answer: It is impossible for us to know, since Morphy himself never explained his decision. Morphy generally shunned static positions and preferred the open game, in which he had no rival. He liked to maintain the pressure, even at the cost of material sacrifices, if necessary. We can only speculate that he might have wanted to practise and learn different positions from the ones he usually obtained. It’s also possible that, since he had been unsuccessful with White against this line, he wanted to improve his understanding of it by playing it with Black.

In one of his ‘family’ games, P.Morphy-E.Morphy, New Orleans 1856, Black chose 5 ... Nf6?! (the only reasonable move here is 5 ... a6, to be able to retreat the bishop to a7 after 6 Ba4 and also have ... b7-b5 available) 6 d4 Bb6 7 Bg5! and his position was already inferior: 7 ... exd4 fails to 8 e5 and 7 ... d6 to 8 d5 when Black regrets not being able to play ... a7-a6 and ... b7-b5. The game continued 7 ... h6 8 Bxf6 gxf6?! (but 8 ... Qxf6 just loses a pawn to 9 Bxc6 dxc6 10 Nxe5) 9 d5 Nd8 10 Nh4! and ended shortly: 10 ... c6 11 Nf5 Qc5 12 b4 Qf8 13 dxc6 dxc6 14 Nd6+ Ke7 15 Qd3 cxb5? 16 Nxc8+ Rxc8 17 Rd1 Qg7 18 Qd7+ Kf8 19 Qxc8 1-0. 6 d4 Bb6 7 Na3 Nd8 8 Nc4 Nf7 9 Ne3 Seeking more than simply degrading Black’s structure with 9 Qa4, followed by Nxb6, when Black must play ... c7xb6. 9 ... c6 10 Nf5 Qf8 149

Black will soon be able to drive the white pieces away from their active positions, but at the cost of moving almost all his pawns, which is of debatable benefit, while the black knights will be forced to take up positions which are far from ideal. Of course, to derive maximum benefit from this will require White to find the correct plan and execute it accurately. 11 Bd3 The earlier game P.Morphy-S.Boden, London 1858, saw 11 Ba4 g6 12 Ne3 d6 13 d5 Bd7 14 dxc6 bxc6 15 Nc4 Rc8 16 b3 Be6 17 Qd3 Qe7 18 Ba3 Bxc4 19 Qxc4 Kf8 20 Rad1 c5 21 Rd3 (alternatively, 21 Ne1, heading for d5 via c2 and e3, was reasonable; as was the idea of bringing the light-squared bishop to a more active position with 21 Bb5 and Bc4, after the white queen makes way) 21 ... Ngh6 and now, instead of playing on the queenside, Morphy tried at all costs to attack the black king, sacrificing a pawn after 22 Kh1?! Kg7 23 Bc1 Rhf8 24 g4? Nxg4, when he only managed to scrape a draw after a long struggle.

11 ... g6 12 Ng3 d6 13 a4! Löwenthal expands on “his” wing, which is the queenside, where he has the advantage in space. 13 ... Bg4 14 a5 Bc7 15 h3 Bd7 16 Qb3 Nd8 17 Re1 Max Lange suggested 17 Bc4, followed by 18 dxe5, but what Löwenthal played looks better, continuing to make progress on the queenside and completing his development before radically altering the pawn structure. 17 ... Be6 18 Qc2 Ne7 19 b4! White continues to gain space on the queenside, while Black has little to boast about. 19 ... Qg7 20 c4 Nf7 21 Be3 0-0 22 d5 Bd7

Löwenthal’s play has been successful; White has gained a lot of space on the queenside at absolutely no cost. Disregarding the placement of some of the pieces, we are in a sort of King’s Indian position (an opening that was practically unknown at the time of this game), in which White’s queenside offensive is far more advanced than Black’s kingside counterplay. 23 Rad1 White has several ways to deploy his pieces. His rook isn’t especially useful on d1 for the natural plan of opening lines on the queenside, although we shall see that Löwenthal has a different idea in 150

mind. There are many possibilities, given that the ... f6-f5 break is prevented for now. One idea was 23 Qb1, keeping the a5-pawn defended, intending 24 dxc6 and 25 b5 or 25 c5, or 24 Rc1 first. Another was 23 Rab1 (or 23 Rec1), planning to open up the queenside with d5xc6, or a5-a6 and then d5xc6. 23 ... Kh8 24 Kh1 cxd5 25 exd5 This was the idea of 23 Rad1, to recapture with the e-pawn and gain a big advantage on the queenside, with the advance c4-c5 in mind. In return White allows ... f6-f5, judging that Black’s kingside counterplay can be contained. The bad position of the black bishop on c7 helps; if it was on g7, defending the long diagonal, White’s plan would have been more than dubious. Having said that, 25 cxd5! looks better, reaching a type of position that became standard in the following century. 25 ... f5 26 Bc1 Rae8 Note that 26 ... e4? was unplayable on account of 27 Bb2 Ne5 28 Nxe5 dxe5 (or 28 ... exd3 29 Nxg6+ Kg8 30 Nxe7+ Qxe7 31 Rxe7 dxc2 32 Rg7+ Kh8 33 Rg6+ and mates) 29 Bxe4 fxe4 30 d6 and wins. But it would have been useful to play 26 ... b6!, increasing Black’s control of the c5-square. 27 Bb2 Ng8 27 ... b6 was still worth considering, both now and on the next few moves, given Löwenthal ‘s slowness in playing c4-c5. 28 Qc3 White builds up the pressure, but it was already possible to play 28 c5!, since 28 ... dxc5? 29 bxc5 Bxa5 fails to 30 Nxe5! Nxe5 31 Rxe5 Rxe5 32 f4. 28 ... Nf6 29 Bb1 Rg8 30 Rd2 Once again, it is not apparent why White delays playing 30 c5!. 30 ... Qh6 31 Nh2 f4

Question: What? Isn’t this a typical mistake in this structure, conceding control of the e4-square? Answer: Logically Morphy would have been aware that he was giving up control of e4, but here that is less serious than usual because of the (albeit temporary) position of the white knight on h2, where it is out of play.

In any case it isn’t clear that there is any better active plan for Black, and he definitely needs to do something to counter White’s progress on the queenside. Now at least he can start to apply some pressure on White’s position and force him to take decisions. In fact Morphy could, and perhaps should, have played this move earlier, on move 29 or 30. Alternatively here, 31 ... b6 was still worth considering. 32 Ne4 Nxe4 33 Bxe4 g5! Now we are definitely in a type of King’s Indian struggle; White is ahead in the attacking race, but there is still a lot of play in the position. 34 f3 Qh4 35 Rf1 Nh6 Morphy abandons for the moment the plan of breaking with ... g5-g4, although it isn’t clear that it would be inferior to the text move. True, after 35 ... h5 36 c5, the position is not yet ready for 36 ... g4? owing to 37 fxg4 hxg4 38 Rxf4! Qe1+ 39 Nf1 Ng5 40 Qg3 Qxg3 41 Nxg3 gxh3 42 Bf5 with advantage to 151

White, because the black king is the one suffering; but Black can prepare it further with 36 ... Rg7!, blocking the ‘X-ray’ on the long diagonal and creating the possibility of 37 ... Reg8 as well. 36 Re2 Nf5

Here the game was adjourned after ten hours’ play. 37 Bxf5 Forced, in view of the threat of 37 ... Ng3+. White loses control of e4 but is now able to play c4-c5 in ideal conditions. 37 ... Bxf5 38 c5 Qh6 The queen must return to the back rank to assist the defence. There is no time to play 38 ... h5, followed by ... Kh7 and ... g5-g4, as White breaks through too quickly after 39 a6!; e.g. 39 ... bxa6? 40 cxd6 Bxd6 41 Qc6 and wins. 39 Rfe1 With the threat of 40 cxd6 Bxd6 41 Rxe5!. 39 ... Rgf8 40 b5 Rc8

Exercise: How do you think White should proceed?

41 Qa3?! “Some of the best analysts in the metropolis came to the conclusion that White should have played 41 c6,” commented Löwenthal, but after 41 ... bxc6 42 dxc6 Rb8 43 Qb4, Black can parry the threat of Bd4 with 43 ... Qg6, so that 44 Bd4 exd4 45 Re7 can be met by 45 ... Rf7. Answer: The move c5-c6 should only be made if there is a concrete tactical justification. It would be better to continue with the idea of undermining the defences of e5 with 41 Qb3, for example, when 41 ... Bxa5? loses to 42 Rxe5! dxe5 43 Rxe5 Kg8 44 d6+ (the point of Qb3) 44 ... Rf7 45 Rxf5 Qg6 46 Ng4! with a decisive attack.

Another way is 41 cxd6 Bxd6 42 Qb3 Rfe8 and then, as well as the immediate 43 Bxe5+ Bxe5 44 Rxe5 Rxe5 45 Qb2!, it’s even better to interpolate 43 b6! a6, when 44 Bxe5+ comes under better 152

circumstances owing to the weakness of b7, while after 43 ... axb6 44 Qxb6, White’s advantage is beyond doubt. 41 ... Kg8 42 b6?! This advance is a good idea, but it fails to achieve much in this particular position; furthermore, White plays it with an incorrect plan in mind. Instead, 42 Qa2! was interesting, with the simple idea of playing 43 cxd6 Bxd6 44 Bxe5, since the white queen would not be under attack from the d6-bishop. 42 ... axb6

Exercise: With which pawn should White recapture?

43 cxb6? A blunder, wasting all his previous good work. It seems that Löwenthal overestimated the possibility of creating a passed pawn on the queenside with a5-a6. Now the pressure on the e5-pawn is reduced and Black achieves the best position he’s had all game. Answer: It was better to weaken Black’s structure with the natural 43 axb6 Bb8 44 cxd6 Bxd6 and now 45 Qb3, although White’s advantage is minimal after 45 ... Rfe8, since Black has the c-file and defends the e5-pawn tactically, due to ... Rc1+ following the exchanges on e5.

43 ... Bd8 44 Rc1 The old recommendation of 44 a6? fails to 44 ... Ra8! 45 axb7 Rxa3 46 b8Q Rb3! and Black wins material. 44 ... Rxc1+ 45 Bxc1 Exercise: What did Morphy play in this position?

Answer: 45 ... Qg6! Simple – this not only supports the infiltration of his light-squared bishop, it also prepares ... h7-h5 again. 46 Qb4 Bd3 Activating the bishop with tempo and preventing the possibility of an eventual sacrifice on a6 to create a passed pawn. 47 Re1 Be7 48 Bb2 Re8 Necessary, to parry the threat of 49 Bxe5. 49 Ng4 h5 50 Nf2

153

Exercise: How did Morphy continue?

Answer: 50 ... g4! Naturally, with this typical King’s Indian pawn break, sacrificing a pawn without hesitation. 51 Qc3?! If White opens the kingside with 51 fxg4 hxg4 52 Nxg4, Black can take the initiative with, for example, 52 ... Bh4 53 Rc1 Bf5, but White would at least have a pawn for his suffering, whereas now Black gets his attack for free. 51 ... Bf5 52 fxg4 It is surprising that Löwenthal never utilized his main idea behind 43 axb6 – the possibility of pushing a5-a6 at some moment. Now that the black bishop has retreated from the a6-f1 diagonal again, 52 a6 was definitely possible, sacrificing the pawn to distract Black’s forces.

52 ... hxg4 53 hxg4 Bxg4 54 Nxg4 Qxg4 55 Rc1

Exercise: And now what?

Answer: 55 ... Kf7! Of course; Black now threatens 56 ... Rh8+ with a mating attack, which forces White to exchange the queens. 56 Qh3 Löwenthal wrote that 56 Qf3 was better, but in that case, rather than exchange queens, Morphy would have played 56 ... Qf5 with the idea of ... Rg8-g3. 56 ... Qxh3+ 57 gxh3 154

Morphy agreed to the exchange of queens because he has acquired two connected passed pawns which, added to his better-placed king, gives him the advantage. Exercise: What did Morphy play now? Answer:

57 ... f3! “Passed pawns must be pushed,” as they wrote many years later. 58 Rf1? This gives Black a free tempo and loses without a struggle, though the position was losing in the long run anyway. Matters are more complicated after Maróczy’s suggestion 58 a6 bxa6 59 b7, but Black still wins with 59 ... Rb8 60 Rc7 Rg8!; for example, 61 Rc1 (or 61 Rc8 f2 and wins, while 61 Rc3 e4 is no better than the game) 61 ... f2 62 Rf1 Bh4 63 Bc1 Rb8 64 Be3 Rxb7 65 Bxf2 Bxf2 66 Rxf2 Kg6 67 Ra2 Rb5 with a winning rook ending. Another defensive idea was 58 Kg1, when the best way to support the pawns seems to be the preparatory move 58 ... Kg6, and if 59 Rc7 then 59 ... e4 60 Bd4 (or 60 a6 e3 wins) 60 ... Bf6 61 Be3 Ra8 and the connected passed pawns win the game for Black. The ending is still very complex, and there are other lines that might offer more resistance than the text move, but it is unlikely that they would have saved the game.

58 ... e4 Black still needs to find a way to terminate White’s resistance, but both his pieces are active and his connected passed pawns are very strong, whereas White’s are harmless. 59 Bd4 Bf6 60 Be3

Exercise: What is Black’s strongest continuation now?

155

Answer: 60 ... Ra8! Condemning White to passive defence, after which Black can continue to improve his position decisively. 61 Bd2 Bd4 62 h4 Kg6 63 Kh2 Rf8 64 Kg3 f2 65 Kg2 e3 66 Be1 Kh5 67 Kg3 fxe1Q+ 0-1 This game lasted twenty hours! Morphy also won the tenth game after Löwenthal blundered in a slightly worse position, bringing the score to 7½-2½; but in the eleventh game Morphy chose the then relatively unexplored Sicilian Defence and landed in an inferior position himself – after 14 moves he was down a piece for two pawns and, despite some ingenious resistance, Löwenthal went on to win. Let’s now look at the twelfth game. Game 21 P.Morphy-J.Löwenthal 12th matchgame, London 1858 French Defence [C01] 1 e4 e6 2 d4 d5 3 exd5 exd5

We know that Morphy preferred an open type of game to a closed one. This Exchange Variation frees the ‘French bishop’ on c8 and leads to a fairly level position; nevertheless, Morphy played it to win, not with the intention of making a draw. 4 Nf3 Nf6 5 Bd3 Be6?! This isn’t the most flexible move – at this stage it’s impossible to be sure whether this is the best square for the bishop. Black usually develops the other bishop first, to e7 or d6, while for those seeking to complicate the games as soon as possible there is 5 ... c5. 6 0-0 Bd6 7 Nc3 c6

Exercise: What do you think Morphy played in this position?

Answer: 8 Ne5 The most ambitious move, occupying a central square with the idea of playing f2-f4. Another idea is 8 Ng5, trying to punish the early development of the bishop to e6, and if the bishop retreats then 9 Re1+. 8 ... Qb6 Exercise: What had Morphy planned against this counter-attack?

Answer: 9 Be3 The house speciality; Morphy has no hesitation in sacrificing a pawn to obtain a lead in development. Today’s analysis engines are initially attracted to an ‘inhuman’ manoeuvre which it is doubtful 156

Morphy would ever have chosen: 9 Re1 0-0 (9 ... Qxd4? is bad on account of 10 Nb5!, which is even better than 10 Nxf7) 10 Na4 Qa5 11 c3 – can it really be good to place the knight on a4? The engines begin to doubt the strength of this original idea after 11 ... Nbd7. 9 ... Nbd7 Rather inconsistent; 9 ... Qxb2 was playable, and if 10 Qe1 Qb6 (10 ... Bb4 11 Bd2 is dubious for Black) 11 Rb1 Qc7, though White would have sufficient compensation after 12 Bg5, intending f2-f4. 10 f4 Supporting the knight, and Black must also take into account the advance f4-f5 at an opportune moment. 10 ... Bxe5! Now White’s f-pawn will never go to f5, and his menacing initiative will lose strength following the exchange of two minor pieces. Instead, 10 ... Nxe5? 11 fxe5 Bxe5 fails to 12 Na4!, winning material; while if 10 ... Qxb2 then 11 Qe1 Bb4 (not now 11 ... Qb6? 12 f5) 12 Bd2 is strong, with the threat of 13 Nxd7 and 14 f5; and 12 Rb1 is attractive too. 11 fxe5 Ng4 12 Qd2 Nxe3 13 Qxe3 Qxb2

Question: Is this a good moment to take the pawn? Answer: Yes. It is less risky now, thanks to the previous exchanges; furthermore, the pawn is some consolation to Black for having less space. Exercise: How did Morphy react to the loss of his b-pawn?

Answer: 14 Ne2 There was no way to exploit the position of the queen on b2 immediately, so Morphy improves the position of his threatened knight, which will now find a good square on f4, from where it will put pressure on the bishop on e6 and might eventually go to h5. White also creates the immediate threat of 15 Rab1 Qa3 (not 15 ... Qxa2?? 16 Ra1 Qb2 17 Rfb1, winning the queen) 16 Rxb7, invading the seventh rank with a big advantage. 14 ... Qa3 15 Nf4 Qe7 Black parries the threat of 16 Nxe6 fxe6 17 Bg6+. This could also have been achieved by playing 15 ... 0-0-0, when White might have replied 16 Rfb1, intending Rb3 and a2-a4, with compensation for the pawn, if no clear advantage. 16 Rab1 0-0-0 Black was probably regretting not having castled on the previous move. Leaving his king in the centre isn’t at all attractive: for instance, 16 ... Nb6?! would allow 17 Nh5! g6 18 Ng7+ (seeking to utilize the f-file, as opposed to 18 Nf6+) 18 ... Kd8 19 a4!, and if 19 ... Nxa4 then 20 Rf6 Nb6 21 Nxe6+ fxe6 22 Rbf1 with a crushing position. 157

Exercise: How did Morphy proceed in this position?

Answer: 17 Be2! Clearing the third rank for the white queen. The target (the black king) is now a fixed one, and White requires the presence of his queen in the attack. Another good idea was 17 a4!, followed by a4-a5, and might be even more accurate as it would prevent Black’s reply (17 ... Nb6? 18 a5 Nc4? now loses to 19 Bxc4 dxc4 20 d5! and Qxa7).

17 ... Nb6 18 Qb3 Rd7 Defending the second rank; 18 ... Kb8 was the standard alternative. 19 Nd3 We are reaching the critical moment. Black’s previous move seemed to encourage this knight manoeuvre, threatening Nc5, but the situation remains rather unclear because it also allows the black knight’s next move. 19 ... Nc4 20 Nc5

Exercise: It’s not hard to see that the black position is a difficult one in view of the weakened position of his king and White’s greater piece activity, but there is a defence. What should Black play here?

20 ... Rc7? Löwenthal fails to offer the most tenacious resistance and his position quickly collapses. Answer: It was essential to play 20 ... Nd2, which looks rather dubious but isn’t easy to refute. If White tries to attack directly with 21 Qa4 then 21 ... Kb8! is both forced and good, since 22 Qxc6? fails to 22 ... Nxb1 23 Rxb1 Ka8!.

Instead, 21 Qg3 (eyeing g7) seems a better try, but after, for example, 21 ... Nxf1 22 Bxf1 b6 (not 22 ... Rc7? 23 Nxb7! Rxb7? 24 Ba6 and wins, nor 22 ... Rg8? for the same reason) 23 Nxd7 Kxd7 24 Qxg7 158

Qf8 25 Qg3, Black is only slightly worse. 21 Qa4 With this attack on the a7-pawn White’s advantage becomes clear. Morphy himself suggested playing 21 Bxc4 first, which is also good though no better than the text. 21 ... b6 After 21 ... b5 22 Qa6+ Kd8 23 Bxc4 dxc4 24 a4, White’s initiative is decisive. From a practical point of view it was perhaps more tenacious to play 21 ... Kb8!? 22 Na6+ Ka8 23 Nxc7 Qxc7, when Black has a pawn for the exchange with hopes of holding on the light squares. 22 Bxc4 Eliminating the most effective defender and opening lines against the black king. 22 ... bxc5 If 22 ... dxc4 then 23 Rxb6! wins. 23 Ba6+ Kd7

Exercise: What is White’s strongest move now?

Answer: 24 Bb7! With a threat that Black overlooks. 24 ... Rd8 This loses quickly, but there was no real improvement. After 24 ... Kd8 25 Bxc6 cxd4 26 Qa5 Kc8, for instance, one way to break in is 27 Qa6+ Kd8 28 Rb8+ Bc8 29 Qa5! Qxe5 (otherwise 30 Qxd5+ etc) 30 Bb7 and there is no defence, since 30 ... Ke7 runs into 31 Re1. Exercise: How did Morphy force Black to resign?

Answer: 25 Bxc6+! 1-0 If 25 ... Rxc6 then 26 Rb7+ Ke8 27 Rxe7+ Kxe7 28 Qxc6 wins. In the thirteenth game Löwenthal had a winning position, with an extra piece, but inexplicably blundered on move 39, overlooking a simple defence involving the capture of a pawn with check, which allowed Morphy to save himself. Let’s now look at the final game of the match. Game 22 P.Morphy-J.Löwenthal 14th matchgame, London 1858 Ruy Lopez [C77] 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 d4

159

This active move is hardly ever played these days. White succeeds in opening the game, but Black obtains adequate resources to neutralize the early offensive. 5 ... exd4 6 e5 Here 6 0-0 has also been tried, when one of the main continuations is 6 ... Be7 7 Re1 0-0 (7 ... b5 is another option, and if 8 Bb3 d6 or 8 e5 Nxe5) 8 e5 Ne8 and Black is able to eliminate the advanced pawn on e5 quickly with ... d7-d6. 6 ... Ne4 7 0-0 Nc5 8 Bxc6 dxc6 9 Nxd4 Question: This pawn structure is the same as in the Exchange Variation, but with White’s e-pawn advanced to e5 – isn’t this a plus? Answer: In the Exchange Variation, White’s superior pawn structure typically compensates for Black’s bishop pair. The 4-3 kingside pawn majority can be important in the endgame, but White’s structure needs to remain flexible; he shouldn’t advance his e-pawn unless this gains a clear advantage.

Exercise: How can Black show that the e4-e5 advance grants White no advantage in this position?

Answer: 9 ... Ne6! Löwenthal plays better than Morphy! A month previously, in T.Barnes-P.Morphy, London 1858, Morphy opted for 9 ... Be7 10 Nc3 0-0 11 Be3 and now the break 11 ... f6, seeking activity on the kingside, but after 12 exf6 Rxf6 13 Qe2 Rg6 14 Kh1 Bd6 15 Rad1 Qh4 16 f4, White had the edge. The fact that Morphy won in the end was not due to his choice of opening.

Thanks to the knight being on c5, rather than g8 (as would normally be the case in the Exchange Variation), Black is able to exchange White’s most active piece and demonstrate that the e5-pawn doesn’t take squares away from Black but in fact concedes them. It is best to do this straight away, since after 9 ... Be7 10 Nc3 Ne6?!, for example, White could play 11 Nf5. 10 Nxe6 160

After 10 Be3 Nxd4 11 Bxd4 Bf5, followed by ... Qd7 and ... 0-0-0, Black has the more comfortable game. 10 ... Bxe6 11 Qe2 Morphy naturally avoids the exchange of queens, seeking middlegame chances based on advancing his kingside majority. 11 ... Bc5 Black rejects the natural 11 ... Be7 in view of 12 Rd1, but 11 ... Qh4 or 11 ... Qd4 (followed by 12 ... Be7) was playable.

12 Nc3

Exercise: What do you think Black should play here?

12 ... Qe7 An inaccurate manoeuvre; Black defends the bishop on c5 and then intends ... h7-h6, defending against the threat of Ne4 followed by Bg5; but it wasn’t necessary to play so cautiously. Answer: Instead, he had 12 ... Qd4!, planning to castle queenside, preventing 13 Ne4 due to 13 ... Qxe5, and not fearing 13 Rd1 either because the exchange of queens that he can offer with 13 ... Qg4 is more than acceptable for Black.

13 Ne4 h6 Exercise: How did Morphy exploit Black’s positional inaccuracy?

Answer: 14 Be3! This eliminates Black’s bishop pair, leaving White with an agile knight ready to support his kingside majority, which is now ready to advance with f2-f4. 14 ... Bxe3 15 Qxe3 Bf5 Understandably, Black didn’t relish 15 ... 0-0 16 f4, when the danger represented by White’s mobile pawn majority is clear; for instance, after 16 ... Rad8 17 f5! Bc4 18 Qg3!, threatening 19 f6. Nor is it possible to seek immediate refuge on the other wing, since 15 ... 0-0-0? loses to 16 Qa7!. However, 15 ... b6!?, to prepare castling, deserves consideration; there are no apparent drawbacks to this idea. Exercise: What did Morphy play in this position?

Answer: 16 Ng3! Of course; he didn’t defend the e5-pawn with 16 f4, since Black would solve his problems easily after 16 ... Bxe4 17 Qxe4 0-0-0. By sacrificing the c2-pawn White gains time to start his pawns rolling forward and retains his knight. Meanwhile the black king remains in the centre. 16 ... Bxc2 17 f4 g6 Castling short isn’t attractive here due to 18 f5. 161

Exercise: How did Morphy continue with his idea, now that Black’s pawn structure has been compromised?

Answer: 18 e6! He exploits the opening of the long dark diagonal, threatening a fork with 19 Qc3. 18 ... Bf5 Although it looks risky, everything indicates that it was better to play 18 ... 0-0!, as the attack with 19 f5 gives White no advantage after 19 ... fxe6!; e.g. 20 fxg6 Bxg6 21 Qxh6 Qg7 22 Qe3 Qxb2. The problem is that the white knight is unable to join in, while the bishop is a good defender. It seems preferable to play 19 Rf2 Bf5 20 Nxf5 gxf5 21 Qb3 Qxe6 22 Qxb7 with a position that remains complicated but not inferior for Black after 22 ... Qd6. 19 Nxf5 gxf5 20 exf7+ Kxf7 21 Qh3 Black has parried the first wave of the attack and has an extra pawn, but White has compensation in the fact that the black king has no comfortable shelter. 21 ... Qf6 22 Rae1 Rhe8

White’s compensation for the pawn is clear, but it isn’t enough for an advantage. Now the struggle begins to see who will be the better at playing his hand. Exercise: How did Morphy maintain the pressure?

Answer: 23 Re5! Activating the rook and gaining space, forcing Black to defend the f5-pawn. Of course 23 ... Rxe5? 24 fxe5 Qxe5? loses to 25 Rxf5+. 162

23 ... Kg6 24 Rfe1 Rxe5 25 Rxe5 Rd8 Question: It seems that Black is on the way to solving his problems. He has already exchanged a pair of rooks and has kept his extra pawn. Isn’t that the case? Answer: You’re not wrong. Black doesn’t stand badly, but don’t forget that he hasn’t yet solved the problem of his king’s safety and this creates practical difficulties. Exercise: What manoeuvre did Morphy find to improve his position?

Answer: 26 Qg3+ It is also essential to attend to one’s own defences. With this and the following move White brings his own king into safety, away from any unforeseen checks. 26 ... Kh7 27 h3! Rd7 Black defends his second rank against the imminent 28 Qe3, but resorting to passive defence is a dangerous ploy. Instead, after 27 ... Rd5!? 28 Re8 Qg7 29 Qxg7+ (White gains nothing from 29 Qh4 due to 29 ... Rd1+ 30 Kh2 Rd2 31 Re7 Rxg2+ with perpetual check) 29 ... Kxg7 30 Re7+ Kf6 31 Rxc7 Rb5, Black should hold the rook endgame with relative ease. 28 Qe3 b6

Exercise: What did Morphy, that paragon of wonderful combinations, play in this position?

Answer: 29 Kh2! The wonderful combinations were the product of a positional advantage. Here the priority is secure the king before entering complications. 29 ... c5 30 Qe2 As well as attacking the a6-pawn, this defends White’s b-pawn, thus granting his rook more freedom. It also contains the idea of playing Qc2, followed by g2-g4, though that would need to be analysed very carefully. 30 ... Qg6?! Here 30 ... a5 was more precise, maintaining the queen on the long diagonal to have ... Qd4 available if the white rook leaves e5.

163

Exercise: How did Morphy respond?

Answer: 31 Re6! Now this move is strong. If 31 ... Qf7 then 32 Qe5!, threatening 33 Re8, or possibly 33 Rf6. 31 ... Qg7? This looks like the losing move. Black needed to find 31 ... Qg8!, keeping an eye on the white rook, while if 32 Qe5 then 32 ... Rf7! and Black is not much worse. Exercise: What is the complement to White’s previous move?

Answer: 32 Qh5! The pressure on h6 and f5 is becoming more and more uncomfortable for Black. 32 Qxa6? Qxb2 would be completely wrong. 32 ... Rd5

Exercise: How did Morphy make progress here?

Answer: 33 b3! A modest move of crushing force. Now Black cannot move any of his pieces and once he has exhausted his pawn moves he will be in zugzwang. 33 ... b5 164

After 33 ... c4 34 bxc4 Ra5, White can exploit the absence of the black rook with 35 Qe2!, threatening 36 Re7, against which there is no satisfactory defence. Or if 33 ... a5 then 34 a4 with zugzwang. 34 Rxa6 Rd6 35 Qxf5+ The queen endgame after 35 Rxd6 cxd6 36 Qxf5+ was also winning. 35 ... Qg6 36 Qxg6+ Kxg6

37 Ra5 The most practical solution. The pawn endgame after 37 Rxd6+ cxd6 is winning too, but requires more calculation. In contrast, White’s task now is very simple. 37 ... Rb6 This passive move isn’t the most tenacious, but the more active defence 37 ... c6 fails to hold the position either after 38 a4 bxa4 39 Rxa4, as White is effectively two pawns up. 38 g4 There is no defence against the advance of White’s connected passed pawns on the kingside. 38 ... c6 39 Kg3 h5 40 Ra7 hxg4 41 hxg4 Kf6 No better is 41 ... c4 42 bxc4 bxc4 43 Ra4 c3 44 Rc4 and wins. 42 f5 Ke5 43 Re7+ Kd6 44 f6 Rb8 45 g5 Rf8 46 Kf4 c4 47 bxc4 bxc4 48 Kf5 c3 49 Re3 1-0 And so the match ended 10-4 in Morphy’s favour (+9, =2, -3). “I am convinced that I was vanquished by superior strength,” was Löwenthal’s comment on it all.

165

Chapter Five The Journey to Paris and the Match against Harrwitz Since it had proved impossible to arrange to a match with Staunton in London, Morphy travelled to Paris to challenge the strongest French players. In informal games one of his opponents was Jules Arnous de Rivière, one of the best French players from the 1850s to the 1870s. Morphy beat him 7½-1½. More testing was the match with Daniel Harrwitz, preceded by a casual game that Morphy lost through a serious blunder as early as the tenth move. The 5th of September saw the start of their formal match, which was scheduled to last until one or other player had scored seven wins. Harrwitz demonstrated that he was a worthy opponent by winning the first game, after a long struggle that finished in a rook endgame. In the second Morphy stood better but, playing in a manner that was most unusual for him in serious games, he burnt his bridges, unjustifiably rushing a pawn break that left his structure damaged. Although another long struggle ensued, Morphy ended up with another loss, so that Harrwitz now led 2-0. Let’s now look at the third game. Game 23 D.Harrwitz-P.Morphy 3rd matchgame, Paris 1858 Dutch Defence [A85] 1 d4 f5 2 c4 e6 3 Nc3 Nf6 4 Bg5 Bb4

In the fifth and seventh games of the match Morphy opted for 4 ... Be7 (see Game 25). 5 Qb3 c5 6 d5?! Although this advance gains space and hinders ... Nc6, White’s centre loses mobility. It was better to keep the tension with 6 Nf3 or 6 e3. 6 ... e5?!

Morphy also commits his central pawn structure prematurely. The natural 6 ... 0-0 was better. Question: So what you’re saying is that Morphy responded with an inaccuracy similar to that of his opponent. What’s the reason? Answer: One (inadequate) explanation might be that Morphy was already planning to exchange his dark-squared bishop and then arrange his pawns on dark squares. But it will become clear that the price is a high one. Black’s position loses flexibility and he soon suffers from having less space.

7 e3 0-0 8 Bd3 d6 9 Nge2 h6 10 Bxf6 Qxf6 11 a3 Bxc3+ Simply retreating with 11 ... Ba5 was to be considered, but it seems that Morphy was resolved to exchange this bishop. 12 Qxc3 Nd7 13 0-0 Qg6 Making way for the knight to come to f6. Morphy is obviously not attracted by the prospect of an equal endgame arising from 13 ... e4 and prefers to seek active chances on the kingside, for which he needs to retain his queen. 14 b4 b6 15 f3 h5 Consistent with 13 ... Qg6; the pawn can later be used as a battering ram. The alternative was 15 ... 166

Nf6, but Morphy doesn’t want the knight to lose control of c5, in order to discourage his opponent from playing b4xc5. 16 Bc2!

An annoying move – the bishop has a good post on a4. 16 ... Bb7?! A surprising positional error from a player of Morphy’s strength. Question: It’s a developing move that connects the rooks; what’s dubious or bad about it? Answer: White’s dream in various positions of the Queen’s Indian Defence is to play d4-d5 and neutralize the bishop on b7. Here, with that advance already made, Black plays ... Bb7, ‘biting on granite’, as the saying goes.

For comparison’s sake, after 1 d4 Nf6 2 c4 e6 3 Nf3 b6 4 a3 Bb7 5 Nc3, the routine move 5 ... Be7?! is considered inaccurate (5 ... d5 is usual) because White can then play 6 d5!, limiting the mobility of the b7-bishop, and in this case Black still has his pawn on c7 and can play ... c7-c6 at an opportune moment. Returning to our game, either 16 ... Nf6 or 16 ... h4 was preferable. 17 Ba4 Qf7 18 Bxd7?! Question: But ... what’s this? Why exchange the active bishop for the passive knight? Answer: You’re right to ask. Harrwitz has an idea that in other circumstances might be a good one, but which is premature here: he wants to open the b-file quickly, thinking that he will be better placed to exploit it. Naturally, he doesn’t want to play 18 bxc5? at once in view of 18 ... Nxc5. Instead of this exchange, White could play 18 Bc6!, followed by doubling rooks on the b-file, ready to open it under more favourable conditions. There is no hurry, because Black’s counterplay on the kingside has barely started. Meanwhile, he can hardly accept the offer to exchange his ‘bad’ bishop, as White would then gain a strong passed pawn on c6.

18 ... Qxd7 Even though Black has been left with the passive bishop, the exchanges have gone some way towards helping him solve his problems of space and mobility. 19 bxc5?! Consistent with his previous move, but weak. White will not manage to exploit the opening of the bfile here. It was better to occupy the b-file with the rooks first and only open it at the most favourable moment. It was preferable therefore to leave b-pawn alone for the time being and play 19 f4 straight away. 19 ... bxc5 20 f4 e4 The pawns are on light squares, so Black must look for some way to activate his bishop. Question: Why push the e-pawn at all? It is then stuck on a light square! Why not just play 20 ... Rae8, for instance? Answer: True, the pawn structure is not so nice for the bishop, and Morphy could instead play 20 ... Rae8, as you suggest; but he didn’t want to be tied to the defence of the e-pawn, and had seen a way for his bishop to obtain some activity in any case.

21 Rab1 Ba6 167

Here we can see another negative consequence of 19 bxc5?! – the passive bishop on b7 is transformed into an active one on a6, putting pressure on c4. But it is understandable if White was not too worried by this attack from the lonely bishop. Exercise: The position has evened out; although it wasn’t part of his plan when he made his 18th and 19th moves, White must now think about defence. What should he play here?

22 Rfc1?! White is thinking only of his own plans and is not taking his opponent’s into consideration. Exercise: How can Black demonstrate that White’s last move was inaccurate?

Answer: 22 ... Qa4! This prevents White’s planned 23 Qa5 and demonstrates that the bishop on a6 is not so lonely after all. Black is now putting pressure on the pawns at c4 and a3. Answer: Going back to the Exercise before last, 22 Qc2! would have been better.

23 Ng3 Harrwitz improves the position of his knight, which is now heading for d2 to support the c4-pawn. The position is no more than slightly unpleasant for White. The computer finds an admirable defensive resource in 23 Rb3!?, when the immediate and logical 23 ... Rab8 can be answered by 24 Rxb8! Rxb8 25 Ng3 Rb3 (to be able play ... g7-g6 without allowing Qf6) 26 Qe1 g6 27 Nxh5! gxh5 28 Qh4 and White’s counterplay seems sufficient for equality after 28 ... Qd7 29 Qxh5 with two pawns for the piece and threats of perpetual check. 23 ... h4 24 Nf1 Rab8 25 Nd2

168

Exercise: How do you think Morphy tried to make progress here?

Answer: 25 ... Rb6! Preparing to double rooks with 26 ... Rfb8. 26 Rxb6 axb6 27 Qb3? A bad decision – in the endgame there are only two results possible. It was necessary to seek counterplay on the kingside, trying to exploit Black’s weakened structure (as in the line beginning with 23 Rb3), here based on a timely g2-g4; for example, 27 Rb1 b5 28 g4 hxg3 29 hxg3 Rb8 30 g4 or indeed 27 g4! at once. 27 ... Qxb3 28 Nxb3 b5 The position is opening up in favour of the bishop, while the protected passed c-pawn that is now created is far more dangerous than White’s passed a-pawn. 29 cxb5 Since “all rook endings are drawn”, White might try to hold the inferior position arising from 29 Nd2 bxc4 30 Nxc4 Bxc4 31 Rxc4 Rb8, but Rubinstein would no doubt have won it with Black – and Morphy too. After 32 Rc1 Rb3 33 Ra1, one way would be 33 ... Rxe3 34 a4 Rb3 35 a5 Rb7 36 a6 Ra7, planning to bring the king to b6 while relying on the passed pawns at e4 and c5 to restrain any white counterplay; another is 33 ... c4 34 a4 c3 35 Rc1 (forced, in view of the threat of 35 ... c2 and 36 ... Rb1) 35 ... Ra3 36 a5 Kf7 37 a6 Rxa6 38 Rxc3 Ra5. 29 ... Bxb5 30 Na5

Exercise: Black needs to combine attack with the defence of the d6-pawn. How did Morphy accomplish this?

Answer: 30 ... Ra8! First the rook heads for a6 to defend d6, while keeping an eye on the a3-pawn; then Black intends to bring his king to e7 to release the rook for attack. 31 Nb7 The only active way to play, but now the knight is almost trapped. 31 ... Ra6 32 Rc3 Kf8 33 Nd8 Bd7 34 Rb3 Ke7 35 Rb8

169

Exercise: White’s pieces are tied up and his monarch is unable to assist the defence. How can Black exploit the situation?

Answer: 35 ... c4! Given the factors mentioned above, Black’s passed pawn increases in strength. 36 Kf2 c3 37 Ke2 Rxa3 38 Nc6+ 38 Kd1 fails to 38 ... Ba4+ 39 Ke2 (or 39 Kc1 Ra1+, winning a piece) 39 ... c2 and the pawn queens. 38 ... Bxc6 39 dxc6 c2 40 Kd2

Exercise (easy): What is the quickest way for Black to wrap up the game?

Answer: 40 ... Rc3! Black will emerge with two extra pawns, including one on the seventh rank. 41 Kc1 Rxc6 42 Rb3 Kf6 43 Ra3 Exercise: There are several good ways to continue; what do you think was Morphy’s choice?

Answer: 43 ... g5! Opening up the kingside to be able to invade with his king. 44 g3 hxg3 45 hxg3 gxf4 46 gxf4 Kg6 47 Ra5 Rc5 170

Morphy guarantees that his king will be able to invade White’s kingside without losing the f5-pawn. The rest is quite simple. 48 Ra6 Rc3 49 Rxd6+ Kh5 50 Rd2 Kg4 51 Rg2+ Kf3 52 Rg5 Rc5 53 Rh5 Kxe3 54 Rh4 Kf3 0-1 Let’s now look at the fourth game. Game 24 P.Morphy-D.Harrwitz 4th matchgame, Paris 1858 Philidor’s Defence [C62] 1 e4 e5 2 Nf3 d6 3 d4 exd4 4 Qxd4 Nc6

In the sixth game Harrwitz switched to 4 ... Nf6. In this game, as we shall see, he had other plans for the f6-square. 5 Bb5 Bd7 6 Bxc6 Bxc6 7 Bg5

A move that seeks to create discomfort in the black position, an ‘agent provocateur’ as Lasker described it. In later years 7 Nc3 became the most popular move here. 7 ... f6?! Question: Well, it appears that 7 Bg5 has done its job; this looks like a concession to me. Answer: Yes, provoking the weakening 7 ... f6 is an achievement for White. But this isn’t forced, nor is it the best in the position.

In the second game of the match Harrwitz opted for 7 ... Nf6 and later won the game, even though he stood worse from the opening: 8 Nc3 (this position can also be reached via 7 Nc3 Nf6 8 Bg5) 8 ... Be7 9 0-0-0 0-0 10 Rhe1 h6 11 Bh4 Ne8?! 12 Bxe7 Qxe7, and now White is slightly better after the prophylactic 13 Kb1 or the more active 13 g4, as suggested by Knaak. Instead, Morphy surprisingly played the hasty 13 e5?!, allowing 13 ... Bxf3. 171

The ‘problem’ with the provocative move 7 Bg5 is that it can be answered by the apparently erroneous 7 ... Be7!, since after 8 Qxg7 Bf6 9 Qxh8 Bxh8 10 Bxd8 Bxb2 11 Bxc7 Bxa1 12 Bxd6 Bxe4, the resulting endgame isn’t advantageous to White. 8 Bh4 Nh6 9 Nc3

9 ... Qd7 In P.Morphy-H.Baucher, Paris (blindfold simul) 1858, Black opted for 9 ... Be7 – see Supplementary Game 24.1, which has various points in common. 10 0-0 Be7 11 Rad1 With this move, centralizing the rook to its most useful square, with ‘X-ray’ pressure against the black queen, White’s ‘semi-forced’ moves come to end – by which we mean those moves obviously need to be played and there is apparently nothing better.

11 ... 0-0 12 Qc4+ This move, on the other hand, marks the start of a concrete plan. 12 ... Rf7 A move that starts an ambitious plan of kingside expansion, which needs to be justified tactically because of the menacing presence of the rook on d1. Yet it is difficult to agree that this is what the position ‘demands’ or, more accurately, permits. The obvious 12 ... Kh8 was better.

Exercise: Did Morphy continue with his idea behind 12 Qc4+, or is there some tactical operation that is stronger and justifies a deviation from his plan? Answer: No, Morphy plays consistently to exploit the trump presented to him by Black’s seventh move; i.e. the e6-square.

13 Nd4! It was tempting to fall in with Black’s provocation and play 13 e5. Then taking the pawn loses the exchange, and otherwise White threatens to advance it with 14 e6, so it looks strong and perhaps it is, objectively. Nevertheless, Black can muddy the waters with 13 ... Qg4!, when White must choose whether or not to enter an endgame – either way taking on f6 or d6 gives him nothing special, and the 172

position after the advance e5-e6 is harder to evaluate, either with the exchange of queens or after interpolating 14 Rd4 Qg6. In the latter case, the position is not clearly advantageous for White after, for instance, 15 e6 Rff8 (threatening to take on f3 and follow up with ... c7-c6 and ... d6-d5) 16 Nd5 Bxd5 17 Rxd5 c6, again followed by ... d6-d5. Morphy preferred the cleanest continuation. 13 ... Ng4 14 h3 Ne5 15 Qe2 Black’s ‘knight on the rim’ has managed to centralize itself and even drive the opposing queen away, reducing White’s control of e6. The knight does remain exposed to a timely f2-f4, among other ideas, but when Black played 12 ... Rf7 he already had a specific idea in mind.

15 ... g5? Question: Is this really so bad? In several positions in the Modern Benoni this move, fortifying the knight on e5, is perfectly normal. Answer: Yes, in general the idea is worth considering, and here there is already a ‘hook’ for the advancing pawn on h3, but the cost is too high – the f5-square is seriously weakened, especially considering that White already has a knight on d4 ready to jump in.

16 Bg3 Rg7 17 Nf5 First, White plays the obvious knight move. 17 ... Rg6

Exercise: And now what? How do you think Morphy continued?

Answer: 18 f4! Question: Is this really such a good move? I can appreciate that it opens the f-file in White’s favour, but I also see that the knight on e5 is now completely secure and Black has the use of the half-open g-file. Besides, surely White had a promising alternative in 18 Nd5 - ? Answer: It’s true that 18 Nd5! was good: after 18 ... Bxd5?! 19 Rxd5 Bf8 20 c4, White dominates the light squares and has the initiative, while the pride and joy of Black’s position, the centralized knight, can be eliminated at an opportune moment with Bxe5, leaving White with a strong knight against a bishop limited in scope.

The text seeks a more active game. “To get squares, you have to give squares,” as Fischer used to say. In return for becoming active on the f-file White concedes certain advantages to Black, but they are very different in value – the white rook, aided by the dominating f5-knight, will be the main beneficiary of the opening of the kingside and Black’s king will be the one that comes under fire. 18 ... gxf4 19 Rxf4 Kh8

173

Black wants to double rooks on ‘his’ half-open file. Exercise: How did Morphy show that it is White who benefits more from the opening of the f- and g- files?

Answer: 20 Rh4! Emphasizing the weakness of the black king. White prevents 20 ... Rag8?? as that now runs into 21 Rxh7+! Kxh7 22 Qh5+ Rh6 23 Qxh6 mate; here we see the decisive assistance rendered by the knight on f5. Once again 20 Nd5 was worth considering, but Morphy prefers to exploit the new elements in the position. 20 ... Bf8 Exercise: How did Morphy make progress in this position?

Answer: 21 Bxe5! The most active enemy piece is eliminated and Black is forced to recapture with the f-pawn, leaving his dark-squared bishop imprisoned. 21 Nd5 was again a good alternative. 21 ... fxe5 22 Rf1 Qe6 Still trying to make use of the g-file, although the black pieces remain uncoordinated. 23 Nb5!? An interesting move, probing the black defences. Must Black defend or can he try to go onto the offensive? 23 ... Qg8?! This was Black’s original idea, but it’s an unequal struggle after this. It was better to go back with 23 ... Qd7 and challenge White to find a way to make progress. He might just retreat his knight to c3 and try something else; if White plays 24 c4, Black would at least gain some practical chances after 24 ... Bxb5 25 cxb5 c6, aiming for ... d7-d5 when appropriate.

174

Exercise: How did Morphy continue his offensive?

Answer: 24 Rf2! Combining attack with prophylaxis: overprotecting g2 and maintaining the threat against the c7-pawn. Morphy wants more than the positional advantage obtained from 24 Nxc7 Rc8 25 Nd5 Bxd5 26 exd5 Qxd5 27 c4. With the second rank defended he threatens 25 Nxc7 followed by retreating with 26 Nb5, if there is nothing better. 24 ... a6 Naturally, ... Bxb5 would mean complete surrender of the light squares and would only be playable if the tactics justified it – for instance, if the shot ... Bc5 were available. In this case it doesn’t seem to work; e.g. after 24 ... d5 25 exd5 Bxb5 26 Qxb5 Rd8, White has 27 d6! (eschewing 27 c4?! Rb6!, when ... Bc5 is suddenly a real possibility) 27 ... Bxd6 (or 27 ... cxd6 28 Qxb7) 28 Nh6, winning material. 25 Nxc7! Rc8 The knight’s retreat to b5 has been prevented, but we shall soon see another important virtue of the prophylactic 24 Rf2!. 26 Nd5 Bxd5 27 exd5 Rc7 Resigning himself to the loss of a pawn. Exercise: How can 27 ... Qxd5? be refuted? Answer: With 28 Rxh7+! Kxh7 29 Qh5+ Bh6 30 Ne7 (White can play this because the f2-rook prevents a check on the g1-a7 diagonal) 30 ... Qe6 31 Nxg6 Kg7 (31 ... Qxg6 loses to 32 Rf7+) 32 Nh4 with an extra pawn and the attack.

28 c4 Be7 29 Rh5 Qe8?

175

Exercise: Black must have felt depressed at the prospect of having passively to await his execution – with 29 ... Bf8, for example – but the move he chose has a clear refutation. What is it?

Answer: 30 c5!! White’s material and dynamic superiority permits an elegant and rapid demolition of Black’s position. 30 ... Rxc5 If 30 ... dxc5 then 31 Qxe5+ and Qxc7, while 30 ... Rd7 loses to, among other things, 31 Rxh7+ Kxh7 32 Qh5+ Kg8 33 Nh6+ Kg7 (or 33 ... Kh8 34 Rf7) 34 Rf7+ Qxf7 35 Nxf7 Kxf7 36 Qf5+, picking up a rook.

Exercise: What was the idea behind the pawn sacrifice, luring the black rook away from c7?

Answer: 31 Rxh7+! Now the black queen will be overloaded with the defence of both the g6-rook and e7-bishop, so White is able to gain a decisive material advantage. 31 ... Kxh7 32 Qh5+ Kg8 33 Nxe7+ Kg7 34 Nf5+ Kg8 35 Nxd6 1-0 Supplementary Game 24.1 P.Morphy-H.Baucher Blindfold simultaneous, Paris 1858 1 e4 e5 2 Nf3 d6 3 d4 exd4 4 Qxd4 Nc6 5 Bb5 Bd7 6 Bxc6 Bxc6 7 Bg5 f6 8 Bh4 Nh6 9 Nc3 Be7

Deviating from Harrwitz’s 9 ... Qd7. 10 0-0 0-0 11 Qc4+ Kh8 12 Nd4

176

The course of the game is similar to what we’ve already seen, but at this point Black has better options. Exercise: What should Black play?

Answer: 12 ... Qd7?! In contrast to the main game above, in which the queen was already committed to d7, here Black could defend the e6-square with 12 ... Bd7, as in J.Löwenthal-D.Harrwitz, London (25th matchgame) 1853, which continued 13 Rad1 c6 14 Ne6 Bxe6 15 Qxe6 Nf7 with only a slight advantage to White. 13 Rad1 Rf7? A strange decision; it’s the same idea as in the main game, but in worse circumstances. It was better to play 13 ... Rae8 or 13 ... Nf7. 14 f4 Intending f4-f5, to support the knight which will head for e6. 14 ... a5 15 f5 Rff8 16 Ne6 Rg8 17 a4 White has a free hand here. Steinitz suggested 17 Rf3 or 17 Nd5 as improvements, but the text doesn’t worsen White’s position.

17 ... Ng4 18 Qe2 Ne5

Exercise: Based on what we saw in the main game above, how do you think Morphy continued here?

Answer: 19 Bg3 Just as before, preparing to eliminate the opponent’s best piece at the right moment. 19 ... Qc8 Black makes way for the c6-bishop to retreat to d7 and challenge the knight. He is not afraid of Bxe5, because he plans to recapture with the d-pawn, releasing his dark-squared bishop. Exercise: What did Morphy play now?

Answer: 20 Bxe5! The activity of the e7-bishop will prove irrelevant. It will soon be apparent that it would have been better to play 19 ... Nf7, planning a timely ... Nd8. 20 ... dxe5

177

Exercise: What is the best way to exploit the strong knight outpost and the lack of coordination among Black’s pieces?

Answer: 21 Rf3! The start of a blitzkrieg. The black king is almost unprotected and the white pieces can rush rapidly into the attack. 21 ... Bd7? This loses, but there was hardly anything better. 22 Rh3! Threatening a quick mate with 23 Rxh7+ or 23 Qh5. 22 ... h6 Exercise: There are many winning moves here – which is the fastest?

Answer: 23 Qd2! Of course, with a double attack on h6 and d7. 23 ... Kh7 24 Qxd7 Bd6

White has an extra piece and a winning position, and Morphy finds the quickest finish. 25 Rxh6+! Kxh6 26 Rd3! Kh5 27 Qf7+ 1-0 It is mate in two moves; e.g. 27 ... Kh4 28 Rh3+ Kg4 29 Qg6 mate, or 27 ... g6 28 Qh7+ Kg4 29 h3 mate. Let’s now look at the fifth game of the match. 178

Game 25 D.Harrwitz-P.Morphy 5th matchgame, Paris 1858 Dutch Defence [A85] 1 d4 f5 2 c4 e6 3 Nc3 Nf6 4 Bg5 Be7 This time Morphy prefers to neutralize the pin on his knight, instead of counter-pinning with 4 ... Bb4 as he previously played as Black (see Game 23).

5 e3 0-0 6 Bd3 Question: Hmm, didn’t I hear you say that it’s better to develop knights before bishops? Answer: Yes, but it isn’t an immutable rule. The reason is that generally the knights have very few options on their first move, whereas the bishops have several. However, that’s not the case here. Harrwitz develops his bishop first to preserve the option of developing the knight to f3 or e2.

6 ... b6 Morphy seeks an active diagonal for his bishop, rather than developing it on d7. Alternatively, 6 ... c5 is playable, but it only postpones the decision about the light-squared bishop. It is hard to believe that Morphy would ever choose a Stonewall set-up with 6 ... d5. 7 Nge2 It is difficult to say whether this is preferable to 7 Nf3. 7 ... Bb7 8 0-0

8 ... Nh5 Question: This isn’t a developing move and isn’t typical of Morphy. How would you explain it? Answer: Morphy prefers to reduce his space problems by exchanging dark-squared bishops before developing his last minor piece.

The alternative 8 ... Ne4?! runs into 9 Bxe4 fxe4 10 Bxe7 Qxe7 11 Ng3 and Black has problems supporting the e4-pawn, since 11 ... d5? can be answered by 12 cxd5 exd5 13 Qb3!, threatening both Nxd5 and Nxe4. Black would have to resort to something like 11 ... Qb4, although after 12 Ncxe4 Qxc4 13 Rc1! Qxa2 14 Rxc7, his position looks very fragile. It wasn’t advisable to play 8 ... d6? either in view of 9 Nf4 Qd7 10 d5!, highlighting the newly created hole at e6 (here we see one of the virtues of 7 Nge2). Instead, 8 ... Nc6 is playable, but it’s not obvious how to follow this up; the black pieces have little mobility and are rather uncoordinated. Harrwitz himself was convinced that Morphy’s ... Nh5 manoeuvre was a good one, and in the seventh game of the match, he opted for 8 Bxf6, as we shall see in Supplementary Game 25.1. 9 Bxe7 Qxe7 10 Ng3!? An interesting decision; White accepts doubled g-pawns so as to be able play g3-g4 later. 10 ... Nxg3 11 hxg3 d6 Planning the mobilize the knight at last by ... Nd7-f6. 12 f4 An ambitious move, keeping both e3-e4 and g3-g4 in reserve. 179

12 ... Nc6 Now 12 ... Nd7 would allow 13 e4, whereas the text restrains that advance. An alternative way was 12 ... c5. 13 g4 This move attracted criticism, but it doesn’t appear to be a bad one. 13 ... Nb4 14 gxf5?! This exchange, on the other hand, is definitely hasty, as it opens the e-file in Black’s favour. It was better to retreat the bishop first, with 14 Be2 or 14 Bb1. 14 ... exf5 15 Qd2 Rae8 16 Rae1

Exercise: What should Black play here?

16 ... Qh4 A questionable decision; although White’s kingside is rather weak, Black can’t exploit this, since he doesn’t have enough pieces lined up for an attack. Answer: Instead, Black would stand well after 16 ... Nxd3! 17 Qxd3 Be4.

17 Bb1 Re6 18 Qf2! Exercise: Why not drive the knight away with 18 a3 - ? Answer: Because Black’s activity, while containable, deserves respect. He would have the strong reply 18 ... Bxg2! 19 Qxg2 Rg6 20 axb4 (not 20 Qxg6 hxg6 21 axb4?? Qg3+ 22 Kh1 Kf7 and mates) 20 ... Rxg2+ 21 Kxg2 Rf6 and Black’s initiative will result in a win of material.

18 ... Qh5

Exercise: How can White restrain Black’s initiative? 180

Answer: 19 d5! Similarly to game three of the match, the b7-bishop is now shut out of play. 19 ... Rh6 20 Qf3 Qh4 After 20 ... Qh2+ 21 Kf2, the white king is in no danger in the centre, and White now has the possibility of a timely Rh1. 21 a3 Na6 22 b4 Since Black’s two minor pieces have been shut out of play, the weakness of the opposing kingside declines in importance, since White can cope with the activity of Black’s queen and rook. But this is only a temporary advantage – if either of Black’s knight or bishop becomes active the situation will become more complicated. 22 ... Nb8 23 Ne2 Nd7 Heading for the kingside without delay. 23 ... c5, controlling d4, was also worth considering. 24 Ng3?! This isn’t the best square for the knight. It was possible to play 24 Nd4 g6 25 Ne6, but after 25 ... Rc8, followed by 26 ... Nf6, or even 26 ... b5 or 26 ... c6, the knight finds itself rather isolated on e6. It seems better to solidify the kingside first with 24 Qg3!. The exchange of queens would leave Black without any play, but after 24 ... Qh5 25 Nd4 Rg6 26 Qf3 Rg4 27 Kf2!, White has the superior position, since his pieces are placed harmoniously, which cannot be said of Black’s forces. 24 ... g6 25 Kf2 Nf6 26 Rh1 This was White’s idea, aiming to exchange rooks and thus eliminate almost all the danger to his king. 26 ... Ng4+ 27 Kg1 Qf6 28 Rxh6 Nxh6 29 Qd1

29 ... Ng4 Morphy didn’t want to abandon the idea of attacking until it became inevitable. Thus he rejected 29 ... Qc3 or 29 ... Qb2, which would almost force White to play 30 Qc1. Then 30 ... Qxc1 31 Rxc1 Ng4 leads to an ending which is slightly better for Black: 32 e4 isn’t possible because the f4-pawn falls, while after 32 Re1, it’s possible to try to activate the passive b7-bishop with 32 ... b5. 30 Qd2 Qh4 Black pursues the dream of attack. Another possibility was 30 ... Bc8, planning to recycle his bishop via d7. 31 Nf1 Re8 32 g3 Qh3 The queen fails to achieve anything positive on this menacing-looking square, since it isn’t possible to bring more pieces to its support. It’s therefore logical to recommend 32 ... Qf6 instead. 33 b5?! White wants to prevent ... b6-b5 forever, but misses the opportunity to play 33 e4! advantageously, exploiting the absence of the black queen from the centre and queenside. 33 ... Nf6 Black retreats the knight to hinder e3-e4, accepting that the chances of a successful kingside attack are minimal. 181

34 Qg2 Qxg2+ Morphy had a plan prepared for the exchange of queens. Besides, it was no longer appropriate to keep the queens on, because White would be able to prepare e3-e4 under favourable conditions again. 35 Kxg2

Exercise: What did Morphy have planned here?

Answer: 35 ... a6! Preparing to open the a-file as an invasion route at an opportune moment. 36 a4 axb5 A debatable move. Question: What? Wasn’t this the idea behind 35 ... a6 - ? Answer: Indeed it was. The point is that the opening of the a-file couldn’t be prevented (b5xa6 by White would damage his structure), so Black could first improve the position of his pieces, ready to open the a-file under better conditions.

One idea was 36 ... Nd7, followed by ... Nc5, ... Ra8, and soon after take on b5. The plan should also include improving the passive bishop before opening the a-file. As we shall see, 36 ... Kf7 would be another useful move. 37 axb5 Ra8 38 Nd2 Ra3 The position is balanced; Black’s rook is the more active, but White is able to cope with this and his space advantage is an important factor, as we shall see. 39 e4 White eliminates the weakness on e3 and opens the e-file in his favour. On the other hand, the black bishop can now view the future with greater optimism. 39 ... fxe4 40 Nxe4 Nxe4 41 Bxe4 The white bishop has also gained in activity. 41 Rxe4 would be less convincing, as after 41 ... Kf7, controlling the rook’s entry point, Black could play ... Bc8 with a slightly more pleasant position. 41 ... Rc3 Black could have interpolated 41 ... Kf7, taking prophylaxis against any invasion of the white rook on the e-file, but he would have no advantage after 42 Bf3 h5 (to prevent g3-g4) 43 f5!.

182

Exercise (easy): How best to defend the c4-pawn?

Answer: 42 Bf3! Indirectly, with the threat of 43 Re8+ and 44 Rb8. 42 ... Kf7 43 Re4 Here 43 Be2! was more precise, so that, after 43 ... Bc8, he can keep the black bishop becoming active by playing 44 g4. If Black tries to prevent this in turn with 43 ... h5?, then 44 Ra1! is awkward to meet. 43 ... Bc8 Exercise: How should White complete his defensive manoeuvre?

Answer: 44 Be2? A surprising error – why allow the black bishop to become active? After 44 g4!, Black would find it much harder to make progress. He could try 44 ... h5!?, but then 45 gxh5 Bf5 46 Re2 (better than 46 hxg6+ Kxg6 47 Rd4 Rd3!) 46 ... Rxc4 47 Kg3 (intending 48 h6) 47 ... Kg7 48 Re7+ Kh6 49 Bg4! looks quite drawish; e.g. 49 ... Bxg4 50 Kxg4 gxh5+ 51 Kf5. 44 ... Bf5 Morphy has finally gained the advantage. 45 Rd4

Exercise (easy): How did Morphy improve his position further? 183

Answer: 45 ... h5! Securing the position of his bishop and preparing to open the kingside with ... h5-h4 at the right moment. 46 Kf2 Kf6 47 Rd2 Exercise: What did Morphy play now?

Answer: 47 ... Bc2! It is not only necessary to think about your own plans, you also need to hinder those of the opponent. This move prevents 48 Ra2 and clears the f5-square for Black’s king. 48 Ke1 After 48 Rd4, Black would have played 48 ... Kf5. 48 ... Be4 Not now 48 ... Kf5?? due to 49 Rxc2!; but 48 ... Bb3 was also strong. 49 Kf2 Kf5 50 Ra2

Exercise: How did Morphy take a decisive step forward here?

Answer: 50 ... h4! Black forces an entrance for his king to make the decisive invasion. 51 gxh4 Kxf4 52 Ra7 Going round the back with 52 Ra8 was equally hopeless; for example, 52 ... Rh3 53 Rf8+ Bf5 54 Bf1 Rxh4 and White can’t even pick off the c7-pawn, since 55 Rf7 Rh2+ 56 Bg2 g5 (with intending 57 ... Rh7) 57 Rxc7 loses to 57 ... Be4.

184

Exercise (easy): What is the quickest way to win?

Answer: 52 ... Rh3 This not only wins the h4-pawn (making the black g-pawn passed), it also corrals the White’s king, which means he again has no time to take on c7, as we quickly see. 53 Rxc7? Rh2+ 54 Ke1 Ke3 0-1 Since if 55 Bf1 then 55 ... Ra2 forces mate. Supplementary Game 25.1 D.Harrwitz-P.Morphy 7th matchgame, Paris 1858 1 d4 f5 2 c4 e6 3 Nc3 Nf6 4 Bg5 Be7 5 e3 0-0 6 Bd3 b6 7 Nge2 Bb7 8 Bxf6

This time Harrwitz prefers to eliminate the black knight, instead of exchanging bishops after 8 ... Nh5, as we saw in the previous game. 8 ... Bxf6 9 0-0 Qe7 10 Qd2 d6 11 f4

Echoes of the previous game; Harrwitz wants to play e3-e4. The immediate 11 e4 wasn’t convincing in view of 11 ... f4!. 11 ... c5! This prevents 12 e4 and is a reminder that, following the exchange with 8 Bxf6, Black is the one better equipped to dominate the dark squares. 12 d5 Na6 13 dxe6 Qxe6 14 Rae1 Bh4 Seeking to provoke a weakness with 15 g3 or else pin the knight after Ng3. But given that White’s 12 d5 has opened the long diagonal, it was possibly better to leave the bishop on f6 and play 14 ... Nb4 15 Bb1 Rae8, when the black pieces are ideally placed; note that 16 a3 Nc6 17 Nd5? would be a mistake, 185

on account of 17 ... Na5!. 15 Ng3 Qg6?! Once again 15 ... Nb4 16 Bb1 was interesting and now, for instance, 16 ... Rad8 17 a3 Nc6 18 Nd5 Ne7 with a good position.

Exercise: There is a snag with Black’s last move – what is it?

16 Nd5? This isn’t it. Answer: The problem with Morphy’s move was that White could have played 16 e4!. Then after, for example, 16 ... Bxg3 17 hxg3 Qxg3 18 exf5 Nb4 19 Bb1, intending Re7, the white pieces are suddenly ideally placed. Let’s look further: 19 ... Rae8 20 a3 Rxe1 21 Rxe1 Nc6 (21 ... Bxg2 loses piece to 22 Qf2!) 22 Ne4, when the d6-pawn is lost and the black position collapses.

16 ... Bxd5! 17 cxd5 Now we’ve reached a sort of Benoni position, and one in which Black has nothing to complain about. 17 ... Bxg3 18 hxg3 Nc7 The knight could have reached a more central post after the pawn sacrifice 18 ... c4!? 19 Bxc4 Nc5, when it is evidently more active, controlling e4, but Morphy didn’t believe it was necessary.

19 Kf2 This move is also reminiscent of the main game. White hopes to create counterplay on the h-file. 19 ... Rae8 20 Rh1 Re7 21 Rh4 Qf7 22 Be2 The alternative was the immediate 22 g4, with complex play after 22 ... Qxd5 23 Qc2 b5.

Exercise: How did Morphy improve his position?

Answer: 22 ... Ne8! 186

The knight goes to f6, from where it attacks the vital squares g4, e4 and d5. 23 Qd3 Nf6 24 Bf3 g6 25 Re2?! Rfe8 Methodical play, putting maximum pressure on the weakness at e3 before taking any other measures. Löwenthal suggestion of 25 ... g5! 26 fxg5 Ne4+ 27 Kg1 Nxg3 28 Re1 Qg7 was also good, or if 28 Rc2 then 28 ... Qe8 with strong pressure.

26 b3 Qg7 Once again 26 ... g5! was possible. 27 Rh1

Exercise: Now that the black queen has left the f-file it appears that ... g6-g5 is no longer on Morphy’s agenda – or is it? What did he play in this position?

Answer: 27 ... h6! Yes, ... g6-g5 is still the plan, but he wants to keep his pawns together. 28 Kg1 g5 The total triumph of Black’s strategy – White’s counterplay on the h-file has vanished and all the black pieces are active. 29 fxg5 hxg5 30 Bh5 Here the simple 30 ... Rf8 is possible with a big advantage, but Morphy demonstrated that it wasn’t the only move.

Exercise: What did he play?

Answer: 30 ... Ne4! Activating his knight and attacking the g3-pawn, while 31 Bxe8? Qa1+ leads to mate. 187

31 Re1 Rf8 And the g3-pawn is indefensible. Black has a winning position. 32 Bf3 Nxg3 33 Rh3 Qe5 34 Rh6 g4 35 Bd1 Kg7 36 Rh4 Exercise: There are several good moves here – what do you think was Morphy’s choice?

Answer: 36 ... Rh8! Now that the h-file is open, Black exchanges the white defender in order to gain control of the file. 37 Rxh8 Kxh8 38 Bc2 Rh7 39 Qd2 Qb2 One of various ways to win; the text threatens 40 ... Rh1+ 41 Kf2 Ne4+. 40 Rd1 Rh1+ 41 Kf2

41 ... Rf1+?? A surprising blunder; Black sees a ‘phantom mate’. He could have played 41 ... Rxd1 42 Qxd1 Ne4+ 43 Kg1 and now there many good moves; for instance, 43 ... Qc3 (but not 43 ... Nc3? 44 Qe1 and White threatens perpetual check) 44 Bd3 Kg7 45 Kf1 Qe5 with a technical win – Black is a pawn up and in total control of the position. 42 Kxg3 Not 42 Rxf1?? Ne4+. 42 ... Qe5+ 43 Kh4 Qf6+ Löwenthal suggested that Morphy might, in his calculations at move 41, have overlooked that after 43 ... Kg7 44 Rxf1 Qh2+ 45 Kg5 Qh6+, the pawn on f5 is no longer defended, so there is no mate and White can play 46 Kxf5 which wins. 44 Kg3 Qe5+ ½-½ By winning the fifth, Morphy took a 3-2 lead in the match. Let’s see the sixth game. Game 26 P.Morphy-D.Harrwitz 6th matchgame, Paris 1858 Philidor’s Defence [C41] 1 e4 e5 2 Nf3 d6 3 d4 exd4 4 Qxd4 Nf6

Varying from 4 ... Nc6, as he played in the fourth game of the match (see Game 24) and which was answered by 5 Bb5. In Supplementary Game 26.1, we shall take a brief look at an example where Black pre-empted the Bb5 pin by playing 4 ... Bd7 first. 5 e5

188

Question: How unusual! It’s not very often that we see Morphy exchanging queens when there are good alternatives available, particularly so early in the game. Answer: You’re right, it is surprising, but it appears that Morphy considered that the resulting endgame offered chances to outplay his opponent.

5 ... dxe5 Steinitz criticized this exchange and suggested 5 ... Qe7 instead, mentioning 6 Be3 (as had been played in D.Harrwitz-J.Löwenthal, 18th matchgame, London 1853), after which 6 ... Ng4 7 exd6 Qxd6 was correct. Many years later, in V.Jansa-E.Ermenkov, Prague 1985, White found an improvement in 6 Be2 dxe5 7 Nxe5 Nbd7 8 Nd3!, when continuing to develop with 8 ... g6 didn’t appeal to Black, probably in view of 9 0-0 (rather than 9 Bf4, which can be met by 9 ... c5!) 9 ... Bg7 10 Re1; so he went for the endgame with 8 ... Qe4, but White achieved a much more pleasant position after 9 Qxe4+ Nxe4 10 0-0 Bd6 11 Bf3 Nec5 12 Re1+ Kd8 13 Nxc5 Nxc5 14 Na3!, heading for b5 or c4. 6 Qxd8+ Kxd8 7 Nxe5 Be6 8 Nc3 Bd6?! Black doesn’t seem to attach any importance to the bishop pair. This move almost forces the retreat of the knight to c4, where it threatens to create a weakness on d6. Simply 8 ... Nbd7 was sounder.

9 Nc4 Bxc4 Having played 8 ... Bd6, Black is logically reluctant to spend another tempo playing 9 ... Bb4; e.g. 10 Bd2 Nbd7 11 0-0-0 and White’s game is more harmonious – but this is objectively better than giving up the light-squared bishop voluntarily. 10 Bxc4 Re8+ 11 Be3 Ke7 Black could exchange the bishop on e3 by playing 11 ... Ng4 12 0-0-0! Nxe3 13 fxe3, but he would remain slightly worse as he is quite a bit behind in 189 development. For example, after 13 ... Rxe3 14 Bxf7

Nd7 15 Rhf1, the inactive rook on a8 is a problem and 15 ... Bxh2? is bad on account of 16 Nd5 Re4 17 Rh1 Rh4 18 Rhe1 and wins. 12 0-0-0 a6 Black is forced to ‘waste’ another move before retreating, since if 12 ... Kf8?! then 13 Nb5.

Exercise: How did Morphy continue?

Answer: 13 Bg5! White had two pleasant options. Morphy wanted more than he could obtain from 13 Nd5+ Nxd5 14 Bxd5, when Black has to acquiesce to the sad 14 ... Nc6 with advantage to White, who might play 15 c3, maintaining Bxc6 as a threat. 13 ... Nbd7 14 Ne4 h6 With the king in the centre it isn’t advisable to embark on tactical complications, such as 14 ... Be5 15 f4! Bxf4+ 16 Bxf4 Nxe4 17 Rhe1 f5 and now 18 Bd5 or 18 g4, when Black’s position collapses. Exercise: Once again there are two attractive options – what do you think was Morphy’s choice?

Answer: 15 Bxf6+ Morphy chooses the cleanest way to a clear advantage, in the style of Capablanca, “clearing all the dead wood from the position”. He prefers to simplify and give up one of his bishops than preserve the bishop pair with 15 Nxd6 cxd6 16 Bh4, which is favourable too. 15 ... Nxf6 16 Nxd6 cxd6

190

Exercise: There now follows a series of moves, after which the advantage Morphy has foreseen will become apparent. How did he begin?

Answer: 17 Rhe1+ First, he continues to simplify. The black king is forced to leave the centre, since f7 must be defended. 17 ... Kf8 18 Rxe8+! Nxe8 Exercise: The knight has also been driven to a passive position in order to defend the d6-pawn. How did Morphy continue now?

Answer: 19 Bd5! Forcing the black rook into passive defence of the b7-pawn. 19 ... Rb8 20 Bf3! Black must lose at least two tempi to remove the pressure on d6 and b7, and the superiority of White’s rook and bishop over Black’s rook and knight is obvious. Furthermore White has only two ‘pawn islands’ to Black’s three. 20 ... g6

Exercise: Although White has the advantage, there is no immediate way to convert this. How can he make further progress?

Answer: 21 c3! The moment has arrived to bring into play the only white piece which isn’t superior to its black counterpart – the king.

21 ... Ke7 22 Re1+! Whenever possible, Morphy never wastes an opportunity to worsen the position of the opponent’s pieces. Now if 22 ... Kd7?!, White replies 23 Bd5, when Black would need to be very brave to play 23 ... f6 (23 ... f5? loses the pawn after 24 Bf7 g5 25 Be6+) 24 Be6+ Ke7 (otherwise 25 Bf7 wins a pawn) 25 f4 etc. 22 ... Kf8 23 Kc2 Nc7 24 Kb3 Another possibility was to preface this with 24 b4. 24 ... Ne6 25 Ka4 b6 Here 25 ... Nc5+ would have slowed down White’s progress, without really changing the situation; e.g. after 26 Ka3 b6 27 Rd1 Ke7 28 b4. 26 b4 h5 27 h4 Ke7 28 Re3! The rook seeks activity along the only rank available, the third, where in addition to overprotecting the c3-pawn it can consider going to f3, tying Black down. 28 ... Kd7 29 Bd5 Re8 30 Kb3 Rg8 31 Rf3 Ke7 32 a4 a5?! 191

33 Bxe6!? Question: A surprising exchange! It’s a pity to give up this bishop, isn’t it? Answer: Yes, Harrwitz was probably just as surprised as you (and me). His last move weakened his queenside, but he had calculated that it wasn’t possible to exploit this immediately with 33 Kc4 Rc8+ 34 Kb5?! on account of 34 ... Nc7+ 35 Kc6 Nxd5+ 36 Kxd5 axb4 37 cxb4 Rc2 with good counterplay. So Morphy gave up his powerful bishop for the knight in order to be able to activate his king.

However, it isn’t clear that this was best. Morphy could have continued to manoeuvre with, say, 33 Re3, followed by g2-g3, retaining the advantage, albeit with a long struggle in prospect. 33 ... Kxe6? Now White’s plan triumphs. Recapturing with 33 ... fxe6! was better, and although White still has the advantage, there is no clear win. After 34 Kc4, it would be unwise for Black to start a race with 34 ... g5? 35 Kb5 gxh4 36 Kxb6 axb4 37 cxb4 Rb8+ 38 Ka5 Ra8+ 39 Kb5 Rb8+ 40 Kc4, since White’s connected passed pawns are too strong. Instead, he should activate his king with 34 ... d5+! 35 Kb5 Kd6!, when 36 Kxb6 axb4 37 cxb4 Rb8+ gives White no advantage, while after 36 Rf7 axb4! 37 axb4 Rc8! 38 Kxb6 d4, the active rook is of little help to White, and Black’s d-pawn is also dangerous; e.g. 39 b5 d3 40 Rf3 Rc3!. After the text, the black king is tied to the defence of f7 for the moment and is unable to become as active. 34 Kc4 g5 35 Kb5! axb4 36 cxb4 gxh4 37 Kxb6 Rb8+ 38 Ka5 Kd5

Black is able to hinder the advance of White’s queenside pawns, but he has already had to give up the b6-pawn and his own passed d-pawn presents no danger here; in fact it will soon be surrendered to the enemy. 39 Rd3+! The advance 39 b5 was also effective. 39 ... Kc4 40 Rxd6 Rxb4 41 Rd4+! 192

41 Rb6 was equally decisive; in either event the queening race is too favourable for White. 41 ... Kxd4 42 Kxb4 f5 43 f4 Another way was 43 a5 f4 44 a6 (or just 44 f3) 44 ... f3 45 gxf3 h3 46 a7 h2 47 a8Q and wins, since if 47 ... h1Q then 48 Qe4 mate. 43 ... Ke3 44 a5 Kf2 45 a6 Kxg2 46 a7 h3 47 a8Q+ Kg1 48 Qf3 1-0 The extra black h-pawn prevents stalemate after 48 ... h2 49 Qg3+ Kh1 50 Qf2! and 51 Qf1 mate. Supplementary Game 26.1 P.Morphy-J.Löwenthal 8th matchgame, London 1858 1 e4 e5 2 Nf3 d6 3 d4 exd4 4 Qxd4 Bd7 Instead, P.Morphy-A.Mongredien, Paris (8th matchgame) 1859, saw 4 ... a6 5 Bg5 f6?!.

This is inferior but is harder to exploit than it might appear. After 6 Be3 Be6 7 Nc3 Ne7 8 Be2 Nec6! (planning to develop the other knight to d7) 9 Qd2 Be7 10 0-0 0-0 11 Rad1 Nd7, White’s advantage was only slight. Morphy decided to exchange a pair of knights with 12 Nd4 Nxd4 13 Bxd4, and now instead of waiting with 13 ... Re8 or 13 ... c6, say, Mongredien took the unjustified risk of opening lines against his own king with 13 ... f5?. Play continued 14 exf5 Bxf5 15 Bc4+ Kh8 16 Nd5 Nf6 17 Nxe7 (17 Qg5! looks more testing) 17 ... Qxe7 18 Rfe1 Qd7 19 Qg5 h6 20 Qh4 Rae8 21 c3 and having defended quite well up to this point, Black suddenly blundered with 21 ... Re4?? 22 Rxe4 Nxe4 (22 ... Bxe4 23 Bxf6 wins a piece) 23 Qxh6+, when White won quickly. 5 Be3 As we shall see, 5 Bf4, to control e5, makes sense here; 5 Bg5 has also frequently been played. 5 ... Nf6 Morphy also had this position with Black. A casual game S.Boden-P.Morphy, London 1858, saw 5 ... Nc6 6 Qd2 Nf6 7 Bd3 Be7 8 Nc3 0-0 9 0-0 h6 (9 ... Ne5 has been suggested, exchanging a piece to relieve Black’s cramped position) 10 h3 Nh7 11 g4 h5 12 Nh2 hxg4 13 hxg4 Ne5 14 f3 g5 15 Kg2 and here, after a series of already risky moves, Morphy committed a glaring error, playing 15 ... c5? (instead of 15 ... c6), after which Black’s light-squares were a permanent problem (even if it was a blunder in the ending that eventually cost him the game). This was Morphy’s only defeat against Samuel Boden in eleven games; four others were drawn, and Morphy won all the rest. 6 Nc3 Be7 7 Bc4 Nc6 8 Qd2 Ne5 Both here, and in Morphy’s game as Black above, we can appreciate the value of the suggestion 5 Bf4. 9 Nxe5 dxe5 10 0-0 0-0

193

White has a slight edge here; his rooks can deploy first on the open central file (e.g. with 11 f3, followed by Rad1), but Morphy was not satisfied with this and decided to sharpen the struggle. 11 f4!? Morphy consents to having the inferior structure after ... e5xf4, in return for open lines and the chance of greater piece activity. 11 ... Bd6 Löwenthal declines the challenge, preferring to maintain his hold on the e5-square before focusing on the weakness of White’s e4-pawn. 12 f5 Morphy has not managed to open the position but has gained some space instead. 12 ... Bc6 13 Qe2! h6 Not 13 ... Nxe4? which loses to 14 Nxe4 Bxe4 15 Qg4 Bc6 16 f6 g6 17 Qh4 etc. The alternative capture 13 ... Bxe4 also fails to inspire confidence; one violent possibility is 14 g4 Bc6 15 g5 Nd7 16 Qh5 with ample compensation for the pawn. 14 Rad1 Qe7 Now after 14 ... Bxe4 15 Nxe4 Nxe4, White has 16 Bxh6, among other things. 15 Bd5! Morphy seeks to exploit the new situation and threatens double Black’s pawns. 15 ... Bxd5?! This keeps Black’s pawn structure intact, but it isn’t really a solution. Black acquiesces to being left with his ‘bad bishop’, which, combined with superior development and a space advantage, gives White the advantage. 16 Nxd5 Nxd5 17 Rxd5 f6?!

18 Qg4? Morphy himself considered this move to be premature, as the threat of 19 Bxh6 can easily be parried. 194

He suggested 18 Kh1 instead; and 18 Rfd1 Rad8 19 Qd3, followed by c2-c4, looks strong too. 18 ... c6! 19 Rd3 Bc5! With the exchange of bishops, Black solves his problems. 20 Qg3 Rad8 21 Rfd1 Rxd3 22 Rxd3 Rd8! 23 Bxc5

Contrary to appearances, Black is able to challenge for the d-file, since 23 Rxd8+? Qxd8 24 Bxc5 fails to 24 ... Qd1+ 25 Kf2 Qxc2+ etc. 23 ... Qxc5+ 24 Qf2 Qxf2+ 25 Kxf2 Rxd3 26 cxd3

The endgame is equal. Morphy continued to play for a win, but without breaking Löwenthal’s resistance; in the end it was Morphy who made the decisive error (51 Kg1??) and resigned eight moves later. We left the match against Daniel Harrwitz with the score standing at 4½-2½ to Morphy, having seen in Supplementary Game 25.1 how Morphy let a win slip through his fingers in the seventh game of the match. In the eighth Morphy won almost without a struggle after Harrwitz played an inferior opening; this game will be quoted in the notes to Game 27 below. So the score was now 5½-2½ to Morphy. The winner was supposed to be the first player to score seven wins, but at this stage Harrwitz reported that he was unwell and considered the match to be over. Apart from the first two games, Morphy’s superiority had been absolute. Morphy now sought fresh challenges and invited Adolf Anderssen to travel from Breslau to Paris for a match. Anderssen accepted, but told Morphy that his teaching duties prevented him from arriving before the Christmas holidays, three months away. Morphy was prepared to wait and meanwhile gave simultaneous displays, some conducted blindfold, and played informal games against Arnous de Rivière and Saint-Amant. His most famous game during that waiting period is the one played at the Paris Opera. Game 27 P.Morphy-Duke Karl of Brunswick & Count Isouard Paris 1858 Philidor’s Defence [C41] 1 e4 e5 2 Nf3 d6 3 d4 Bg4?

195

“Nowadays every schoolboy knows that this is bad, but in those days it was even played by Harrwitz!,” commented Kasparov. Fischer wrote: “Later Steinitz said you shouldn’t move out your bishops before you bring out your knights, a very good rule for beginners.” This is because the bishops generally have more options than the knights.

Based on that rule, 3 ... Nf6 is effectively the main line, while 3 ... Nd7 is also possible, and even 3 ... exd4 is better than the move in the game. Exercise: How can White show that Black’s last move is inaccurate?

Answer: 4 dxe5 Now to avoid losing a pawn, Black is forced to exchange his bishop, which is his only developed piece, leaving him behind in development already. 4 ... Bxf3 5 Qxf3!? Question: What? You mark this move only as ‘!?’. You can’t seriously be considering taking on f3 with the pawn, accepting doubled pawns, not when Qxf3 is available, surely? Answer: I understand your surprise, but matters are actually not so clear in this case. The first World Champion, Wilhelm Steinitz, suggested 5 gxf3, based on the fact that, after 5 ... dxe5 6 Qxd8+ Kxd8, White has 7 f4 and the resulting endgame is better for White, thanks to his bishop pair and with the position opening up. In support of his assessment he gave this sample line (not forced): 7 ... Nf6? 8 fxe5 Nxe4 9 Bg2 Nc5 10 b4, followed by 11 Bxb7 and wins.

5 ... dxe5 6 Bc4 Nf6? This is a blunder, overlooking White’s strong reply. It was better to defend f7 with the queen, but this has its drawbacks too. For instance, 6 ... Qd7 leaves the queen on the open d-file, so White can play simply 7 0-0, or else 7 Qb3!?, followed by Nc3; either way a rook will soon land on d1 with a clear advantage. Note in the latter line that an attempt to disrupt White’s smooth development with 7 ... c6 8 Nc3 b5 would be met by 9 Nxb5! cxb5 10 Bxb5 Nc6 11 Be3, when White has very strong play for the piece, with ideas of Rd1-d3-c3.

When Fischer gave a simul in 1970 in Sarajevo, two of his games saw 6 ... Qf6 7 Qb3 b6 8 Nc3.

Here 8 ... Ne7? loses to 9 Nb5 Na6 10 Qa4 Nc5 11 Nd6+! Kd8 12 Qe8 mate – which, as Kasparov noted, “was pointed out long ago in Greco’s treatise!” Fischer’s opponents naturally preferred 8 ... c6, controlling d5. Exercise: How do you think Fischer continued in both games? Answer: There are several ways forward; one possibility is 9 Nd5! exd5 (or 9 ... Qd8 10 Nxb6) 10 Bb5+! Nc6 11 Bg5! Qg6 12 exd5 etc. Fischer opted instead for 9 Bg5!, based on the variation 9 ... Qxg5 10 Bxf7+ and 11 Bxg8, with an extra pawn and a winning position. His opponents both replied with 9 ... Qg6, and after 10 Rd1 (10 0-0-0 isn’t playable, because the bishop can then be captured with check) 10 ... Be7 (if 10 ... Nd7, White’s superior development permits the combination 11 Bxf7+! Qxf7 12 Qxf7+ Kxf7 13 Rxd7+, again with an extra pawn and a winning endgame; and 11 Nb5! cxb5 12 Bxb5 looks even stronger) 11 Bxe7 Nxe7.

196

Exercise: How can White’s dynamic advantage be highlighted this time? Answer: By the combination 12 Bxf7+! Qxf7 13 Rd8+ Kxd8 14 Qxf7 and wins.

Exercise (easy): What was the strong move that Black had overlooked?

Answer: 7 Qb3 With a double attack on f7 and b7. Black’s position is already on the edge of collapse. 7 ... Qe7 Seeing that 7 ... Qd7 8 Qxb7 wins easily, Black seeks some relief by exchanging queens with ... Qb4+. In the eighth game of the Morphy-Harrwitz match, Paris 1858, Black played 7 ... Bd6, resigning himself to allowing 8 Bxf7+, which prevented a ‘miniature’ but not his ultimate defeat. Exercise: There are several attractive continuations here. What do you think Morphy selected?

Answer: 8 Nc3 The ‘house speciality’, development before anything else. As we mentioned, Black was intending 8 Qxb7 Qb4+ etc, which of course isn’t bad for White, who is just a pawn up. Another option is to throw in 8 Bxf7+, when 8 ... Kd8? 9 Qxb7 Qb4+ 10 Qxb4 Bxb4+ 11 c3 reaches the same endgame with White now two pawns up, while 8 ... Qxf7 9 Qxb7 wins the exchange – except 197

that 9 ... Bc5! makes things quite messy here; for example, 10 Qc8+ (10 Qxa8?! is worse after 10 ... 0-0, intending ... c7-c6 and ... Qc7 or ... Ng4) 10 ... Ke7 11 Qxh8 Bxf2+! etc. White may still be winning but, from the practical point of view, it’s hardly worth choosing this continuation over the others. As Lasker wrote about 8 Bxf7+: “That would have been a butcher’s method, not an artist’s.” 8 ... c6 9 Bg5 “Now Black is in a zugzwang position here. He can’t develop his [b8] knight because his pawn [on b7] is hanging, the bishop is blocked by the queen,” commented Fischer. 9 ... b5? Too optimistic; Black’s lag in development doesn’t permit such a weakening move. However, it was no better to play 9 ... Na6 10 Bxa6 bxa6 11 Qc4, nor 9 ... h6 10 Bxf6 gxf6 11 0-0-0; in both cases White’s moves are simple – all he has to do is keep developing. Steinitz’s suggestion of 9 ... Qc7 doesn’t help either as, after 10 0-0-0,

Black has nothing better than 10 ... b5 anyway, which leads to an inferior position; e.g. 11 Bxf6 gxf6 (not 11 ... bxc4? due to 12 Qb7!! and wins) 12 Nd5! Bh6+ 13 Kb1 cxd5 14 Bxb5+ Kf8 and now either 15 Qxd5 or 15 Rxd5!? gives White a big advantage. Exercise: If Black tries 10 ... Bc5 in this line, what would be the refutation? Answer: White has the same tactic as in Fischer’s simultaneous games: 11 Bxf7+! Qxf7 12 Rd8+ and wins.

Exercise: What did Morphy play here?

Answer: 10 Nxb5! White’s lead in development permits and even demands sacrifices. In contrast, the passive 10 Be2? 198

allows Black to get back on his feet with 10 ... Qb4! again. 10 ... cxb5 11 Bxb5+ Nbd7 White can answer 11 ... Kd8 with simply 12 0-0-0+, followed by Rd3, or else 12 Bxf6 gxf6 13 Qd5+ Kc8 14 0-0-0! (much better than 14 Qxa8 Qb4+). 12 0-0-0 With the simple threat of 13 Bxf6, followed by taking on d7, or indeed 13 Bxd7+ straight away, exploiting the fact that both black knights are pinned. 12 ... Rd8 Unfortunately, 12 ... 0-0-0 loses to 13 Ba6+ and Qb7 mate.

Exercise: How did Morphy increase his dynamic advantage?

Answer: 13 Rxd7! Preparing to bring the only missing piece into play, the king’s rook, and already planning the final combination. Morphy’s handling of the final stage of this game attracted nothing but praise: “White lands successive blows, and each time with gain of tempo.” (Euwe) “Morphy is in his element. The brilliant combination with sacrifices makes this game one of the most beautiful achievements in the entire history of chess.” (Maróczy) 13 ... Rxd7 14 Rd1 Qe6 There is no time for 14 ... Qb4? in view of 15 Bxf6, threatening 16 Bxd7 mate. Exercise: Out of several winning continuations, what do you think was Morphy’s choice?

Answer: 15 Bxd7+! “Morphy was looking for a brilliancy. 15 Qxe6+ fxe6 16 Bxf6 is an easy winning ending.” (Fischer) “15 Bxf6 was a drier alternative” (Kasparov). 15 ... Nxd7

199

Exercise: What was the key to Morphy’s idea?

Answer: 16 Qb8+!! Do you recall the move 12 Qb7!! in the note to Black’s ninth move? 16 ... Nxb8 17 Rd8 mate “A sparkling finish!” (Euwe)

200

Chapter Six The Match Against Anderssen December arrived at last, and with it the match against Anderssen. This contest between the two best players of the time was also to be decided by the first player to score seven wins. The first game was played on the 20th of December 1858. Just as in his two previous serious matches, against Löwenthal and Harrwitz, Morphy began with a defeat. The second game ended in a draw, as we shall see in Supplementary Game 29.1. Once again the match had not begun well for Morphy, but just as in the previous two matches, once Morphy started to win games he was unstoppable. He won no less than five games running, including several of high quality. Let’s look at game three of the match. Game 28 P.Morphy-A.Anderssen 3rd matchgame, Paris 1858 Ruy Lopez [C65] 1 e4 e5 2 Nf3 Nc6 3 Bb5 Nf6 4 d4

This is one of only two recorded games where Morphy faced the Berlin Defence, which has been all the rage since the year 2000, following its successful use by Kramnik in his World Championship match against Kasparov in London. The main debate has focused on the variation 4 0-0 Nxe4 5 d4 Nd6 6 Bxc6 dxc6 7 dxe5 Nf5 8 Qxd8+ Kxd8, reaching the ‘Berlin Endgame’, which is so complex that Kasparov described it as “a queenless middlegame”. 4 ... Nxd4 This reply has found few followers; 4 ... exd4 is normal. 5 Nxd4 exd4

Exercise: How did Morphy meet Anderssen’s idea?

Answer: 6 e5 This advance creates more problems than the quiet 6 Qxd4. Nearly a century and a half later the zwischenzug 5 ... c6 was tried a few times, the point being that after 6 Bc4 exd4 7 Qxd4 Qb6, Black has avoided Morphy’s e4-e5 idea, since 7 e5? would fail to 7 ... Qa5+ followed by 8 ... Qxe5. 6 ... c6 7 0-0 Question: There are several options here; why did Morphy choose this move? Answer: It can’t be said that this was a strange choice for Morphy, since it speeds up his development and brings his king to safety, which is something that Black cannot imitate easily.

Alternatively, 7 Qxd4 was playable, when it’s advisable for Black to play 7 ... Nd5 rather than enter an inferior endgame after 7 ... cxb5?! 8 exf6 Qxf6 9201Qxf6 gxf6 10 Nc3, as the extra black pawn does not

compensate for his horrible structure. On the other hand, 7 exf6 Qa5+! is less convincing and provides some justification for preferring 7 00. 7 ... cxb5 Anderssen had faith in this move and returned to it in several future games. It was even played by Steinitz. Given the course of the game, though, it looks more prudent to play 7 ... Nd5 again.

Exercise: What did Morphy choose in this position?

Answer: 8 Bg5! Rapid development looks like the most appropriate recipe, as the most direct continuation offers nothing clear. After 8 exf6 Qxf6 9 Re1+ Be7 10 Qe2, White can prevent Black from castling immediately, but with his own poor development he has no way to increase the pressure. As well as the optimistic (if correct) 10 ... a6, Black can return the pawns with 10 ... d5! 11 Qxb5+ Qc6 12 Qb4?! Be6 13 Qxd4 0-0, obtaining the superior position. 8 ... Be7 8 ... h6?? isn’t possible due to 9 exf6 hxg5 10 Re1+. 9 exf6 Bxf6 10 Bxf6 Qxf6 11 Re1+ Kf8

Some sources say that this position was reached via 10 Re1+ Kf8 11 Bxf6 Qxf6. There isn’t any vital difference between these two move orders, since 11 ... Kd8 is worse than placing the king on f8. Exercise: What do you think Morphy played in this position?

Answer: 202

12 c3 An unsurprising decision, isn’t it? Morphy sacrifices a pawn to speed up his development. Morphy commented in A.C.M. 1859: “Up to this move the game coincides with one played by Lange in his Schach-partieen.” In that game White played 12 Na3 a6 13 Qe2, but after 13 ... g6, followed by ... d7-d5 and/or ... Kg7, Black doesn’t stand badly. Morphy’s move enables his knight to reach a much better square than a3. Both Anderssen and Steinitz faced 12 Qe2. A.Carstanjen-A.Anderssen, Cologne 1859, continued 12 ... Qe6 13 Qd2 Qb6 14 c3 d5 15 cxd4, reaching a position similar to the game (after 14 Nc3), with a bad bishop vs. a good knight. Here it would also be unwise to underestimate the weakness of the dark squares, as Anderssen did: 15 ... Be6 16 Nc3 h5 17 a4 b4 18 Na2 a5 19 Nc1 Rh6 20 Nb3 Rg6 21 Nc5 Kg8 22 Re5 Bh3 23 g3 and now after 23 ... f5? 24 Rae1, Black’s position collapsed in a few moves. In another game against the same opponent Anderssen played 16 ... h6 instead, followed by 17 ... g6 and 18 ... Kg7, and achieved a better result. Steinitz solved his problems in a less weakening way. First of all he secured the position of his king by playing 12 ... g6 at once; then after 13 Nd2 d6 14 Qxb5 a6 15 Qd5 Bf5 16 Ne4 Bxe4 17 Rxe4 Kg7 18 g3 Rac8, the game was equal and soon drawn, H.Bird-W.Steinitz, London (14th matchgame) 1866. 12 ... d5 Taking the pawn is very risky: after 12 ... dxc3 13 Nxc3 d6? 14 Nd5 (threatening Nc7) 14 ... Qd8 15 Rc1, Black can’t cope with White’s advantage in development and activity; e.g. 15 ... Be6 16 Rxe6! fxe6 17 Qf3+ Kg8 18 Rc7 and wins. 13 cxd4 Be6 14 Nc3 a6 15 Re5

White’s compensation for the pawn is adequate; he has the superior development and a firm grip on the dark squares. 15 ... Rd8 Understandably Anderssen is reluctant to let go of his extra pawn and resign himself to an inferior, albeit tenable position after 15 ... g6 16 Nxd5 Bxd5 17 Rxd5 Kg7. 16 Qb3 Morphy continues to mount pressure on the d5-pawn. His queen move makes way for the a1-rook to come to d1, while creating the possibility of playing a timely a2-a4, demonstrating that Black’s 15 ... Rd8 left his queenside slightly weak. 16 ... Qe7 This defends b7 against the threat of a2-a4, although the situation of the black king remains a problem. If 16 ... g6 here, White could play 17 a4 bxa4 18 Qxb7 Kg7 19 Qxa6, when he has regained the pawn, remaining with the more pleasant position. 17 Rae1 g5? A move reminiscent of the 23 ... f5? in the Carstanjen-Anderssen game mentioned above. Black weakens his kingside without obtaining anything in return; we can safely say that Morphy would never have played this. Instead, White would have some advantage after 17 ... g6 18 Ne2 Kg7 19 Nf4 (if this knight manoeuvre was what Anderssen wanted to prevent, he has paid too high a price by making other 203

concessions) 19 ... Rhe8 20 Qg3, followed perhaps by h2-h4-h5, but Black’s position would be sounder than in the game.

Exercise: What do you think Morphy played here?

Answer: 18 Qd1 This immediate transfer of the queen to the kingside was criticized, even though it’s a reasonable move, as it was considered more precise to activate the knight first. There are two ways to do this: one is via the queenside with 18 a4! bxa4 19 Nxa4 Rg8 20 Nc5, which offers a slight advantage; but 18 Ne2! is more energetic, heading for g3; for example, 18 ... Rg8 19 Ng3 (threatening Nf5, followed by Qa3+) 19 ... Qd6 20 Qe3 and White’s superiority is clear. 18 ... Qf6 19 R1e3 Creating an immediate threat which Black fails to spot. Another option was to prepare the f2-f4 break with 19 Ne2, followed by Qd2. 19 ... Rg8?? It was essential to play 19 ... Kg7 and the struggle continues, although White’s advantage would not be in doubt – he could play 20 Rg3, for instance, with the idea of h2-h4, seeking to exploit the weaknesses in Black’s camp.

Exercise (easy): What had Anderssen overlooked when he played his last move?

Answer: 20 Rxe6! 1-0 If 20 ... fxe6 then 21 Rf3 wins. 204

In the fourth game of the match, the players repeated the opening moves of the second game. Game 29 A.Anderssen-P.Morphy 4th matchgame, Paris 1858 Ruy Lopez [C77] 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 d3 Bc5

In those days, when the theory of the openings was in its infancy, Morphy was critical of 5 d3, on the grounds that it allows the development of this bishop to c5, leading to equality. 6 c3 b5 7 Bc2 More frequently the bishop retreats to b3, although that position is usually reached by 5 ... b5 6 Bb3 Bc5 7 c3. At the time of writing, 7 Nc3 is more popular via that move order, and if 7 ... d6 then 8 Nd5. In the only game where Morphy reached a similar position (via 5 ... Bc5 6 0-0 b5 7 Bb3), after his retirement, against J.Sicre in Havana 1862, he played 7 ... d6 8 c3 h6.

7 ... d5! The retreat of the bishop to c2 seems to encourage this move, although Steinitz considered it to be premature. It comes as no surprise that Morphy should seek active piece play, as long as there no clear reason not to. 8 exd5 In the 16th game of the Steinitz-Chigorin World Championship match in Havana 1892, Steinitz played 8 Qe2, although he failed to gain any advantage after 8 ... 0-0 9 Bg5 dxe4 10 dxe4 h6; and Black has a good alternative in 9 ... d4, played successfully by Tarrasch and more recently by Ivan Sokolov. 8 ... Nxd5 9 h3 Preventing the threat of ... Bg4, which would gain in strength once White has completed his plan of advancing with d3-d4. 9 ... 0-0 10 0-0 h6 It is possible that this move, which weakens the kingside and might have been very important in the game, could have been postponed. Instead, the prophylactic retreat 10 ... Bb6! was interesting. 11 d4 This was Anderssen’s idea. The alternative was 11 Nxe5 Nxe5 12 d4, when Morphy may have rejected 12 ... Bd6, which concedes a slight initiative to White after 13 dxe5 Bxe5 14 f4 Bf6 15 Qd3 g6 16 f5 g5 17 Nd2, and instead chosen 12 ... Qf6, with good piece play for the pawn after 13 dxc5 Bb7, followed by ... Rad8. 11 ... exd4 12 cxd4 Bb6 13 Nc3

205

13 ... Ndb4?! The move isn’t a bad one in itself, but it forms part of a dubious plan. Question: Really? It looks like a “free move” that forces the bishop to retreat to the not very advantageous b1-square; then in order to evict the knight from b4 White would have to play a2-a3, another move of debatable utility. Isn’t that so? Answer: What we have here is a slightly atypical isolated queen’s pawn (IQP) position, with a black pawn on b5 (rather than b7) and a bishop on b6 (rather than e7). The absence of the e7-bishop means that Black’s kingside is slightly weakened. Nevertheless, his position is a good one; he should be able to cope with any problems on the kingside, while the bishop on b6, aiming at d4, can be an advantage.

As for the utility of a2-a3 and the inconvenience of having to retreat the bishop to b1, the negative side dissipates when we consider that one of the typical plans for White in IQP positions is to play Bc2 (or Bb1) and Qd3, for which it’s useful to insert a2-a3 to prevent a subsequent ... Nb4. There were several good moves instead of 13 ... Ndb4. The most appropriate, with a view to a long struggle, was 13 ... Nf6!, and after 14 Be3, Black can choose between 14 ... Bb7, 14 ... Be6 and 14 ... Re8.

14 Bb1?! White offers the d4-pawn; defending it with 14 Be3? would entail simply leaving the isolani as a weakness after 14 ... Nxc2. However, 14 Be4! was a good zwischenzug, and if 14 ... f5 then 15 Bb1, when the inclusion of ... f7-f5 naturally weakens Black’s position; after 15 ... Nxd4 16 Nxd4 Qxd4 (not 16 ... Bxd4?? 17 Qb3+) 17 Qf3, followed by a2-a3 and Rd1, White would have more than enough compensation for the pawn, as Maróczy pointed out.

14 ... Be6? This move is based on a tactical error. Although it wasn’t his intention, Black should have played 14 ... Nxd4, and even if 15 Nxd4 and 16 Qf3 still gives White some compensation, Black certainly isn’t worse. 15 a3 Nd5

Exercise: This position was reached in both games two and four of the match. Surprisingly, neither player noticed a strong continuation 206

for White here. Can you find it?

16 Be3?! In game two Anderssen played 16 Ne2? Nf6 17 Be3, which we shall examine in Supplementary Game 29.1. The text move is a slight improvement, defending the centre pawn by developing the bishop at once and maintaining the harmony of White’s position. The petite combinaison 16 Nxb5 is harmless: Black can choose between 16 ... Nf6 17 Nc3 Nxd4 18 Nxd4 Bxd4 with an equal position; or 16 ... axb5 17 Qc2 Nf6 18 Qxc6 Bd5! 19 Qc3 (taking the second pawn is risky, as after 19 Qxb5 Bxf3 20 gxf3 Ra5, the d4-pawn is lost and White’s king is left very vulnerable) 19 ... Re8, when his great piece activity compensates for the pawn. Answer: But 16 Qc2! is very strong. Now 16 ... Nf6? loses to 17 Ne4!, based on the weakness of the loose knight on c6; and 16 ... f5 is no better, since White can exploit another tactical weakness (the a2-g8 diagonal) with 17 Re1 Qf6 18 Rxe6! Qxe6 19 Nxd5 Qxd5 20 Ba2. Black would have to resign himself to playing a pawn down after 16 ... g6 17 Bxh6. Question: Isn’t it odd that neither of the two players saw this over the board, and even odder that it wasn’t found after game two? Answer: Yes, it’s very strange. But times have changed and things are very different today, when top games can be watched ‘live’ in the Internet and a blunder is flagged up by the engines the second after it is made.

16 ... Nf6 Anticipating the white battery after 17 Qc2 or 17 Qd3. 17 Qd2 Now Black has to check whether 18 Bxh6 is really a threat or not. 17 ... Re8 Bringing the rook into play and giving his king an escape square that might prove useful in some lines. Black concludes that the sacrifice on h6 can’t be prevented, but neither does it bring White any advantage; for example, 18 Bxh6 gxh6 19 Qxh6 Nxd4 20 Rd1 (20 Ng5? is parried by 20 ... Nf5) 20 ... Bb3 (or 20 ... Bc4) 21 Ng5 (or just 21 Qg5+ Kf8 22 Qh6+ etc) 21 ... Qd6 22 Bh7+ Nxh7 23 Qxh7+ Kf8 24 Re1 Qf6 25 Qh4 Kg8 26 Qh7+ Kf8 with a draw.

18 Rd1 Bd5 Morphy opts for a strategy of blockade, at the cost of his light-squared bishop. Instead, 18 ... Bb3! was interesting, to drive the rook off the central file and weaken the defences of the d4-pawn. After 19 Rc1 Qd7, Black seems to be fine, since 20 Bxh6 can now be met by 20 ... Nxd4! straight away, as well as taking on h6. 19 Ne5 Faithful to his style, Anderssen offers a pawn. A more positional player would have considered 19 Nxd5 Qxd5 20 Ba2 with a good position for White.

Exercise: What did Morphy play here? Do you think he accepted the pawn? Answer: No.

19 ... Qd6! The greedy 19 ... Nxe5? runs into 20 dxe5 Rxe5 (or 20 ... Bxe3 21 fxe3 Rxe5 22 e4) 21 Bxb6 cxb6 22 Ba2, winning a piece due to the pin on the d-file. The attempted counter-pin after 22 ... Qe7 (or 22 ... Qe8) 23 Bxd5 Rd8 fails to 24 Re1! Nxd5 25207Rxe5 Qxe5 26 Rd1 and Black is pinned again.

Alternatively, White can throw in 23 f4!, eliminating all counterplay before capturing on d5. 20 Qc2? Anderssen insists on attacking, but Morphy will show that this is unsound. Here 20 Bxh6 was also inadequate in view of 20 ... gxh6 21 Qxh6 Nxe5 22 Nxd5 (not 22 dxe5? Bxf2+! 23 Kxf2 Ng4+ or 23 Kh1 Rxe5 and wins) 22 ... Qxd5 23 Qxf6 Rad8!. Despite the engines’ interesting resource 24 Bh7+! Kxh7 25 Qh4+ Kg8 26 dxe5, aiming for perpetual check, Black can play 26 ... Bd4 27 Qg5+ Kf8 28 Qh6+ Ke7 29 Qh4+ f6!, when he still has some chances. If White lines up further forward with 20 Qd3, Black might strengthen his position with 20 ... Rad8, as suggested by Zukertort. White can’t then support his e5-knight with 21 f4? because of 21 ... Bc4! 22 Nxc4 bxc4 23 Qd2 Qe6, winning; so he should forget about the attack and play something like 21 Nxd5 Qxd5 22 Nf3, followed by Ba2, with equal chances. But in this case 20 ... Nxe5 21 dxe5 Rxe5 looks even better for Black, as 22 Bf4 can be met by 22 ... Qc5! 23 Qd2 Rh5, keeping the extra pawn.

Another possibility was 20 Nxd5 Nxd5, when one complex line, wholly in the style of Anderssen, was 21 Qc2 Nxe3 22 Qh7+ (22 fxe3!? is also possible) 22 ... Kf8 23 Nxf7! Kxf7 24 fxe3 Ke7 25 Be4, followed by Rac1. Despite having only a pawn for the piece, White has obvious compensation as both Black’s minor pieces are out of play and his king is weak.

Exercise: How did Morphy refute Anderssen’s idea?

Answer: 20 ... Nxd4! After the necessary calculation, Morphy exploits the greater centralization of his pieces. 21 Bxd4 Bxd4 22 Nxd5 22 Ng4 isn’t dangerous in view of 22 ... Bb7 or 22 ... c6, and with his rook on a1 White’s attack is going nowhere after 23 Nxf6+ Qxf6 24 Qh7+? Kf8 25 Qh8+ Ke7 26 Nxd5+ cxd5. The analysis engines consider 22 ... Nxg4 to be even better, seeing no danger to the black king after 23 hxg4 Rad8; the queen check on h7 is again harmless, as White is unable to back it up. 22 ... Qxe5 23 Nxf6+ Qxf6 24 Qh7+ Kf8 25 Be4

208

White is finally ready to bring his last piece into play, but at the cost of a pawn and weaknesses at b2 and f2. This is one of the rare instances where the presence of opposite-coloured bishops fails to help the attacker. 25 ... Rad8! Rather than take on f2 and risk trouble on the f-file, Black prepares to win the b2-pawn safely. Not yet 25 ... Bxb2? due to 26 Rd7! and White’s attack springs to life. 26 Kh1 After 26 Rf1 or 26 Rd2, capturing on b2 is still fine for Black. 26 ... Bxb2 27 Rab1 Rxd1+ 28 Rxd1 Qxf2 29 Qh8+ Ke7 30 Qh7 Be5 Three pawns up Morphy is content to simplify to an easily winning position. Otherwise 30 ... Rd8 was good, since 31 Rxd8 loses to 31 ... Qe1+ 32 Kh2 Be5+ etc. 31 Bf3 Qg3 32 Kg1

32 ... Qg6 This is good enough to win, even if Morphy might have been expected to seek to conclude the struggle in the middlegame; for example, 32 ... Bd4+! 33 Kh1 (or 33 Rxd4 Qe1+ 34 Kh2 Qe5+) 33 ... g6! 34 Qxh6 Rh8 35 Qd2 Rxh3+! 36 gxh3 Qxh3+ 37 Qh2 Qxf3+ 38 Qg2 Qxd1+ 39 Kh2 Be5+ etc. 33 Qxg6 fxg6 34 Bb7 Rb8 35 Bxa6 c6 36 Kf2?! Anderssen doesn’t defend in the most tenacious manner. It was better to play 36 Rb1, intending a3-a4 to free the bishop, even if it means Black gaining one or two passed pawns as a result. 36 ... Bd6 37 Rd3?! Here 37 Ra1 was more stubborn, with the same idea of playing a3-a4. 37 ... Kd7 38 Ke2 Ra8!

Another pawn falls and the rest is simple. All Black needs to do is advance the connected passed pawns with the necessary modicum of care. 39 Bb7 Rxa3 40 Rd1 Kc7 41 Bc8 Ra2+ 42 Kf3 Bc5 43 Be6 Rf2+ 44 Kg3 Rf6 45 Rd7+ Kb6 46 Bg4 Bd6+ 47 Kh4 c5 48 Bf3 209

c4 49 Rxg7 Rf4+ 50 Bg4 c3 51 g3 Rxg4+ 0-1 Supplementary Game 29.1 A.Anderssen-P.Morphy 2nd matchgame, Paris 1858 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 d3 Bc5 6 c3 b5 7 Bc2 d5 8 exd5 Nxd5 9 h3 0-0 10 0-0 h6 11 d4 exd4 12 cxd4 Bb6 13 Nc3 Ndb4 14 Bb1 Be6 15 a3 Nd5

As we saw in the main game, White could have gained the advantage here with 16 Qc2!. 16 Ne2?

Although this move threatens to win with 17 Qc2, that’s a poor return on a move that retards White’s development, and one to which Black can respond with something useful. 16 ... Nf6 Black defends h7 in advance and starts to apply pressure to the isolated queen’s pawn.

17 Be3 Re8! There were various attractive options here, such as 17 ... Bd5 and 17 ... Bc4. Morphy prefers include another piece in the struggle, postponing the above-mentioned alternatives to try to lend them more power. 18 Ng3 Bc4!

Once again it was difficult to choose between this move, 18 ... Bd5, 18 ... Na5, etc; although all of these are good, the text move forces White to take an important decision.

19 Nf5!? Objectively this sacrifice is unsound, but at least it complicates the struggle. Instead, 19 Re1 allows 19 ... Nxd4! 20 Nxd4 Bxd4 21 Bxd4 Rxe1+ 22 Qxe1 Qxd4, demonstrating one of the points of 17 ... Re8. And exchanging the light-squared bishops with 19 Bd3, so as not to lose material, would represent a positional capitulation, as after 19 ... Bxd3 20 Qxd3 Qd5, White would have nothing in return for the weakness of his IQP. 210

19 ... Bxf1 20 Qxf1 Recapturing with 20 Kxf1 would make little difference. 20 ... Ne7 Exchanging pieces is usually a good measure when ahead on material. For his part, White agrees to the exchange in return for retaining a knight on the f5-square, which always offers hopes. 21 N3h4 Nxf5 22 Nxf5 Qd7 This prepares the development of the a8-rook and is a good move. Alternatively, Zukertort recommended 22 ... Ne4, followed by 23 ... Qf6.

23 Bxh6! Anderssen had this second sacrifice in mind when he played 19 Nf5. It is almost forced, since it represents White’s only chance to complicate the struggle before Black completes his development with 23 ... Rad8, or plays 23 ... Ne4. 23 ... gxh6 Black is now a rook up, and “a rook is a rook”, so it would have been difficult for Morphy to opt instead for the equally winning continuation 23 ... g6, when 24 Qc1 can be answered by 24 ... Ne4. 24 Qc1 Bxd4 This is correct, involving the bishop in the defence. In contrast, 24 ... Nh7 25 Qxh6 f6 26 Ba2+ Kh8 27 Nh4 Rg8 28 Bxg8 Kxg8 29 Re1 is quite unclear. 25 Qxh6 Re1+ Once again the best move, pinning the bishop to the rook. 26 Kh2

Anderssen has managed to complicate the struggle, although Black’s position is still objectively winning. Exercise: The moment has come for concrete measures, as White threatens mate. What should Black play here? 211

26 ... Ne4? A calculation error. Answer: The only winning chance was with 26 ... Qxf5! 27 Bxf5 Rxa1, gaining two rooks and a knight for the queen. Once Black manages to get organized, even if he loses a pawn or two, his material advantage should prove decisive; e.g. 28 Qg5+ (White has to start checking; 28 Qf4 Rd8 29 Qxc7 Rd5 is worse) 28 ... Kf8 29 Qh6+ Ke7 30 Qf4 Rd8 31 Qxc7+ Ke8 32 Bc8 (32 Qc6+ Kf8 33 Qxa6 Be5+ 34 g3 Rd2 35 Kg2 Bd4 wins easily) and now after 32 ... Re1 or 32 ... Rd5, Black’s advantage is indisputable.

27 Bxe4 Rxe4 Forced, since the advantage passes to White after 27 ... Rxa1? 28 Nxd4!. Despite being two exchanges down, White’s three attackers form a very strong team; note that 28 ... Qxd4? loses to 29 Bh7+ Kh8 30 Bg6+ and mate in two moves. 27 ... Be5+? is also inferior: 28 f4 Rxe4 29 Qg5+! Kf8 30 fxe5 and White has a dangerous initiative; for example, after 30 ... Qd5 (the only move) 31 Qe7+ Kg8 32 Nh6+ Kg7 33 Qf6+ Kf8 34 Nxf7 Qxf7 35 Qh8+ Ke7 36 Qxa8, White has all the chances in the major piece endgame. 28 Qg5+ Kf8 29 Qh6+ Ke8 30 Nxd4 Qd6+ White was threatening 31 Qh8+ and Qxa8. Keeping the queens on with 30 ... Qd5 (defending the a8rook) isn’t promising after 31 Rd1, when the knight is taboo in view of 32 Qh8+. Taking the knight at once is no better: 30 ... Rxd4?? loses to 31 Re1+, while after 30 ... Qxd4 31 Qc6+ Ke7 32 Qxa8, the best Black can hope for is a draw. 31 Qxd6 cxd6 32 Rd1

White has a pawn for the exchange and his strong centralized knight defends the vital entry points on the second rank (c2 and e2). Since the knight can’t easily be dislodged, Black will find it virtually impossible to make progress (32 ... Rc8 33 g4 Rc4 34 Nf5 doesn’t help him at all). 32 ... Kf8 33 Rd2 Rae8 34 g4 R8e5 35 f3 Re1 36 h4 Rd5 37 Kg3 a5 38 h5 Kg8 39 Kf2 Re8 40 Kg3 Re7 41 Kf4 Kh7 42 Kg3 Re3 43 Kf4 Re8 44 Kg3 Re3 ½-½ Game 30 P.Morphy-A.Anderssen 5th matchgame, Paris 1858 Scandinavian Defence [B01] 1 e4 d5

After his defeat in the third game, Anderssen never again replied to Morphy’s 1 e4 with 1 ... e5 during the match. He played two Scandinavians, one Sicilian, and one irregular defence. This was a sort of tribute to Morphy’s mastery in the open games, as Anderssen later admitted – though as we will see, he lost all these games as well. They did subsequently play a series of informal games with 1 e4 e5, but those were all with the King’s Gambit, where the opening moves 1 e4 e5 2 f4 were obligatory. 2 exd5 Nf6 3 d4 The theory of the time, in the Handbuch, recommended 3 Bb5+, a move also employed by Fischer on the twentieth century. 3 ... Nxd5 4 c4 Nf6 Another idea is 4 ... Nb6; the position is then similar to some lines of Alekhine’s Defence (1 e4 Nf6) 212

in which, with the long a1-h8 diagonal left open, Black tries to apply pressure on d4 with ... g7-g6 and ... Bg7. Naturally, this is only one of the many development options available to Black. 5 Nc3 Bf5 6 Nf3 e6

7 Be3 A flexible move, not yet committing the other bishop. Alternatively, 7 Qb3 is typical of some positions in the Queen’s Gambit with an early ... Bf5, although here Black can reply 7 ... Qc8 and White hasn’t achieved much. Another thematic idea, borrowed from the Caro-Kann in this case, is to play 7 Bd3 Bxd3 8 Qxd3, as in the game P.Keres-O.Novotny, Prague 1943, which continued 8 ... Be7 9 Bf4 c6 10 0-0-0 Qa5 11 Rhe1, when White has the freer game and a slight advantage, but no greater than after the other options at his disposal. 7 ... Bb4? In this case the pin isn’t at all annoying for White; it was better to play 7 ... c6 or 7 ... Be7. Exercise: How did Morphy show that Black’s last move was inaccurate?

Answer: 8 Qb3 This move is now stronger than on the previous move, since the indirect pressure on b7-pawn is harder for Black to deal with. Morphy also considered 8 Qa4+ Nc6 9 Ne5 but decided that the text move was stronger, since after 9 ... 0-0 10 Nxc6 Bxc3+ 11 bxc3 bxc6, “whether White takes the pawn or not, in either case his opponent has an equal game”, as indicated by Maróczy. Today’s analysis engines agree with the assessments of Morphy and Maróczy. After 12 Qxc6 Rb8, for instance, White is behind in development; he can’t even restrain Black’s activity with 13 Bf4? because of 13 ... e5! 14 Bxe5 Re8 15 Qf3 (15 Be2?! Rb2 is worse) 15 ... Be4 16 Qe3 Bg6, reaching a position that both Anderssen and Morphy would have gladly played as Black. 8 ... Bxc3+ 9 bxc3! Question: Really? I was thinking that the idea of 8 Qb3 was to play 9 Qxc3, keeping his structure intact. Why recapture with the pawn? Answer: Taking with queen was no doubt a good alternative. Morphy’s move is more ambitious, since White is attacking b7, as we mentioned above. There is no obvious way for Black to try to exploit the doubled c-pawns, and on the positive side the pawn on c3 bolsters White’s centre, so that in the event of a future ... c7-c5 or ... e6-e5, White is ready to recapture on d4 with a pawn, retaining his central superiority.

9 ... Be4 Here 9 ... b6 was perhaps a better way to implement Black’s idea of placing the bishop on the long diagonal, so that after ... Be4 the bishop can retreat immediately to b7 if attacked. 10 Nd2 Bc6

213

Exercise: How did Morphy continue his development?

Answer: 11 Bd3 Of course; he doesn’t need to play f2-f3. If Black takes the g-pawn with 11 ... Bxg2, then after 12 Rg1 Bc6 13 Rxg7, it is his own king who is more inconvenienced by the opening of lines. 11 ... Nbd7 Exercise: How did Morphy continue now?

Answer: 12 Qc2 One should constantly ask oneself: “what is my opponent planning to do?”. Here the automatic 12 00?! runs into 12 ... Nc5!, exchanging the strong white bishop. 12 ... h6?! A move of debatable utility; it doesn’t seem an improvement on 12 ... 0-0, or even 12 ... Ng4 13 0-0 Nxe3 14 fxe3. White would stand slightly better in all cases. 13 0-0 0-0

Exercise: There are many possibilities here. What do you think Morphy played?

Answer: 14 Rae1 Morphy is faithful to his style, not hiding the fact that his target is the black king. Another way to 214

deploy the rooks is on the central files; Zukertort preferred 14 Rfe1, followed by 15 Rad1. 14 ... b6 15 h3 Morphy wants to control g4, so that he can play f2-f4 without having to worry about ... Ng4. But that’s not the only point of this move, as we shall see. 15 ... Qc8? Zukertort was very critical of this, one idea of which is to play ... Qb7, putting pressure on the long diagonal. He wrote: “One of those moves which a player may evolve from his inner consciousness, which baffle any attempt to explain them to the outer world. 15 ... Kh8 was very much to the point.” Anderssen’s problem is that his position is already unpleasant and all he can do is wait. One can suggest moves, such as Zukertort’s 15 ... Kh8, or else 15 ... Bb7 to allow ... c7-c5, but after 16 f4, with the idea of f4-f5 or Nf3-e5, White is still better.

Exercise: How did Morphy counter Black’s idea?

Answer: 16 Kh2! Question: That’s a surprise, and you’ve given it an exclamation mark; is it the best move? Answer: It isn’t clear whether it’s the best or not, but it’s definitely a move that matches Morphy’s style of play, and it’s consistent with 13 h3 as well. Since Black has helpfully provided a ‘hook’ with 12 ... h6, it should come as no surprise that Morphy considers breaking with a timely g2-g4-g5 advance.

In answer to 16 ... Qb7 Morphy didn’t want to play simply 17 f3, which would nullify Black’s battery and retain the advantage. He wanted to do something rather more aggressive; i.e. 17 Rg1, followed by g2-g4-g5. There were other ways to play, such as the natural 16 f4 Qb7 17 Rf2, intending f4-f5 or once again g2-g4-g5 (but not 17 Nf3 as that would allow 17 ... Be4!, weakening White’s attack). 16 ... Kh8 17 Rg1 Rg8 18 g4 g5

215

Anderssen takes radical measures to prevent g4-g5. Otherwise 18 ... Bb7 19 g5 hxg5 20 Rxg5 wasn’t pleasant. Exercise (easy): How did Morphy continue now?

Answer: 19 f4 Of course; opening lines against the weakened black kingside and, if only for a few moves, “chess is easy again”, as Najdorf used to say. 19 ... Qf8 Exercise: There are several possibilities here –

what do you think was Morphy’s choice? Answer: 20 Rg3! The idea is play Nf3, increasing the pressure on g5 to force Black to exchange on f4, opening the ffile and the diagonal c1-h6 in White’s favour. 20 Ref1! was of similar strength. 20 ... Rd8 On 20 ... Qd6, analysed by Zukertort, White would prepare Nf3 with 21 Rf1; for example, 21 ... Rg7 22 fxg5 hxg5 23 Nf3 Bxf3 24 Rxf3, followed by Qf2, when all the white pieces are attacking on the kingside. 21 Nf3 Bxf3 22 Rxf3 Qd6 23 Kg2

23 ... Nh5 216

Trying to complicate the game and confuse his opponent. An optimistic Löwenthal commented: “A very brilliant conception, and one that would probably have resulted successfully against a less formidable opponent.” Others were more critical: Zukertort described it as “a thoroughly unsound sacrifice”, while Steinitz, even more harshly, commented that it was “throwing away a clear piece without a shadow of justification.” Although those two masters were right, it is hard to suggest anything better; for instance, after 23 ... c5 24 fxg5 hxg5 25 Qd2, the black position is indefensible. 24 fxg5 hxg5 25 gxh5 g4 26 hxg4 It is quite correct to take the pawn; White has more than sufficient defensive resources to cope with the opening of the g-file. Alternatively, he might have played 26 Rxf7, since 26 ... gxh3+ 27 Kh1 is nothing to worry about either; e.g. 27 ... Rg7 (or 27 ... Rg2 28 Bf4 Qc6 29 Be4) 28 Bf4 e5 29 Rxg7 Kxg7 30 Rxe5 and there is no defence. 26 ... Rxg4+ 27 Kf1 f5 28 Qf2 Ne5

Exercise: What was the most crushing way for White to win?

29 dxe5 Morphy handles this phase of the game, following Anderssen’s piece sacrifice, less accurately, which perhaps goes some way towards justifying Löwenthal’s assessment of 23 ... Nh5. Answer: The text move is good enough, but 29 Bf4! was stronger, as after 29 ... Nxd3 (or 29 ... Rxf4 30 Rxe5!) 30 Bxd6 Nxf2 31 Be5+, White wins more easily.

29 ... Qxd3+ 30 Qe2 Qe4 31 Bf2 Qc6 32 Rd1 Rxd1+ 33 Qxd1 Qxc4+ 34 Qd3 Here 34 Qe2 was playable, retaining the a2-pawn, but Morphy’s move is also correct and contains a good idea ... 34 ... Qxa2

217

Exercise: What was Morphy’s idea?

Answer: 35 Rg3 Forcing the exchange of either the rooks or the queens, and making the technical task simpler. 35 ... Qc4 36 Qxc4 Rxc4 37 Rg6 Rc6 After 37 ... Rxc3 38 Rxe6, the passed e-pawn is decisive; e.g. 38 ... Rh3 39 Rh6+ Kg7 40 e6! Kf8 41 Kg2 Rd3 42 Rg6, followed by h5-h6 etc. 38 c4 a5 39 Ke2 Rxc4 Black has to do this now, before Kd3 leaves his rook with no squares. 40 Rxe6 Rc2+ 41 Kf3 a4 42 Rg6 Instead, 42 Re8+ Kg7 43 e6 was stronger: White’s two passed pawns are much faster than Black’s; but the text move doesn’t change the result. 42 ... Rc4 43 Rg1 a3 44 e6 a2 45 Ra1 Here 45 e7 Re4 46 Bh4, followed by Ra1 and Rxa2, was quicker. 45 ... Re4 46 Rxa2 Rxe6

Black has made some progress, but not enough. Although White has only the h-pawn pawn left, he has the ‘right’ bishop to control its queening square. 47 Kf4 Rd6 48 Kxf5 Rd5+ 49 Kg4 b5 50 Ra8+ Kh7 51 Ra7 Rd7 52 Bg3 Rg7+ 53 Kh4 Rf7 54 Rxc7 1-0 In the sixth game Anderssen resorted to playing 1 a3, trying to avoid open positions altogether; and he was successful, in so far as he gained a winning advantage, but he went on to squander it, eventually even allowing the draw to escape him, as we saw from the extract in Chapter One. 218

In the seventh game Anderssen once again played the Scandinavian Defence. Game 31 P.Morphy-A.Anderssen 7th matchgame, Paris 1858 Scandinavian Defence [B01] 1 e4 d5 2 exd5 Qxd5 3 Nc3 Qa5 4 d4 e5

Question: I find this move surprising. Can it be sound to open up the game like this as Black, having only developed the queen? Answer: Indeed, it seems too ambitious and feels suspect for the reason you mentioned. It looks inferior to the usual lines beginning with 4 ... c6 or 4 ... Nf6. On the other hand, Anderssen’s idea creates immediate tension and allows him to develop his minor pieces quickly.

5 dxe5 A reply that one might expect from Morphy, taking up the challenge and aiming for speedy development. In S.Tarrasch-J.Mieses, Gothenburg 1920, White played differently, developing quickly but not objecting to an early exchange of queens, relying on his more active pieces even in the endgame: 5 Nf3 Bb4 6 Bd2 Bg4 7 Be2 exd4; although Black has been able to mobilize both his bishops quickly, White’s advantage in development and piece coordination soon makes itself felt: 8 Nxd4 Qe5 9 Ncb5! (the alternative was 9 a3, but Tarrasch evaluated that in the ensuing endgame his pieces are better placed to give him the advantage) 9 ... Bxe2 10 Qxe2 Bxd2+ 11 Kxd2 Qxe2+ 12 Kxe2 Na6 13 Rhe1 0-0-0 14 Nxa7+! Kb8 15 Nac6+! and White went on to win. 5 ... Qxe5+ Subsequent attempts to rehabilitate this line have focused on accelerating Black’s development with 5 ... Bb4 or 5 ... Nc6. 6 Be2 In a 1913 simul – i.e. before his above-mentioned game with Mieses – Tarrasch played 6 Qe2 Qxe2+ 7 Bxe2, followed by Bf4 and 0-0-0, when the white pieces were definitely more active in the ending, but against a strong opponent it wouldn’t have been as easy to increase this slight advantage. 6 ... Bb4 In O.Duras-C.Schlechter, Bad Pistyan 1912, Black tried to revive the line with 6 ... Bg4, and although the game was drawn, it seems that White’s play can be significantly improved at various points. After 7 Be3 Bxe2 8 Ngxe2 (here 8 Qxe2, followed by Nf3, would give White a big lead in development) 8 ... Bc5 9 Bxc5 Qxc5 and now, instead of 10 0-0 (as in the game), it looks appropriate to play ‘à la Tarrasch’ with 10 Qd4! Qxd4 11 Nxd4, followed by 12 0-0-0 or 12 Ndb5, when White stands better.

219

Exercise: How do you think Morphy responded to the attack on c3?

Answer: 7 Nf3! You’re not really surprised by this, are you? Morphy offers a pawn in order to speed up his development, gain the bishop pair, and open the b-file in his favour. Steinitz wrote that this was a unnecessary sacrifice; but after the natural 7 Bd2, Black can play 7 ... Bg4 under better conditions than in Duras-Schlechter, since White can’t respond with Be3. Thus Morphy’s pawn sacrifice has its logic. 7 ... Bxc3+ 8 bxc3 Qxc3+ 9 Bd2 Qc5 10 Rb1 The rook is activated, both putting pressure on b7 and envisaging an eventual transfer to the kingside via Rb5 and possibly Rg5. 10 ... Nc6 11 0-0 Nf6

12 Bf4 Question: There are several playable moves here; is Morphy’s choice best? Answer: There are three main possibilities: Morphy’s move, 12 Rb5, and 12 Bg5.

Maróczy criticized Morphy’s choice and suggested 12 Rb5 Qd6 13 Re1 0-0 14 Qc1, but after 14 ... b6, mobilizing the bishop, for instance, it is not clear how White gains the advantage; the rook can’t really achieve anything along the fifth rank. However, 12 Bg5 looks more attractive, planning to damage Black’s pawn structure; e.g. 12 ... 0-0 13 Bxf6 gxf6 14 Qd2 (heading for the kingside) or 14 Rb5 Qd6 15 Bd3, when Black’s weaknesses and the agility of White’s pieces provide compensation for the pawn. The same would apply to the endgame arising after 12 ... Qd6 13 Qxd6 cxd6 and here, in addition to 14 Bxf6, White has 14 Nd2, planning Bf3 and/or Nc4, again with sufficient compensation. 220

12 ... 0-0! Steinitz didn’t consider it necessary to return the pawn. Unfortunately, he didn’t suggest an alternative; 12 ... Qe7 would have been answered by 13 Re1, maintaining the pressure. Despite the criticisms Anderssen appears to be defending well; in any case he soon reaches a position with equal chances. 13 Bxc7 Nd4! A good choice from among several reasonable moves, such as 13 ... Re8 or 13 ... Ne4, intending ... Nc3. Anderssen’s move eliminates the c7-bishop that might have dominated the dark squares. 14 Qxd4 Qxc7 15 Bd3 Bg4?! But this move can definitely be criticized. Instead, with 15 ... b6!, followed by 16 ... Bb7 or even 16 ... Be6 (attacking a2), Black could have neutralized White’s pressure.

Exercise: How did Morphy bring the position to life?

Answer: 16 Ng5! Preventing the exchange on f3, setting up the obvious threat of 17 Nxh7, as well as the possibility of 17 Ne4. 16 ... Rfd8 This move was criticized too, but the suggested improvement 16 ... Bh5 17 Ne4 Ng4? (adorned with an ‘!’ by some annotators) 18 Ng3 b6 is unsound. Exercise: How can White gain a big advantage after 18 ... b6 - ? Answer: After 19 Be2!, Black can’t prevent the loss of material; e.g. 19 ... Rfd8 (or 19 ... f5 20 h3!) 20 Qb4! and if 20 ... Nxh3!? then 21 Rfe1! Bxe2 22 Rxe2 f5 23 Re7 and wins.

It was also suggested that Black might head for a rook ending a pawn down with 17 ... Nxe4 18 Qxe4 Bg6 19 Qxb7 Qxb7 20 Rxb7 Bxd3 21 cxd3 Rfd8. Although the phrase “all rook endings are drawn” applies in many cases, this isn’t one of them. Instead of the passive 22 Rd1, White can play 22 Rc1 (or 22 Re1), intending to double rooks on the seventh rank; or if 22 ... Rxd3 then 23 Rxa7. 17 Qb4 Bc8? A perhaps demoralized Anderssen makes a move that retains material equality but concedes the initiative.

It was more tenacious to sacrifice a pawn with 17 ... h6! 18 Ne4 Nd5! 19 Qxb7 Rac8, when Black’s activity would make it difficult to breach his position.

221

Exercise: How did Morphy proceed here?

Answer: 18 Rfe1 By activating his last piece, of course, with the threat of 19 Re7. 18 ... a5 Now 18 ... h6 would be met by 19 Re7!, when 19 ... Rd7? allows mate in three with 20 Re8+! Nxe8 21 Bxh7+ Kh8 22 Qf8, while 19 ... Bd7 20 Nxf7 Nd5 fails to 21 Nxh6+! (Maróczy) 21 ... gxh6 and now the quickest win is 22 Bh7+ Kf8 23 Re8+! Kxe8 24 Bg6 mate. 19 Qe7! Question: Hmm, exchanging queens? Answer: Yes, thanks to White’s great dynamic advantage this is the quickest way to win material; Black can’t defend f7 satisfactorily. 19 Qc4!, with the same idea, was also good.

19 ... Qxe7 20 Rxe7 Nd5? This hastens the defeat, though Black’s position was very unpleasant in any case; if 20 ... Rf8 then 21 Bc4. More tenacious was 20 ... h6 21 Nxf7 and only now 21 ... Nd5 (rather than 21 ... Kf8?, which is refuted by 22 Nxd8 Kxe7 23 Nxb7 Rb8 24 Re1+ and Nxa5). Even then Black has a difficult task ahead; for example, after 22 Nxd8 Nxe7 23 Nxb7 Rb8 24 Be4 Bf5 25 Bf3 Bxc2 26 Rc1 Bg6 27 Nxa5. 21 Bxh7+ Kh8 22 Rxf7 With two extra pawns the result is not really in any doubt. 22 ... Nc3 23 Re1 Nxa2

222

Exercise (easy): Black has managed to recoup one of the pawns, but material is no longer important. What is the quickest way to win?

Answer: 24 Rf4 Morphy opts for a mating attack on the h-file, and with a timely Nf7+ hanging over Black as well. Obviously there were numerous other ways to win. 24 ... Ra6 25 Bd3 1-0 In the eighth game Anderssen again resorted to 1 a3. This time Morphy stood better at one point but lost control of the situation; after a complicated struggle (in which the computer points out that Anderssen missed a win when Morphy put his king on the wrong square), the game was drawn in 51 moves. The seventh and eighth games were both played on the same day, the 25th of December 1858. The ninth game was a miniature and is possibly the most well-known of the match. Game 32 P.Morphy-A.Anderssen 9th matchgame, Paris 1858 Sicilian Defence [B32] 1 e4 c5 The Sicilian Defence wasn’t very popular in the middle of the nineteenth century, so the contributions of Anderssen, Paulsen, De la Bourdonnais and Staunton were especially meritorious in opening up paths that were relatively unexplored and viewed with some scepticism. 2 Nf3 Nc6 3 d4 cxd4 4 Nxd4 e6 This move order was used by Kasparov at the start of his career, with the idea of transposing to a Scheveningen Variation, ‘normally’ reached via 1 e4 c5 2 Nf3 e6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 d6. Question: This route allows Morphy’s next move; what’s the advantage of playing this way? Answer: One of the Kasparov’s motives was to avoid the Keres Attack, 6 g4, which at the time was a main weapon of his great rival, Karpov.

5 Nb5 If there is a problem for Black with this move order, this is it – a move favoured by both Fischer and Karpov in their day. Indeed, Karpov played it against Kasparov, although on one occasion he used a ‘relative’ of the Keres Attack instead. This was in the 14th game of their second match, where he played 5 Nc3 and answered 5 ... d6 with 6 g4!?, despite there being no knight on f6 for the advancing g-pawn to attack.

5 ... d6

6 Bf4 This was Fischer’s favourite move, with which he beat Najdorf, Taimanov (twice) and Petrosian, although the last of these wins was somewhat fortunate (see below). It was revived for a while with some new ideas towards the end of the twentieth century. 223

White forces Black to play 6 ... e5, weakening the d5-square. In the current game it had a decisive influence on the result, but from the 1970s onwards, which saw the rising popularity of the Sveshnikov Variation (2 ... Nc6 3 d4 cxd4 4 Nxd4 Nf6 5 Nc3 e5 and so on), it began to be realized that such positions are very complex and the ‘hole’ on d5 is only one of the important factors. Karpov preferred 6 c4, reaching a Maróczy structure, and Fischer employed this too. 6 ... e5 7 Be3 f5? “Anderssen with his aggressive style wanted to hasten a crisis in the centre: previously such methods had always worked for him,” commented Kasparov.

In the twentieth century it was discovered that the correct way for Black is simply to continue his development with 7 ... Nf6. The historic game that discouraged the White supporters of this line continued 8 Bg5 Be6 9 N1c3 a6 10 Bxf6 gxf6 11 Na3 and here the spectacular counter-strike 11 ... d5! was played in the first game of the Fischer-Petrosian match, Buenos Aires 1971; Petrosian soon gained a clear advantage but ended up losing anyway. Later White sought new paths with 9 Nd2, leaving the c3square free for the b5-knight.

Exercise: How did Morphy respond to the threat to drive back his bishop with 8 ... f4 - ?

Answer: 8 N1c3! Morphy makes development the priority, seeking to prove, successfully, that the advance 7 ... f5 was a loss of tempo. Naturally, it is essential that the tactics work in support of his decision. 8 ... f4 If 8 ... a6 then 9 Nd5! axb5 10 Bb6 Qh4 11 Nc7+ Kd7 12 Nxa8 Qxe4+ 13 Qe2 is decisive according to Kasparov; and 12 Qd5 (threatening Qe6+ and mates) might be even better: the queen also defends e4, while after 12 ... Ke7, White can interpolate 13 g3 Qg4 (Black’s position after the exchange of queens would be untenable owing to his material deficit) 14 Be2 before capturing the rook. Exercise: How would you respond to 8 ... Nf6, controlling d5? Answer: White can play 9 Bg5!, when 9 ... Be7 fails to 10 Bxf6 gxf6? (but otherwise the d6-pawn drops off) 11 Qh5+ with a decisive invasion; e.g. 11 ... Kf8 12 Qh6+ Ke8 13 Qg7 Rf8 14 Nd5 with a crushing position. 9 ... Be6 is also insufficient: after 10 Nd5 Bxd5 11 exd5 Ne7 12 Bxf6 gxf6 13 Qh5+, Black must give up a pawn with 13 ... Ng6, since 13 ... Kd7 14 Qf7 is even worse. Exercise: What was Morphy’s idea?

Answer: 9 Nd5! Of course; the obvious threat is to check on c7, followed either by the capture of the a8-rook or an attack on the black king. 9 ... fxe3 Consistent; the sad 9 ... Rb8 would leave Black in a most inferior position, in which he has no compensation for the weakness on d5, and indeed the whole a2-g8 diagonal. One possible continuation 224

is 10 Nbc7+ Kf7 11 Bd2, followed perhaps by Bc4 and Bc3, and/or g2-g3 to open lines, and so on. 10 Nbc7+ Kf7

There are two candidate moves now: a materialistic one, and one seeking rapid development and an attack. It isn’t easy to know which is the better course of action without a very deep analysis, and that would be very difficult to do over the board. Exercise: What do you think was Morphy’s choice in this position?

Answer: 11 Qf3+?! Morphy chooses the most attractive option, going for the attack. Objectively, it isn’t the best, although from the practical viewpoint Black’s position is very difficult to hold. The materialistic 11 Nxa8! was stronger; e.g. 11 ... exf2+ 12 Kxf2 Qh4+ 13 g3 Qxe4 14 Bg2 and White’s advantage is clear. 11 ... Nf6 12 Bc4 Nd4! Question: Can this really be a good move, allowing White to take on f6 with check? Answer: Yes; surprisingly this counter-attacking move by Black is a good one, enabling him to activate his pieces and slow down White’s attack. In fact, although it isn’t pleasant allowing Nxf6 with check, this is the only possible try to complicate the game, while avoiding a quick mate, so in fact Black has no choice but to be bold.

If 12 ... Kg6, White can take the rook with advantage, but 13 Qg3+! is even stronger, and after 13 ... Ng4, as well as the simple 14 f3, White can play 14 f4, threatening 15 f5+. 13 Nxf6+

Exercise: Black’s position appears to be falling apart. What defence had Anderssen prepared here? 225

Answer: 13 ... d5! A magnificent counter-blow: for the modest price of a pawn, Black activates his f8-bishop, gains the d6-square, and also forces White to occupy d5 with his bishop, preventing Nd5+, which would be very strong in several lines. Admittedly, Black has little choice: a) 13 ... Ke7 loses to the sequence 14 Nfd5+ Kd7 15 Qf7+ Be7 (15 ... Kc6 16 Nb4+ leads to a quick mate) 16 fxe3 and the black king can’t be defended; e.g. 16 ... Rf8 (or 16 ... Nxc2+ 17 Kd2, threatening 18 Bb5 mate) 17 exd4! Rxf7 18 Bb5 mate. b) 13 ... Kg6 also fails, to 14 Qh5+ Kxf6 15 Ne8+! Qxe8 (or 15 ... Ke7? 16 Qf7 mate) 16 Qxe8 Nxc2+ (if 16 ... d5, hoping for 17 ... Bb4+, then simply 17 0-0-0! wins) 17 Kf1 e2+ (17 ... Nxa1 18 h4! and there is no defence against Qf7 mate; if 18 ... e2+ then 19 Kg1! e1Q+ 20 Kh2 etc) 18 Bxe2 (not 18 Kxe2?? Bg4+) 18 ... Nxa1 19 g4! “and Black, despite the favourable material balance, is helpless against the new wave of the attack”, according to Kasparov. 14 Bxd5+

Exercise: How should Black proceed now?

Answer: 14 ... Kg6? After conducting a difficult defence so far, Anderssen commits a blunder which costs him the game straight away. One suggestion here was 14 ... Qxd5 15 Nfxd5+ Nxf3+ 16 gxf3 exf2+ 17 Kxf2 Bc5+ 18 Ke2 Rb8, arguing that the two bishops provide compensation – but as Najdorf’s auntie used to say, “better one pawn more than one pawn less, old chap”. Here after 19 b4 Bf8 (or 19 ... Bd6 20 Nb5) 20 Rhd1 (or 20 c4) 20 ... Bd7 21 c4, Black “has a technically lost ending”, according to Kasparov. The strongest move, and one which seems to equalize, is Zukertort’s suggestion in Chess Monthly: 14 ... Ke7!, as the black king is better defended in the centre than on the kingside. Let’s look at this: Hunting the king with 15 Ng8+?! Kd6 16 Qf7 achieves nothing after 16 ... Nxc2+; e.g. 17 Ke2 Bg4+ 18 f3 Nd4+ 19 Kxe3 Qxc7 20 Qxc7+ Kxc7 21 Rac1+ Kb6 22 fxg4 Bd6 and Black wins material. A better try is 15 Qh5 gxf6 16 Qf7+ (there is no advantage in playing 16 Nxa8 as Black can bring his knight back to strengthen the defence: 16 ... Nxc2+ 17 Ke2 Nd4+! 18 Kd3 Ne6!, threatening 19 ... Nf4+, or else 19 ... Bd7) 16 ... Kd6 17 Nxa8! (17 Ne8+? now loses after 17 ... Qxe8 18 Qxe8 Nxc2+ 19 Kf1 e2+! 20 Kxe2 Bg4+ or 20 Kg1 Nxa1) 17 ... Nxc2+ 18 Ke2.

226

Here Maróczy analysed 18 ... Qe7 19 Qxe7+ Bxe7 20 Rac1 Nd4+ 21 Kxe3 Bd7 22 Rc7! Rxa8 23 Rxb7 Bc6 24 Bxc6 Nxc6 25 Rc1 Nd8 26 Rd1+ Ke6 27 Rc7 Rb8, commenting that “White’s position is slightly better”, whereas Kasparov considered that: “after 28 b3 Black cannot hold out.” Actually, Black can improve on this by playing 23 ... Rc8 24 Rxa7 Rc2, when he seems to have sufficient counterplay; but the analysis engines find the best line: 18 ... Nd4+! 19 Kd3 exf2 20 Bc4 Bh6 and the most probable result is a draw, which can now be forced by 21 Qd5+ Ke7 22 Qf7+ Kd6 23 Qd5+ etc. If White tries 19 Kxe3, Black has 19 ... Bh6+! 20 Kd3 Bd7, when White’s best course of action seems to be to set up the perpetual check mechanism again with 21 Bc4, intending 21 ... Qc8 22 Qd5+ Ke7 23 Qf7+ etc; but in this line Black can keep the game alive with 21 ... Bc6. 15 Qh5+ Kxf6

Exercise: Morphy found a way to win here – what is it?

Answer: 16 fxe3! Opening the f-file is the simplest. We already know that 16 Ne8+? is bad due to the familiar resource 16 ... Qxe8 17 Qxe8 Bb4+; but 16 f4! was equally strong. It was thought that Anderssen was expecting 16 Qf7+, when 16 ... Kg5 surprisingly appears to hold. However, Maróczy excused him, commenting that “the game was played very quickly, otherwise Anderssen would undoubtedly have noticed this simple move.”

16 ... Nxc2+ If 16 ... Qxc7, one way to win is 17 0-0+ Nf5 (or 17 ... Ke7 18 Rf7+) 18 exf5 (threatening 19 Qh4+) 18 ... Ke7 19 Rad1! and Black is defenceless. 17 Ke2 1-0 Anderssen resigned in view of the imminent check on f1; for example, 17 ... Nxa1 18 Rf1+ Ke7 19 Qxe5+ Kd7 20 Be6+ Kc6 21 Rc1+ and mates. 227

“An impressive defeat of the champion of the Old World!,” commented Kasparov. The game only lasted half an hour and left Morphy leading 7-2. Now he needed only one more win to seal the match. Once again beginning with 1 a3, Anderssen won the tenth game. It was a long struggle that lasted 77 moves and his comment showed a great sense of humour: “Morphy wins in 17 moves, whereas it takes me 77. However, this is still bearable ... ”. This was a great comeback by Anderssen, closing the gap to 7-3, but the following game turned out to be the last of the match. Game 33 P.Morphy-A.Anderssen 11th matchgame, Paris 1858 Irregular Defence [C00] 1 e4 e6 2 d4 g6

Question: Hmm, I understand that Anderssen was possibly the second or third strongest player in the world at the time, but this ... ? What is it? Answer: I agree that to us today this doesn’t look like the play of one of the world’s best players. Nevertheless, although this particular move is unusual, it leads to a formation frequently played in those days by Anderssen’s main rival for the number two position, Howard Staunton.

Black is employing a provocative move order that gives him the option of reaching what has been called the ‘Hippopotamus’ set-up. This consists of playing ... g7-g6 and ... b7-b6, fianchettoing both bishops, and deploying the knights on d7 and e7; generally Black waits to see what White is doing before deciding where to place his king. This was actually an interesting decision on Anderssen’s part – he was aiming for a closed position, on the basis that Morphy’s forte was in open games. Although it’s debatable whether Anderssen handled closed positions any better than his opponent, his results as White in his three games with 1 a3 (transposing, after 1 ... e5 2 c4, to a Sicilian Defence with reversed colours and an extra move for White) were encouraging, whereas his score with the other defences we’ve seen in the earlier games was depressing. 3 Bd3 Both players already had experience with White in this position; they both played 3 Be3 Bg7 4 Nd2 Ne7 5 Bd3 b6 6 Ne2 Bb7 7 0-0.

228

The central structure remains indeterminate; a whole range of very different position types can arise, depending on whether Black continues with ... d7-d6, ... d7-d5, ... c7-c5, etc. Note that White hasn’t played Nf3, in order to keep the option of playing f2-f4; while the other knight has been developed to d2, so as to be able to play c2-c3 and try to neutralize the g7-bishop. In a consultation game Anderssen/Horwitz/Kling vs. Boden/Kipping/Staunton, Manchester 1857, the Black team followed the scheme outlined above: 7 ... d6 8 c3 Nd7 9 Qb3 0-0 10 f4 d5 11 e5 Rb8 12 Rac1 c5, when they had achieved a reasonable position and went on to win after a long, manoeuvring struggle. In P.Morphy-A.Meek, US Congress, New York 1857, Black played 7 ... d5, which is more committal. After 8 e5 0-0 9 f4 f5?!, Morphy was able to form a clear plan: to break with g2-g4, to which end he played 10 h3! Nd7 11 Kh2 c5 12 c3 c4 13 Bc2 a6 14 Nf3 h6 15 g4 Kh7 16 Rg1, when White’s advantage was unquestionable. 3 ... Bg7 4 Be3 c5 5 c3 cxd4 6 cxd4

6 ... Nc6 It isn’t clear that Black’s decision to play an early ... c7-c5 and eliminate the tension by exchanging on d4 was the best; White has gained the c3-square for his queen’s knight, but what has Black gained? In A.Anderssen-H.Staunton, London (semi-final) 1851, Black tried to prove that he had weakened the opposing structure by playing more aggressively with 6 ... Qb6?!, attacking two pawns. However, the time lost in winning one of them was too much: after 7 Ne2 Qxb2? 8 Nbc3 Qb6, White had developed his pieces to natural squares and his position was already almost winning. There are now many strong ideas based on Nb5 and a knight invasion of d6. Anderssen opted for 9 Rc1 (the immediate 9 Nb5 is also good) 9 ... Na6 10 Nb5 Bf8 (a sad retreat; but 10 ... d6 was certainly no better, due to 11 Qa4! Bd7 12 Bd2!, threatening Ba5, and the black position collapses) 11 0-0 d6 12 d5 and achieved an overwhelming advantage. Although he played quite inaccurately from this point on, allowing Staunton to put up tenacious resistance, Anderssen nevertheless won the game eventually. 229

7 Ne2 As we’ve already seen, Morphy liked to retain the possibility of advancing his f-pawn. 7 ... Nge7 The knight is rather passive here, so it’s reasonable to suggest 7 ... Nf6 instead – but as we shall see, Anderssen had another idea for activating this piece. 8 0-0 0-0 9 Nbc3 d5

Black prevents any ideas of d4-d5 and, after White pushes the e-pawn, gains the f5-square for his knight. However, this comes at a high price, since the dark squares f6 and d6 are left weaker and his g7bishop is restricted. Either 9 ... d6 or 9 ... b6 seems preferable. 10 e5 f6 11 f4 fxe5 Anderssen again resolves the central tension (as he did with 5 ... cxd4). In general, unless some sort of advantage is gained thereby, it’s better to maintain it. He could have played 11 ... a6 straight away or 11 ... Bd7, say, since the exchange e5xf6 would not be favourable for White. Now, for his part, White has the f4-square available (even if it’s not as useful as the c3-square conceded by 5 ... cxd4), while his e3-bishop gazes down the now open c1-h6 diagonal. 12 fxe5 a6 Having opted for 11 ... fxe5, exchanging on f1 as well, before White can connect his rooks, looks more precise. 13 Qd2 Nb4

Exercise: How did Morphy reply to the attack on his bishop?

Answer: 14 Bg5! The light-squared bishop isn’t of vital importance in this position, so Morphy focuses on seizing the 230

weak dark squares on Black’s kingside. Here we can see one of the snags of Black’s decision to exchange on e5. 14 ... Nxd3 15 Qxd3 Bd7 Exercise: How did Morphy improve the situation of his pieces with his next move?

Answer: 16 Qh3! Transferring the queen to what will become the principal battle front: the kingside. White threatens 17 Qh4 and sets up the possibility of a timely Bh6, exchanging an important defender of Black’s kingside. 16 ... Qe8 17 Ng3

This is aimed against ... Nf5. Another possibility was the immediate 17 Rxf8+ Qxf8 18 Rf1, when 18 ... Nf5? loses to 19 g4. 17 ... Rc8 18 Rxf8+! Since there aren’t any more useful preparatory moves available, White ensures that he has a (temporary) numerical superiority on the kingside. 18 ... Qxf8 Perhaps 18 ... Bxf8 is slightly better, but it isn’t a big improvement. 19 Rf1 Qe8 20 Qh4 Nf5? Morphy will demonstrate that this move, which seeks to close the f-file, is inferior; 20 ... Nc6 should have been preferred. 21 Nxf5 Question: Well, surely this represents a small defensive achievement for Black? Now the f-file is closed. So what was so bad about 20 ... Nf5 - ? Answer: Yes, the file is now closed, but this isn’t the only change to the position, as we shall see.

21 ... gxf5

231

Exercise: What great advantage has Morphy gained in return for the closure of the f-file? How did he highlight this?

Answer: 22 Rf3! “The cure was worse than the disease” – the g-file has been opened, to White’s exclusive benefit; furthermore, the rook can go to h3 as well as g3. White’s advantage is decisive. 22 ... Bb5 Steinitz criticized this move, but there is no denying White’s advantage and there is nothing significantly better available. 23 Rg3 Rc7 24 Bf6

24 ... f4 A pawn sacrifice, which is as futile as any another move. If instead 24 ... Kh8 then 25 Qh6! Qf8 26 Nxb5 axb5 27 Rxg7! (not yet 27 Kf2?! in view of 27 ... Qxf6!) 27 ... Rxg7 28 Kf2! with two plans after exchanging on g7: one is to exchange all the pieces and send the king to c3 to capture the leading black b-pawn and then create a passed pawn; the other is to play the queen endgame by capturing on e6 (after Bxg7), first taking precautions so that the black queen can’t become too active. We can see the first idea in action after, for example, 28 ... Qf7 29 Ke3 Qg6 30 Qxg6 hxg6 31 Kd3 Kg8 32 Bxg7 Kxg7 33 Kc3 etc. The other plan might follow after 28 ... Qf7 29 h3 Qg6 30 Bxg7+ Qxg7 31 Qxe6 Qf4+ 32 Ke2. White wins easily in both cases. 25 Qxf4 Qf8 26 Nxb5 The simplest; Black could answer 26 Qh6 with 26 ... Bd3, intending ... Bg6, which should also lose, but Morphy doesn’t want to allow even that chance. 26 ... axb5 27 Qh6 Kh8 28 Rxg7 Rxg7 29 Kf2 Kg8 30 Qxg7+ Qxg7 31 Bxg7 Kxg7 232

With an extra pawn the endgame presents no difficulties. Morphy first consolidates the kingside, so that his king is free to go after the black b-pawns. 32 Kf3 b4 33 g4 b6 34 h4 b5 35 Ke3 b3 36 a3 1-0 By winning this game, Morphy also won the match (+7, -2, =2), and demonstrated the great difference between himself and all the other chess players of his time. He was quite rightly considered the world number one, without any rival who could offer him serious resistance.

On the following day they played eight informal games, using only one opening, the King’s Gambit. Morphy won 7-1, confirming that Anderssen had been justified in avoiding the open games for most of their match.

Morphy left Paris in the midst of great acclaim – he was called the “king of all kings” – and with that his serious chess career came to an end. In contrast, Anderssen went on competing for a further twenty years. He was frequently one of the three best players in the world, had a great success winning the Baden-Baden tournament in 1870 (ahead of Steinitz, whom he beat in both their games), and was always amongst the top ten. When Anderssen retired in 1879, he was still number five.

233

Chapter Seven Retirement On his return to New York, Morphy was received with the same admiration and recognition as Robert Fischer in 1972 after defeating Boris Spassky in Reykjavik.

Kasparov commented that “for the first time in the history of chess a victory was recognized as an event of national importance!”Morphy continued playing just as much but no longer ‘officially’. He gave blindfold simultaneous displays and played games or whole matches at the odds of ‘pawn and move’ or even a knight, even against some of his opponents from 1857; he won all these matches.

Let’s look at a game from the last informal match (back in Paris) that he played against the best player in France, Jules Arnous de Rivière – a match which Morphy, already in retirement, won 9-3. Game 34 P.Morphy-J.Arnous de Rivière Paris 1863 Giuoco Piano [C53] 1 e4 e5 2 Nf3 Nc6 3 Bc4 Bc5 4 c3

Question: This choice is rather strange, isn’t it? I thought Morphy played the Evans Gambit much more frequently than 4 c3. Answer: The former World Champion Max Euwe explains the reason: “At the time when this game was played Morphy had already withdrawn from public chess. Arnous de Rivière was his personal friend and they played occasionally to investigate particular variations.”

Having said that, out of the three times he reached this position as White in the match, Morphy chose the Evans Gambit on the other two occasions. 4 ... Qe7 In another game, as Black, Morphy played the standard 4 ... Nf6. After 5 d3 d6 6 h3 h6, Arnous de Rivière expanded on the queenside with 7 b4 Bb6 8 a4, whereupon Morphy opted for 8 ... a6 (there is nothing wrong with 8 ... a5 either, as in the current game) 9 Na3 0-0 10 Nc2 Be6 11 Qe2?! and then seized the initiative with 11 ... d5!.

5 d4 Bb6 6 0-0 Question: It looks to me as if 6 Ng5 would be awkward for Black here. Why didn’t Morphy play it? Answer: Both on this move and the next, Morphy could indeed have played Ng5, forcing ... Nh6, when the black knight is badly placed. However, this is not all roses for White because he has the problem of how to defend the d4-pawn; for example, 6 Ng5 Nh6 7 dxe5?! would give Black good play after 7 ... Nxe5, while the advance 7 d5, even though it gains a tempo on the c6-knight, increases the scope of the b6bishop.

6 ... d6 7 h3 It’s essential to prevent ... Bg4 and keep the central tension; but it was possible to insert 7 a4 a6 (or 7 234

... a5) and then play 8 h3, since 7 ... exd4?!, hoping for 8 cxd4 Bg4, runs into 8 a5! Ba5 9 cxd4 (threatening 10 d5 and 11 Qa4+) 9 ... Bb6 10 Nc3, followed by 11 Nd5, which is very unpleasant for Black (or if 10 ... Nf6 then 11 Bg5). 7 ... Nf6 8 Re1

8 ... h6 “To prevent White from playing Bg5, a move which isn’t a threat,” was Euwe’s comment, which is true; although the move ... h7-h6 is useful and has been played many times, it isn’t necessary just yet, since there isn’t a white knight on c3 to threaten Nd5 after Bg5. For instance, the classic game S.Tarrasch-A.Alekhine, Baden-Baden 1925, continued 8 ... 0-0 9 a4 a6 10 Bg5 h6 11 Be3, which Alekhine answered with a famous regrouping 11 ... Qd8!!, vacating e7 for the knight in case of d4-d5, and preparing the possibility of 12 ... exd4 13 cxd4 d5!. (Tarrasch actually played 12 Bd3.) If White tries to maintain the pin with 11 Bh4, it seems okay to play 11 ... g5, since 12 Nxg5 hxg5 13 Bxg5 exd4! looks fine for Black. Nevertheless, it is useful to know a typical manoeuvre for when ... g7g5 isn’t immediately advisable, which is to prepare it with 11 ... Kh8!, followed by ... Rg8 and then ... g7-g5. Instead of 10 Bg5, White can consider playing 10 b4 (or indeed 9 b4 on the previous move) with Ba3 on the agenda, as in the next note. 9 a4 a5 This is playable, though Black has more often preferred 9 ... a6 to avoid weakening the b5-square; e.g. 10 b4 0-0 11 Ba3 Nd7 12 b5 Nd8 13 Nbd2 Qf6 14 Nf1 Ne6 with an unclear position, R.SpielmannE.Eliskases, Semmering (5th matchgame) 1936.

Exercise: What do you think Morphy played in this position? 235

Answer: He carried on developing, which should come as no surprise.

10 Na3 Euwe made an interesting comment here: “Morphy always brought every piece into play as quickly as possible. To play the same piece twice in the opening was a breach of principle. Only under the influence of such players as Réti and Breyer, about 1920, was it realized that in some cases it can be good. The present position is such a case: White could obtain a definitive advantage with 10 Bb5! Bd7 11 Na3, planning 12 Nc4.” There is no doubt that development was important to Morphy; there are a great many examples demonstrating this, and relatively few examples where he moved an already developed piece, unless it achieved something tangible. In this particular case it isn’t clear that 10 Bb5 is better than 10 Na3. Let’s see: After 10 Bb5, it doesn’t appear that taking on c6 is a real threat, so there’s no need for ... Bd7. Instead, 10 ... 0-0 is playable, as after 11 Bxc6 bxc6 12 Nbd2 (or 12 Na3), Black has 12 ... Ba6, controlling c4; while 11 Na3 exd4! 12 cxd4 Nb4 13 Nc4 Ba7! is also very pleasant for Black; 14 Nxa5 can be met by 14 ... Nxe4!, intending ... c7-c6, since it isn’t possible to exploit the pin on the e4-knight. In the latter line, White should probably sacrifice with 12 Nc4 dxc3 13 Nxb6 cxb2 14 Bxb2 cxb6, when he has definite compensation for the two pawns, but Black’s position is more than playable. 10 ... Nd8? Question: Hmm, now that’s a strange move, isn’t it? Answer: You’re right; it looks as if Morphy’s friend didn’t ascribe the same value to development and piece activity as Morphy. This retreat, which prepares ... Be6 without losing a piece to d4-d5, is clearly inferior to 10 ... 0-0.

11 Nc2!

This contradicts Euwe’s opinion quoted above; Morphy moves an already developed piece. Instead of playing, for instance, 11 Be3 (which is perfectly good), Morphy sends his knight on a sort of ‘Spanish tour’, but instead of b1-d2-f1-e3, as in the Ruy Lopez (or ‘Spanish Game’), the knight takes the route b1a3-c2-e3, albeit with the same intention, to settle on d5 or f5. 11 ... Be6 12 Ne3 Bxc4 It is obvious that, with his king still in the centre and facing a rook on e1, 12 ... Nxe4 would be very risky; White can gain a big advantage with, for example, 13 Nd5 Bxd5 14 Bxd5 f5 and now, among many options, 15 Be3 is strong. 13 Nxc4 Nd7 Black supports the e5-pawn and also avoids doubled pawns after Nxb6. 14 Ne3 Now it isn’t possible for Black to defend both weak squares, d5 and f5, at the same time. 14 ... g6?! Black opts to defend f5, but this meets a concrete refutation. Protecting d5 with 14 ... c6 was no better, since after 15 Nf5 Qf6 16 dxe5 dxe5 17 Nd6+ Ke7 18 Nf5+ Ke8, now that Black has lost the right to castle, White can continue with 19 Qd3, followed by Rd1, or simply 19 Be3, bringing the bishop in play; Black can’t defend all his weaknesses. 236

Perhaps the most tenacious course was to accept the deterioration of his pawn structure after 14 ... Ne6 15 Nd5 Qd8 16 Nxb6 cxb6, although this would naturally be unpleasant too. 15 Nd5 Qe6 The sad retreat 15 ... Qf8 provides no relief after 16 Nxb6 cxb6 (once again 16 ... Nxb6 loses the e5pawn); but White’s position is so superior that he can already try for more with 16 Be3, continuing his development, given that the black king is stuck in the centre.

Exercise: How did Morphy exploit the lack of coordination among the black pieces?

Answer: 16 Bxh6! Winning a pawn, since 16 ... Rxh6 loses the queen to 17 Ng5. 16 ... f6 Guarding against the threat of 17 Ng5. 17 Bg7?! Simply 17 Be3 was much stronger, as pointed out Maróczy. Exercise: What is Black’s best defence here?

Answer: 17 ... Rh5! Not 17 ... Rg8, which would occupy the queen’s retreat square and allow 18 Bxf6!, with the threat of 19 Ng5 again; whereas now Black is threatening 19 ... Kf7. 18 g4 This weakens White’s castled position, but ... he needs to do something; the following moves are forced. 18 ... Rxh3 19 Nxf6+ Nxf6 20 Ng5

237

Exercise: What should Black play now?

20 ... Qd7? Answer: Instead of submissively returning the piece, it was better to head for the endgame with a pawn for the exchange; i.e. 20 ... Qxg4+ 21 Qxg4 Nxg4 22 Nxh3 Kf7 23 f3 Kxg7 24 fxg4 exd4 25 Kg2 Ne6, as indicated by Maróczy.

21 Bxf6 Rh4 22 f3 exd4 23 cxd4 White is now two pawns up for nothing, so the game is decided. 23 ... Rh6 24 Kg2 Preparing an invasion on the h-file. 24 ... Nf7 25 Rh1 Nxg5 25 ... Rxh1 26 Qxh1, followed by Qh7, was certainly no better. 26 Rxh6 Nh7 27 Qh1! Nxf6 28 Rh8+ Ke7 29 Rxa8 Two exchanges down, Black could safely resign. 29 ... Bxd4 30 Qh6 Qc6 31 Rc1 Qb6

Exercise: There are many ways to win here – what do you think Morphy played?

Answer: 32 Rxc7+! Of course; now 32 ... Qxc7 loses to 33 Qg7+. “It was unthinkable for a Morphy game to finish without fireworks.” (Euwe) 32 ... Ke6 White now mates in three moves. 33 Re8+! Nxe8 34 Qxg6+ 1-0 238

Mate follows with 35 Qf5.

On the 10th of July 1884, Paul Morphy died from a stroke in his native city of New Orleans. He was only 47 years old.

239

Paul Morphy’s Results The only games of Morphy’s that could be considered “official” are: First American Chess Congress, New York, 6 October-10 November 1857 Rd.1: P.Morphy 3-0 J.Thompson Rd.2: P.Morphy 3-0 A.B.Meek Semi-final: P.Morphy 3½-½ T.Lichtenhein Final: P.Morphy 6-2 L.Paulsen (+5, =2, -1) Match, London, 19 July-22 August 1858 P.Morphy 10-4 J.Lowenthal (+9, =2, -3) Match, London, September 1858 P.Morphy 5½-2½ D.Harrwitz (+5, =1, -2) Match, Paris, 20-28 December 1858 P.Morphy 8-3 A.Anderssen (+7, =2, -2) Match, Paris, February-March 1859 P.Morphy 7½-½ A.Mongredien Morphy played many games at various odds – a knight; pawn and move; pawn and two moves; pawn and three moves – and won most of them. He also played multiple games blindfold with success. His most remarkable results giving odds of the queen’s knight were against two former rivals: New York, May 1959 Morphy 5½-3½ J.Thompson (+5, =1, -3) New York, July 1959 Morphy 6½-2½ T.Lichtenhein (+6, =1, -4) Other significant results New Orleans, May 1850 P.Morphy 3-0 J.Löwenthal (or 2½-½) New York, 28-30 November 1857 (Morphy giving pawn and move) P.Morphy 4½-½ C.H.Stanley London, April 1959 (Morphy giving the queen’s knight) P.Morphy 7-2 T.H.Worrall (+7, =0, -2) New York, May 1859 (Morphy giving the queen’s knight) P.Morphy 5½-½ F.Perrin London, August 1959 (Morphy giving pawn and move) P.Morphy 6-1 J.Owen (+5, =2, -0)

240

Index of Complete Games Anderssen.A-Morphy.P, 2nd matchgame, Paris 1858 Anderssen.A-Morphy.P, 4th matchgame, Paris 1858 Arnous de Rivière.J-Morphy.P, Paris 1863 Bird.H-Morphy.P, London 1858 Harrwitz.D-Morphy.P, 3rd matchgame, Paris 1858 Harrwitz.D-Morphy.P, 5th matchgame, Paris 1858 Harrwitz.D-Morphy.P, 7th matchgame, Paris 1858 Löwenthal.J-Morphy.P, 9th matchgame, London 1858 McConnell.L-Morphy.P, New Orleans 1852 Morphy.P-Anderssen.A, 11th matchgame, Paris 1858 Morphy.P-Anderssen.A, 3rd matchgame, Paris 1858 Morphy.P-Anderssen.A, 5th matchgame, Paris 1858 Morphy.P-Anderssen.A, 7th matchgame, Paris 1858 Morphy.P-Anderssen.A, 9th matchgame, Paris 1858 Morphy.P-Arnous de Rivière.J, Paris 1863 Morphy.P-Arnous de Rivière.J, Paris 1863 Morphy.P-Ayers.T, Mobile 1855 Morphy.P-Baucher.H, Blindfold simultaneous, Paris 1858 Morphy.P-Boden.S, London 1858 Morphy.P-Boden.S, London 1858 Morphy.P-Bonford.J, Blindfold simultaneous, New Orleans 1858 Morphy.P-Duke Karl of Brunswick & Count Isouard, Paris 1858 Morphy.P-Harrwitz.D, 4th matchgame, Paris 1858 Morphy.P-Harrwitz.D, 6th matchgame, Paris 1858 Morphy.P-Lichtenhein.T, First American Congress (s-final, game 2), New York 1857 Morphy.P-Löwenthal.J, 12th matchgame, London 1858 Morphy.P-Löwenthal.J, 14th matchgame, London 1858 Morphy.P-Löwenthal.J, 2nd matchgame, London 1858 Morphy.P-Löwenthal.J, 4th matchgame, London 1858 Morphy.P-Löwenthal.J, 8th matchgame, London 1858 Morphy.P-Löwenthal.J, New Orleans 1850 Morphy.P-Morphy.A, New Orleans 1849 Morphy.P-Morphy.E, New Orleans 1850 Morphy.P-NN, Blindfold simultaneous, New Orleans 1858 Morphy.P-Paulsen.L, First American Congress (final, game 1), New York 1857 Morphy.P-Paulsen.L, First American Congress (final, game 5), New York 1857 Morphy.P-Potier.W, Blindfold simultaneous, Paris 1858 Morphy.P-Schulten.J, Blindfold game, New York 1857 Morphy.P-Schulten.J, Blindfold game, New York 1857 Morphy.P-Stanley.C, Blindfold game, New York 1857 Paulsen.L-Morphy.P, First American Congress (final, game 6), New York 1857 Schulten.J-Morphy.P, Blindfold game, New York 1857 Thompson.J-Morphy.P, First American Congress (rd.1, game 1), New York 1857 Thompson.J-Morphy.P, First American Congress (rd.1, game 3), New York 1857

241