BOTVINNIK'S BEST GAMESVolume 3: 1957 -1970 Mtkhatl Bolvinnik Mikhail Botvinnik BOTVINNIK'S BEST GAMES Volume 3: 1957-1970 (Analytical & Critical Works) Translated and Edited by Ken Neat Olomouc 2001 Published in the Czech Republic in 2001 by PUBLISHING HOUSE MORA VIAN CHESS P.O. Box 101, 772 11 OLOMOUC 2 Czech Republic This book is an authorised translation ofMM Botvinnik Analiticheskie i kriticheskie raboty 1957-1970 (Moscow 1986) ©English translation copyright Ken Neat 2001 Chess Agency CAISSA-90 All rights reserved ISBN 80-7 1 89-405-2 Contents From tournaments and matches - to understanding the essence of the game ... 7 Selected Games 1957-1970 252 253 254 255 256 257 258 259 260 26 1 262 263 264 265 266 267 268 269 270 27 1 272 273 274 275 276 277 278 279 280 28 1 282 Botvinnik-Smyslov, 1 957, 5th match game Botvinnik-Smyslov, 1 957, 9th match game Botvinnik-Smyslov, 1 957, 1 1th match game Botvinnik-Smyslov, 1 957, 13th match game Smyslov-Botvinnik, 1 957, 18th match game Smyslov-Botvinnik, 1 958, 1 st match game Botvinnik-Smyslov, 1 958, 2nd match game Botvinnik-Smyslov, 1 958, 6th match game Smyslov-Botvinnik, 1 958, 7th match game Smyslov-Botvinnik, 1 958, 9th match game Botvinnik-Smyslov, 1 958, 12th match game Botvinnik-Smyslov, 1 958, 14th match game Botvinnik-Smyslov, 1 958, 16th match game Smyslov-Botvinnik, 1 958, 21st match game Smyslov-Botvinnik, 1 958, 23rd match game Botvinnik-Raizman, 1958 Citrone-Botvillllik, 1 958 Botvinnik-Pomar, 1 958 Botvinnik-Alexander, 1 958 Botvinnik-Duckstein, 1 958 Uhlmann-Botvinnik, 1958 Botvinnik-Donner, 1 958 Tal-Botvinnik, 1 960, 3rd match game Tal-Botvinnik, 1 960, 5th match game Botvinnik-Tal, 1 960, 8th match game Tal-Botvinnik, 1 960, 9th match game Botvinnik-Tal, 1 960, 20th match game Tamburini-Botvinnik, 1960 Tarnows!Q-Botvinnik, 1960 Neikirch-Botvinnik, 1 960 Portisch-Botvinnik, 1 960 3 22 25 27 29 32 36 40 44 48 49 53 58 63 66 68 71 73 75 81 83 85 87 90 93 96 99 102 105 107 . 108 111 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 3 12 3 13 3 14 315 316 3 17 3 18 3 19 320 321 322 323 Botvinnik-Schmid, 1 960 Botvinnik-Paclunan, 1 960 Botvinnik-Tal, 1 961, 1st match game Botvinnik-Tal, 1 961, 3rd match game Botvinnik-Tal, 196 1 , 7th match game Botvinnik-Tal, 1 96 1, 9th match game Tal-Botvinnik, 1 96 1 , 1 0th match game Botvinnik-Tal, 1 96 1, 1 1th match game Botvinnik-Tal, 1 96 1 , 1 3th match game Botvinnik-Tal, 1 96 1, 1 5th match game Tal-Botvinnik, 1 96 1 , 18th match game Botvinnik-Tal, 1 96 1, 2 1st match game Botvinnik-Paclunan, 1 96 1 Botvinnik-Unzicker, 196 1 Botvinnik-Wade, 196 1/62 Littlewood-Botvinnik, 1 96 1/62 Robatsch-Botvinnik 196 1 /62 Botvinnik-Bisguier, 1 96 1 /62 Skold-Botvinnik, 1 962 Botvinnik-Lundin, 1 962 Botvinnik-Soderborg, 1 962 Unzicker-Botvinnik, 1962 Filip-Botvinnik, 1962 Botvinnik-Robatsch, 1 962 Botvinnik-Fischer, 1 962 Petrosian-Botvinnik, 1963, 1 st match game Botvinnik-Petrosian, 1963, 4th match game Botvinnik-Petrosian, 1963, 8th match game Botvinnik-Petrosian, 1963, 10th match game Petrosian-Botvinnik, 1963, 1 3th match game Botvinnik-Petrosian, 1963, 14th match game Botvinnik-Petrosian, 1963, 16th match game Krutikhin-Botvinnik, 1 963 Gipslis-Botvinnik, 1963 Taimanov-Botvinnik, 1 963 Kholmov-Botvinnik, 1 963 Botvinnik-Van Scheltinga, 1 963 Botvinnik-Donner, 1 963 Botvinnik-Petrosian, 1 964 Botvinnik-Smyslov, 1964 Botvin,nlk-Stein, 1964 4 1 12 1 15 1 17 1 20 1 22 125 1 29 132 135 137 141 143 146 1 49 153 155 1 57 160 162 165 1 67 169 171 174 177 184 1 87 1 89 1 93 1 95 1 98 203 206 208 2 10 2 13 217 219 22 1 225 227 324 325 3 26 327 328 329 330 33 1 3 32 333 334 335 3 36 337 338 339 340 341 342 343 344 345 3 46 34 7 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 Botvinnik-Medina, 1 964 Letelier-Botvinnik, 1 964 Aloni-Botvinnik, 1 964 Yanofsky-Botvinnik, 1 964 Botvinnik-Gligoric, · 1 964 Ciocaltea-Botvinnik, 1 964 Botvinnik-Larsen. 1965 Trifunovic-Botvinnik, 1 965 Botvinnik-Donner, 1 965 Botvinnik-Langeweg, 1965 Gipslis-Botvinnik, 1965 Botvinnik-Tolush, 1 965 Yudovich-Botvinnik, 1 966 LiberLon-Botvinnik, 1 966 Botvinnik-Szilagyi, 1966 Sz.abo-Botvinnik, 1966 Botvinnik-Pomar, 1 966 Botvinnik-Zuidem� 1 966 Botvinnik-Robatsch, 1 965 Botvinnik-Smyslov, 1 966 Botvinnik-Keres, 1966 Spassky-Botvinnik, 1 966 Botvinnik-Balashov, 1 9666/67 Liberzon-Botvinnik, 1 967 Levit-Botvinnik, 1967 Botvinnik-Polugayevsky, 1 967 Taimanov-Botvinnik, 1 967 Botvinnik-Boleslavsky, 1 967 Botvinnik-Toran, 1 967 Medina-Botvinnik, 1967 Botvinnik-Diez del Corral, 1 967 Botvinnik-Bednarski, 1 967 Gligoric-Botvinnik, 1967 Botvinnik-Matulovic. 1 967 Donner-Botvinnik, 1967 Botvinnik-Larsen, 1 967 Botvinnik-Padevsky, 1968 Botvinnik-Larsen, 1968 Benko-Botvinnik, 1 968 Botvinnik-Portisch, 1 968 Botvinnik-Kholmov, 1969 5 230 232 234 238 240 245 247 250 253 257 259 262 264 266 269 27 1 273 275 279 28 1 285 287 29 1 294 297 2·99 302 305 308 311 3 13 3 15 3 19 32 1 324 .326 330 333 336 338 341 365 Ostojic-Botvinnik, 1969 36.6 Botvinnik-Ree, 1969 · 367 Botvinnik-Van Scheltinga, 1969 368 Medina-Botvinnik, 1969 369 Botvinnik-Lombardy, 1969 353 357 359 370 Ciric-Botvimtlk. 1969 371 Langeweg-Botvinnik, 1969 362 365 372 Botvinnik-Kavalek, 1969 373 Botvinnik-Kurajica, 1969 374 Matanovic-Botvinni� 1969 368 372 377 375 Botvinnik-Ostojic, 1969 376 Matulovic-Botvinnik, 1969 380 383 377 Botvinnik-Suttles, 1969 378 379 380 381 344 348 351 387 Matulovic-Botvinnik, 1970 Botvinnik-Spassky, 1970 Botvinnik-Larsen, 1970 Spassky-Botvinnik, 1970 389 391 396 Training Games Postscript Tournament and Match Cross-tables 401 428 429 Summary of Tournament and Match Results (1957-1970) 457 Translator's Notes 459 Index of Openings 463 6 From tournaments and matches - to understanding the essence of the game ... from my final competitive appearances. This was a highly intense period. Over a return match. The question was - should this right be used? Pressure was put on me - they tried period of 14 years I played five matches to persuade me to give up the return Collected in this book are the games for the World Championship, and I took match. But the decision depended on part in many international tournaments, one thing alone: could I be confident four Olympiads, and also other team that I would be successful in a new competitions. encounter? In parallel with this I continued my After analysing the games from the scientific work in the field of electrical match I decided that 'a cat may look at a engineering, and from 1960 in the field king', and that it was possible to com of cybernetics. I also directed a junior pete with the new champion: The return chess school, and a considerable amount of time was taken up by writing. It all began with a failure in a World Championship match. The battle pro ceeded with alternating successes, but then I failed to withstand a difficult test, and Vasily Smyslov brilliantly won the match took place in the Spring of 1958. I made complete use of my analytical powers and was excellently prepared. It was easy to prepare - the lost match had given a wealth of preliminary infor mation. The champion, moreover, was over-optimistic, and also he had become accustomed to searching only when title of champion. Possibly I underestimated my oppon sitting at the chess ent - much to my cost! In the period board during a game ... By the 15th game I already had an from 1953 to 1958 Smyslov did not know any failures; this was the summit advantage of four points, and this game of his chess achievements. At the end of was adjourned in a position that was hopeless for White. It only remained for the event I was aiming only to 'rid myself of the match as soon as I could. m� to go off to my dacha for a night It was no longer possible to save it, but by the FIDE rules then in force a time analysis - this procedure had invariably proved · successfhl. But the defeated champion had the right to a difference in the scores had its effect: I 7 stayed in Moscow, analysed super ficially, and during the resumption became rattled. Even so. when ·1 was considering my 55th move, there were probably still hopes of winning. I learned about this when the FIDE Congress of 1959 abolished the right of the champion to a return match (begin ning in 1963). Although this decision was against the interests of the chess world, on egotistical grounds I did not protest - how long can one go on fight ing for the chess crown! Two years passed� and I faced a new opponent in a match (Vasily Smyslov had twice won the Candidates Tourna ment, but on the third occasion he was unsuccessful). In the late 1950s Mikhail Tai stag gered the chess world with a series of brilliant tournament successes. He won enormous popularity not only with his competitive results, but also his except ionally lively and clever play. His first prizes in two USSR Cham-· pionships, the Interzonal Tournament and the Candidates Tournament were fully justified. All this was crowned by Tai' s victory in our 1960 match, which brought him the title of World Champion, a result which was fully expected by chess enthusiasts. However, if one analyses things from the creative viewpoint, our match also gave a wealth of material for studying the deficiencies in the young champion's play. Even when it was not in the spirit of the position, Tal would aim for open play. He would go in for difficult positions, merely in order to obtain great mobility for his pieces, when his unique ability to calculate variations might tell, as well as... the opponent's lack of time for thought. Such a utilitarian approach . to chess secured the achievement of his goal, but Botvinnik Smyslov The resumption had taken a tense course, but finally the queens had been exchanged, and I breathed a sigh of relief - a feeling which arises when the time control has been reached. And so here I decided that the right number of moves had been made, and... I over stepped the time limit! Instead of being five points ahead, I had to be content with three. My mood was spoiled, of course, but the title of champion was regained. My colleagues (and not only they) were unhappy about this. They realised that in a match it was possible to defeat the old champion� but in a return match the art of preparation might still tell ... And behind my back a campaign to abolish the return match was begun. 8 at a high price. It led to a prejudiced style of play, to a narrowing of his creative possibilities, and harboured the possibility of future failures. A year later, in the return match, I was able to demonstrate that the creative defects in the young Latvian ' s play were more significant than his phenomenal calculating ability. In the first match I had been able to demonstrate this only once - in the 9th game. The return match was coming to an end. To general surprise, the young world champion, who everyone regard ed as a genius, was losing. . . Towards the end of ·the match some curious things happened. Before the 20th game the score was 1 1 Y2-7Y2. White had the advantage, but after the first adjourn ment I was able to analyse the position better, and I could have gained equal chances. But I made an incorrect move, and Tal again achieved a winning position. Then in a rook ending White played insufficiently energetically, the game was adjourned for a second time, and again analysis was required. I faced a second successive sleepless night. For a few hours the position seemed hopeless. The f5 and c6 pawns are weak, and the passed a6 pawn is powerful. But the thing that Black has to fear most is the possibility of the white king invading at b6 .. Even so, by morning an unexpected tactical chance was found: if White were to go in for a natural variation, Black would be saved by stalemate! And again the agonising search continued: what if White were to notice this stalemate? Finally, here too drawing chances were found. Botvinnik Tai I was in no doubt that my opponent would have slept peacefully all night. But Tal was Tal, and even at the board he might have noticed this tactical trick and chosen the correct, stronger contin uation. In order to lull my opponent's vigilance, when the game was resumed I did not take with me my usual thennos flask of coffee - to suggest that within a few moves I would be resigning. And so the game was resumed. 89 90 �a2 a7 l:lb5 My analysis had shown that the strongest continuation was to lure the black rook to a6: 90 l:f8 %%.a5+ 9 1 �b3 llxa6 92 llxf5 �d7 93 l:t f6 J:l.a l 94 f5 .D.d l 95 'ittc3 l:c l + 96 <it>d2 :!fl 97 l:tt7+ 'it>d8 98 'it>c3 l:tb 1 . Even so, it would appear that here too Black gains a draw, e.g. 99 f6 :n 1 00 'it>b4 'it>e8, or 99 <it>c2 llb4 100 �d3 l:.b3+ 101 �e2 :h3 102 f6 l:th4 103 <it>e3 l:.e4+ 104 @d3 : r4. 90 9 • • • lla5+ 91 92 @bJ .D.f8 ci>b7 llb5+ 95 96 It is important here to force White to decide where .to take his king, since it is on this that Black's further actions depend. For example, if 93 �c3 there would have followed 93.. Jla5. in order to attack the enemy king from the side (93 ... �xa7 94 l:.xf5 :as 95 :n+ <iti>a6 96 1lc7 or 95 ... <iti>b8 96 'iti>b4 :b5+ 97 �a4 is bad for Black) 94 l:.xf5 rl;xa7 95 :n+ (95 l:tf6 <it.">b7 96 f5 :a3+) 95 ... <it>a6 96 l:tc7 l:a3+ 97 <t>b4 l:.a4+. and the black king, and this is very important, is not driven onto the eighth 4Ulk. White, however, continues playing 'more naturally'. 93 94 <i>a4 llxf5 Af6 f5 'it>b7 l:lal+ First the king is driven away from the queenside, so that there should be no question of it invading at b6. 97 98 99 100 'it>b4 <it>cJ <it>d2 <it>eJ :bt+ Itel+ :n <i>xa7 llbl And only here did Tal notice that after 95 l:tf7+ 'iii>a6 96 :c7. as he was intending to play, there follows 96 ... .:tb4+! ! 97 �xb4 - stalemate (see Now the king is safely cut off along the f-file, and Black can wait. diagram). There was no longer any way of gaining winning chances, although White could still hope for some mistake by his tired opponent. 100 101 102 103 10 :.n+ <it>e2 �dJ <it>c7 <it>ds :r4 . l:lfJ+ . 104 �d2 not for chess. Tasks, which people have to try and solve every day and every hour, ·are search problems.· This applies in· particular to economics, where well known mathematical efforts prove ineffective. Chess is a very complicated search problem of great dimension (I happened to read an article by a foreign ex"Pert on cybe�etics, abo1=1t the fact that chess is more complicated than even economics or politics). So if one could identify how a master plays, and employ similar methods in applied search problems, this would be both a great theoretical, and a practical achievement! This is the problem to which I have devoted myself since 1 964. When I sensed that taking part in competitions prevented me from wor king successfully, I gave up tournament play ( 1970). And when I realised that my engineering work was also limiting my successes in creating a chess com puter program, I also gave up. directing the laboratory ( 1 974 ). What is the cru.x of the problem? It is positional understanding. It is this factor that distinguishes the play of a master from the play of the existing chess programs. And although, in searching for a move, mathematicians used computers that could perform on average as many as 2 50 million oper ations a second, due to the absence of positional Wlderstanding the optimum results were not achieved; I set myself the objective of develop ing a chess program which would act like a chess master, i.e .. possessing positional understanding. Most of the �cs And a draw was agreed on the 12 1 st move! To celebrate (after the return match) I played in the Christmas tournament in Hastings. Here in 1934/35 I had made my first appearance abroad, and .. . failed! On this occasion I rehabilitated myself, easily winning first prize. But in the following match, against Tigran Petrosian, I lost. And, it would seem, justifiably - I had already lost matches against both Smyslov, and Tai.'.. Age has its effect. And there was no longer any return match, thank God. In style Tigran Petrosian was not like other players. Only Flohr can be called his predecessor, and possibly Karpov Iris successor. What .unites them is a subtle and original positional under standing, and also a striving to achieve success without undue risk. It was always difficult to get at Petrosian's pieces, so skilfully did he defend his position. In· the first half of the 1960s Petrosian played exceptionally strongly, but the features of his style reduced his popularity among the broad mass of players. Before the match with Petrosian, doubts arose: should I play, or not? I had become fascinated by the problem of the artificial chess player, and matches for the World Championship took a great deal of effort. And only a year had passed after the loss of my title when I began finding the essence of the 'mysterious' way that a chess master searches for a move. This problem is of exceptional practical significance, and, of course, 11 experts regarded this as utopia. But the years have passed, and the first results have been obtained. The computer is now capable (this has been tested on several examples of various types) of choosing, on the basis of positional principles, 3-4 moves (out of 25-30), which are worth considering. Evidently the time is not far off when a computer will play like a chess master. On the basis of this research, . a method of retrieviJ,1.g algorithms for sol ving search problems was formalised. Practical tasks began to be solved using tliis method: the planning (with com puter aid) of the repair of power station equi ment �d the load levelling of electrical power system schedules. But let us return to chess. After I had opted out of the battle for the World. Cham ionship, for the first time since 1 948 I had time for events that were not even indirectly associated with official competitions for the champion's title. A small tournament in Amsterdam ( 1 963) was the first such 'free' event. The 1 965 tournament in the small Dutch town of Noordwijk was a happy event in my career. Wonderful walks across the deserted sandy North Sea beaches, good playing conditions, and tlie grandmaster's calculating apparatus - his head - working excellently. Tiris tournament was dedicated to the 70tl1 a.Ilniversary of the Leiden Chess Club . I won first prize, and together with it I received a certificate stating that I was an honorary member of the Leiden Chess Club. And then came my final appear ances... After 1 963 I sharply cut down on my· research work (in the field of preparing for competitions). From the creative standpoint I largely· had to live off my old (research) know ledge. And I played quite successfully, because I emp.loyed openings that I had studied and prepared before 1 963 . The USSR Spartakiad ( 1 967) was quite a good training event before the tmunament in Palma-de-Mallorc� which began at the end of November. Unfortunately, as often happens, Smys lov and I flew in late to Spain. Mean while, it was obvious that, on account of the difference between December in Moscow and the Mediterranean, we shoii.19-· have arrived in good time. The weather was Hot, we played in a stuffy building and we had to live in a room with the blinds drawn, in the semi darkness, to make it a little cooler. At the start I suffered a disaster. p p Botvinnik Damjanovic 12 In this position with an over whelming adv_antage for Black (two pawns up!) I contrived to lose. More over, it cannot be said that I blundered anything away. It was simply that with every move I worsened my position. Then I nevertheless adapted and offered serious competition to Bent Larsen, who in those years was enjoying his best results. In the 14th round we met. I had an overwhelming advantage in the middlegame, missed at least five winning continuations, but after an error by Black I reached an ending where, on the resumption (with Smyslov's help in the analysis of the adjourned position) I nevertheless won. There now seemed to be real chances of catching the leader, since we were separated by only half a point, but a weak finish by me (draws with Tatai and Calvo) led on the final day of the tournament to Bent being more fortunate and becoming the winner. A few months passed, and again Smyslov and I set off. for an inter national tou rnament, but whereas in November we flew via Paris to Spain, this time in April we flew to Monte Carlo. There was roughly the same group of players (only .instead of the Spanish national masters, here they were French), the same heat, and almost the same results. I played two very interesting games (with Portisch and Benko), but on this occasion I was unable to convert a winning position against Larsen, and as a result he again finished half a point ahead. In January 1969, along with Keres and Geller, I took part in tl1e main tournament of the Beverwijk Festival, which was being held for the 31st time. I should like to say a few words about these Dutch festivals. They are financed by the Hoogoven metallurgical firm, which then employed nearly 19,000 staff. Its products were largely exported, including to the Soviet Union and the USA. Wo_rking in the administration of the factory was an engineer by the name of Rueb, the son of the late Alexandre Rueb, the first President of the Inter national Chess Federation (and, which is less well known, a problem composer). Rueb-junior was the director of the festival committee (sometimes, and in particular in 1969, the tournaments were held in the neighbouring resort of Wijk aanZee). An interesting line-up had been assembled in the main tournament. Apart from the Soviet grandmasters, particular mention should be made of Lajos Portisch and Fridrik Olafsson. The Soviet players firmly held the lead. But our trio was closed on by Portisch, who after a poor start had as though acquired a second wind. A tense situation arose when he won against Keres. Portisch became especially danger ous, when he gained a winning position in his game with me. After somehow holding out until the time control, I adjourned the game, but I did "riot even bother to analyse it - the position was hopeless. During the first adjournment session the play became somewhat more complicated, but prospects . remained poor. After the seco�d adjournment .13 'No, that too does not help' came the unconsoling reply. Analysis showed · that Geller had evaluated the position correctly. By 'normal' means it could not be saved, but was there not some exceptional possibility? There was a knock at the door, and Keres came in. It turned out that he too had agreed a quick draw. 'Well, shall we have a look at your position?' And so we began analysing: I in bed, on my pocket set, and Paul Petrovich sitting at a table, on a big set... We quickly established that only one move could have been sealed. session there was no time to analyse it I had to eat quickly and again sit down at the board: round 13 was beginning. Portisch agreed a quick draw in his next game and disappeared . Of course, he had huni.ed off to analyse his end game with me. There was nothing I could do, and so I offered Benko a draw - the game would be resumed within five hours! I was lying on my bed. Towards the end of the tournament I had caught a chill, and after the illness I still felt weak. On my pocket set I tried to establish the truth. - . . ;Botvinnik 73 .1\a2 73 74 75 'it>e4 e6 Now White must be forced advance his e5 pawn. Therefore to li)c6 li)e7 Now the g5 pawn is lost, but Black improves the position of his king: 'iti>c5 a5 <ifi>d6 <ifi>e5 75 76 77 78 79 'it>e5 �6 'iti>xg5 .i.b3 Portisch White is preparing zugzwang. When Portisch was sealing his move, it seemed to me that Black had every chance of gaining a draw, but during dinner Geller gave me some unpleasant news: Black would lose due to zug zwang. 'Excuse me', I tried to object, 'can't the black king blockade the white e-pawn?' 79 80 81 .i.al .i.bJ! a4 a3 Here we both fell silent and sank into thought. Indeed, what can Black do? If he moves his king, White will occupy f6, and if the knight moves, the white king will break through via g6 ... 'Paul Petrovich', I remarked timidly� 'there is one drawn.position: if White's 14 king is at f7 and his pawn at g5, Black can play .. . @e5-f5, and after g5-g6 give check with his knight from e5, with a draw . . . Only how can this be achieved? Keres, an experienced study com poser, immediately formulated the idea: Only here did Portisch deviate: 83 �g7 83 84 85 86 87 g5 <it>h6 .la2 <i;g7 But this problems. 81 82 83 84 a2. @g6 <t;n thc6! the7+! thc6 87 88 caused any thc6 'iit>f5 the7 <li>e5! 'IW5 thg6 'iW7 88 . ..l'Dc6 was also possible. Here the players agreed a draw in view of the variation 89 .i.b1 + <i>xg5 90 .ltxg6 a2 9 1 e7 al 'if 92 e8'i' 'i'f6+. As a result, Geller and I shared first place, with Keres and Portisch half ·a point behind. This was my last successful tourna ment. By this time I had come round to the thought that it was time to give up playing chess. And when I was visiting my friends from the Leiden Chess Club (the reader will no doubt remember that in· 1965, after the toumament in Noord wijk, I was made an J1onorary member of this club), and they suggested that in g5 g6 longer Or 87 g6 �6 88 g7 tt:tl"5+. Or 84 e7 ED.xe7 85 <Ji;xe7 <itf4 86 .i.e6 8.4 85 no 'IW5 the5+! We laughed for some ten minutes - it turned out to be a very simple and elegant solution! When the game was resumed, strictly speaking, nothing new occurred: for the first l 0 moves both players followed exactly the given analysis. It rarely happens that in analysis both -players agree on the same 10-move optimal variation, but it is even more rare for their evaluations of it to be different. 15 the Spring of 1970, to commemorate the Fischer had again demanded that the 75th anniversary of their club, l should play a friendly match with match be staged with att!unlimited num Bobby ber of games, with which, as before, I Fischer, I promptly asked: could not agree. It has to be assumed that the future World Champion was 'But are you sure that with the intrac table grandmaster it will be possible to find a common language?' 'Yes, yes', I heard in reply. 'We have considered everything� all the negotia tions will be made through lawyers, and everything will be fine.' 'Then it will be the last .event I will seeking a pretext for avoiding the match. Apparently already then it was hard for Fischer to overcome that psy chological barrier, which often preven ted him from beginning an event. At the end of this year, which had take part in.' to take part in an international tour begun so successfully, I also happened The negotiations proceeded favour nament in Yugoslavia (for the first and last time). It was an extremely poor ably. First Fischer insisted that we should play to six wins with an un tournament for me. However I tried, I could not force myself to put up a restricted number of games; I suggested genuine fight. It was evident that in my sub-conscious the thought had taken that the 'players should meet 16 times at the age of 59 I would be unable to withstand a longer event. Finally the root that within a few months I would two sides reached a compromise: the be giving up playing chess. For a long time I lagged behind the winner would be the player who scored other participants, after suffering two more points out of 18 games. successive I took with me several hundred of defeats. Then I slightly Fischer's games and set off to the Black Sea for two weeks to prepare for the improved my affairs, only to lose again, to Janosevic. match. Fischer's games did not provide any surprises with their strategic ideas. Botvinnik These were known to any experienced player from old games (even before Fischer). But the American grand master's tactical resourcefulness, his energy in carrying out his plans. and his striving for activity at the very first opportunity were staggering. In September I received from the Leiden Chess Club's lawyer the formal agreement to the match, which had to be signed by the participants and organ isers. However, soon the Dutch sent me Janosevic a letter, ·in which I was informed that 16 This defeat was especially unpleas ant, both in terms of the play, and the competitive consequences. Although I am temporarily two pawns up, it is obvious that I have to fight for a draw - Black's position is so cramped, and he has numerous weak nesses. ·After a number of poor moves, Black finally makes a good one, removing his king from the eighth rank. 27 28 29 l:ttbt .i.xb5 <it>h7! b5 .D.xb5 32 33 34 :b7 :d7 lba7 34 35 36 37 i.g5 :ds+ :.xd6 • • • i.f6 38 39 fxe3 . . e3! • Or 3 9 :xn e2 40 :xg7+ �h6. 1 Or 2 9 �xb5 i..xa6 30 J:txb8 .i.xc4 as in the game, Black sacrifices the exchange in order to eliminate the dangerous passed a-pawn. 29 30 38 But strangely enough, altb.ough this move is undeniably pretty, .: it throws away the win. Black cannot take the bishop, since then he loses. his f7 bishop, and after it his pawns, one after another. But for the first time since the 20th move (! ) Black gains an oppor tunity to activate his king's bishop, and this gives him sufficient counterplay. 39 40 • . .i.h6 • l:laJ White defends against the threat of 40 . . . i.xe3+ 4 1 @hi lifl mate. He is also not afraid of 40 . . .tf8 in view of 4 1 l:ta8 ! (4 1 . . . .i.xd6? 4 2 l:th8 mate). Another possible continuation - 40 J::lxf7+ @g8 41 i..gS ! @xf7 42 i.xh6 �h5, which in fact later occurred in the game, should also have led to a draw. llxb5 .i.xa6 . The aim is achieved, but at a high price - the white rooks will operate very strongly along the 7th and 8th ranks. .tc4 31 :aS .i.xd5 i.f7 34 �a6 was simpler, when the d6 awn is immediately lost. @g8 l!f5 �h7 c4 Although Black has sufficient mater ial compensation for the exchange, his position is lost, since all his pawns are isolated. It would have been simplest for White to play his bishop to c3 (at the 4ame time it would be useful to place ·the king on e2), and then a concerted attack on the e5 pawn would have decided the outcome.: :' . Here I still had 1 5 minutes left for my. last move before the time control, but I made a mistake that had a psy chological context. . The.: .point is that Black as though buried himself alive ! If 17 I had only imagined that a drawing con tinuation .existed, I would have found it: 40 ... c3! 4 1 l:txc3 i.c4! 42 l:ld7+ @gs 43 l:dl (the interposition of 43 h3 g3 or 43 g3 llf3 does not change anything) 43 .. J:txf6 44 l:xc4.txe3+ 45 @h l.td4 46 lk7 :n 47 :xf7 �xf7 48 g3 <ifte6. By placing his pawn at e4, Black prevents the activation of the white king, and the rook will be tied to the defence of the h2 pawn. And if the h2 pawn is exchanged for the g4 pawn, a draw is inevitable, since in the fol lowing position (the maximum that White can achieve) he has no hopes of winning. The sealed move. Since 42 :a3 c3, as has already been established, leads to a draw (during the break before the re sumption, this also became clear to the Yugoslav grandmaster), White makes another attempt to convert his slight positional advantage. . 42 ltxf7+ <t>g8 . 43 .i.g5 In this way White obtains a superior rook ending; Black cannot play 43 ... .i.xg5 44 llxf5 gxf5 (44...i.xe3 + 45 llf2), since after 45 ltg6+ he remains a rook down. 43 <J;}xf7 44 �xh6 :b5! White is now forced to give back the bishop; he does not achieve anything with 45 l:d7+ 'iii>e6 46 :b7 c3 ! But he tries to obtain as much as possible for the piece! 45 <it>n! A subtle move, which· I under estimated in my adjournment analysis. After 45 l:c6 l:xh6 46 llxc4 l:h4l 47 g3 l:h8 48 J:.xg4 l:ta8 and 49... l:a2, or 45 .i.f8 :hs ! 46 lk6 <46 :d7+ �xf8 4r7 lk7 c3) 46...l:txf8 47 l:txc4 :as � l:xg4 l:a2 Black would ·have gained a draw without difficulty. llxh6 45 46 � g3 Here I calculated a lengthy variation: 46 ... c3! 47 l:tc6 e4! 48 :xc3 'iitf6 49 l:tc8 :h7 50 :£8+ �e5 51 l:f4 g5 52 l:xg4 �5 53 h4 gxh4+ 54 l:txh4 l:tg7+ 55 � (or 55 'it?h3 :a7 56 g4+ <it>e5 57 l:h5+ <it>e6 58 �g3 :f7 59 l:h3 <ii?e5) 55 .. .%la.7 56 llf4+ �e5 57 g4 :a3 (but not 57 ... :a2+ 58 �g3 :a3 59 :£5+ lt>e6 60 'itif4), but I was. unable to However, let us return to the previous diagram, where Black made a move that demonstrated his low spirits. lLg7 40 41 :a7 A comedy of errors! After 41 i.xg7 �xg7 42 l:ta5!, in view of the threat of 43 e4 l:tf6 44 l:td7 l:.e6 45 l:aa7 l:tf6 46 l:te7 Black would have lost his e5 pawn (in the event of 42 ... e4 his position is also hopeless). i.h6 41 • • • • • • 18 • • • • • . · neither player had won 6 games, the player who was leading at that point would be declared the ·winner. Fischer also turned down this initiative. Then, instead of a match, the Dutch organisers decided to hold a tou rnament of four grandmasters in four cycles, which in fact took place in the Spring of 1970. However, before this event I also took part in the so-called 'Match of the Century'. In it a team from the Soviet Union met a team from the rest of the world. This was an old idea. Although earlier I had been opposed to it, now, since I considered that the Soviet team faced a difficult task, I decided to take part. I learned to my surprise that I had been assigned board eight in the team. This was clearly unreasonable, since on results (after all, the tou rnament in Belgrade had been my only failure) I should have been given board four (on average my successes were superior to at least four participants who played higher than me). Apparently, in the USSR Chess Federation they were so confident of the success of the Soviet team, that did not consider my participation to be obliga tory. But I held a different opinion, and I considered myself obliged to play in the match. The Yugoslav Chess Union staged the match on a grand scale, and it pro voked enormous interest. Unfortunately, my suppositions were justified. Between our participants there was not that writy, which is so necessary in a team compe tition. The top boards of the Soviet team suffered a fiasco. The mate� was won evaluate correctly the resulting position. It appears that the e4 pawn is weak and that the g4 pawn is very dangerous. But in fact, if this variation is continued slightly- 58 :f8 l:tb3 59 l:te8+ 'ittd5 60 g5 :a3 61 g6 I:ta7 ! , it can be seen that Black must be able to save the ending. After missing this opportwrity, the ending is completely lost for Black, since both his pawns are weak, and his king is passive. 46 47 48 49 :c6 �xc4 l!e4 llh8 :as �a3 etc. Towards· the end of the Belgrade tournament I to some extent pulled myself together, but even so I finished only seventh - the worst result in my entire chess career. During the tournament the question was again raised about a match between me and Fischer. The Leiden Chess Club returned once more to the problem of a match with an wrrestricted number of games. I suggested another compro mise: the winner should be the first to win six games, but if after 18 games 19 with a minimal advantage,' thanks to those participants on whom they had not been depending, the greatest contri bution being made by Paul Keres, who had only been assi ed the bottom board. .. .Leiden, April 1970. Apart from the two Soviet participants, Spassky and Botvinnik, also playing were Bent Larsen and Jan Hein Donner. We lived by the sea in Noordwijk and played in Leiden. This decision of the organisers was regrettable, since a car journey before a game is something one can do without. In addition, the tournament was held in a damp, unheated school hall. I was also hindered by being conscious of the fact that this event was my last. In the first cycle I suffered an 'accident' in my game with Larsen. threat of 34 :as+ would have been most simply eliminated by the energetic 33.. .g5! The f6 pawn is immune (34 'i'xf6+? "i'xf6 35 l::txf6 .t.xf2+ 36 �h l .i.d4, and in view of the threat of 37 l:.xh2 mate, White can resign), and disaster on the f2 square is unavoidable. But I decided to defend the back rank without gaining a tempo. Botvinnik A move which I overlooked in time trouble. gn .. . '33 34 • • • <.t>hl 11fd8 .i.xf2 34 ... l:xf2 was more energetic, and if 35 :xf2 :xn (35....txf2 leads to a position that occurred in the game), then White cannot play 36 'flxc7 on account of 36 ... :txfl+ 37 <at>g2 1id2+ 38 @h3 'i'h6+ etc. This means that the queen must modestly retreat, and Black's attack continues. Now the "initiative passes to White. 35 35 36 ltd3! • • • i.xd3 l:bd3 ltb2 A new mistake, due to the fact that I again overlooked White's reply. Best in the given position was the counter threat 36 ... .i.d4, after which there would have still have been all to play for. 37 38 39 40 41 11fxc7 'ffxb7 l:tcl l:tc7 1!fd5 1!fg8 lla2 l:ta8 Jitb8 And White won. Thus my last tournament ended in failure. . . Some consolation was provid ed by the fact that I was proclaimed an honorary member of the Dutch Chess Union. Larsen White has been completely out played. He cannot defend his f2 pawn, and as a result his second rank will inev itably be exposed. The only counter- 20 Since that time a master's playing This book concludes the publication of my games, and it is possible to try algoritl� has in principle changed little. All that has occurred is a·-deepening and and answer the question: what, strictly speaking, was the contribution of the author to chess, and what did he do that others did not do? This is not an easy question. But first · expanding of positional understanding, and here in the 1930s to the 1950s Soviet :qiasters made a weighty contri bution to the theory and praxis of chess. most The art of preparing for competitions the became so refined, that a master, in the history of chess. Chess achieved a high quiet qf his study, unrestricted by time a slight significant digression, that has about the occurred in level when masters acquired a thorough (not like during a game!), was able to positional understanding. It was then find more complete evaluations of a that they began to create games which variety of chess positions. This also made an artistic impression, even from developed positional understanding and the viewpoint of a strict appraiser. Paul simultaneously led to good practical Morphy was a pioneer of this positional results. trend; he acted intuitively and did not It is this� evidently, that constitutes explain how he played chess. However, he played only open games, as was my contribution to chess. And the reader, after studying the games in these accepted in those days. three volumes, will certainly be able to Wilhelm Steinitz not only extended understand why I was victorious (when positional understanding to the closed this happened). It was because I was games, but he also explained how to act. superior to my opponents in the under After this masters began playing more standing of positions of the most varied strongly. nature. 21 Selected Games 1957-1970 7 8 9 10 11 Game 252 Botvinnik-Smyslov World Championship Match Moscow 1957, 5th game King's Indian Defence 1 c4 2 . lbcJ J g3 4 .A.g2 5 d4 6 lbfJ .i.g2 e3 d5 .i.xf3 lbc6 lbd7 e5 It would seem that, in order to exploit the advantage of the two bishops, White should aim to open up the position, but in fact his lack of development rules out this plan. For example, 1 1 0-0 exd4 12 exd4 tt'Jxd4 1 3 ..txb7 �b8 14 ..tg2 ltJe5, and Black has an excellent game. By blocking the position, White completes his development without hindrance, and the d5 pawn will control the light squares in the enemy camp. li)f6 g6 .A.g7 0-0 d6 .i.g4 In the Griinfeld Defence this man oeuvre is quite good, but here it is dubious. Black achieves a comfortable mobilisation of his forces, but the absence of his light-square bishop may tell. Black's early castling al�o does not go well with the exchange on f3, since White's chances of attacking on the kingside are improved. Even so, at present the text move occurs, along, of course, with the more popular continuations 6 . . . tiJc6, 6 . . . c5, 6 . . . tt'Jbd7 and 6 . . . c6, which, incidentally, was played in the next game. 7 .lxf3 11 • • • lbe7 It is well known that in the Ki.Jig's Indian Defence the activity of a knight at e7 is limited. Therefore Petrosian was correct in recommending 1 1 ... tiJcb8. 12 13 e4 h4! f5 This refutes Black's opening strategy. Now the exchange ·in the centre ( 13 . .. fxe4 14 tt'Jxe4) would weaken still further the light squares in his position. The threat of h4-h5xg6 is also unpleasant - this is where Black's castling is S\.tn to be premature. Smyslov takes the correct decision and restricts· the mobility of White's h3 I thought that this was the only way to play, since after 7 0-0 'i'c8 Black's idea would be justified. However, later theory gave a new recommendation: 7 'ib3 1fc8 8 ti)es dxe5 9 .i.xb7. 22 18 19 20 queen's bishop, although now his knight at e7 has altogether no future. 13 14 .ilh3 f4 l:tf6! . .i.d2 .ll.xd2+ lL\�6 White's advantage on the queenside and control of the h3-c8 diagonal, to gether with the weakness of the enemy f3 pawn, should guarantee him a win. Apparently played merely in order to vacate the f8 square for the knight, but in fact Black provokes his opponent into the variation 1 5 .i.e6+ 'it>h8 1 6 h5 g5, after which the possible exchange sacrifice on e6 would have led to him seizing the initiative. 15 1fe2 .i.h6 16 <i&?xd2 aJ 20 21 22 23 c6 ifdJ l:tab1 l:thcl lt\c7 l:tb8 a5 This merely leads to the creation of a new weak pawn at a5. 23 . . . b6 was somewhat better. lDc5 Superficially played. Of course, it is tempting to prevent the opponent from castling, but iil the given case it is not worth it. Black parts with his last chance of gaining counterplay on the f-file, wastes time on· an unsuccessful knight manoeuvre, and presents White with a tempo for connecting his rooks. Is this not too much? It was essential to play 16 . . . a5. although after 1 7 etJa4 Black has a difficult position 24 b5 25 b6 c5 After 24 . . .cxb5 25 cxb5 b6 26 ltJa4 the a-file would have been under White's control. An essential move, otherwise Black consolidates his position on the queen side with . . . b7-b6. 25 lhe8 lt\g7 26 :el 2 7 l:Ie3 First the enemy heavy pieces. are tied to the defence of the f3 ·pawn. f3 b4 iffl Only not 18 ii'd l because of 1 8 . . . 17 18 . tt:\d3+ 1 9 'it>f1 .i.xd2 20 Wixd2 liJxb4. 23 27 1if8 28 29 l:tb5 lha4 It becomes clear that one of the two pawns f3 or a5 - will be lost. 'ilfi h5 kl.b8 'lfb3 lhdJ :.et 9a4 �c2 1b7 ..i.xg4 45 46 47 48 49 ile7 :ff8 li)eS 1!i'b5 tiJb2 li)dl 1id8 tiJd7 li)f6 11fb5 a5 <li>d3 11fb2 l:tgt 1We7 'tlb7 lUi ltlh5 In order to radically neutralise the possible threat of . . . ttlxg3. 49 50 51 At just the right time. After 39 ...tLlef6 the exchange would have no longer been so effective. 39 40 1!f a5! Black had to reject the capture of the b6 pawn, since after 44... 1Wxb6 45 'ifxb6 ttlxb6 46 ttle3 or 44 .. . tt:lxb6 45 ttle3 the game would have gone into an ending where he is bound to lose material. �b7 lhg8 lh1l6 ttJg4 Black has finally managed to block the h3-c8 diagonal, but now this is merely of 'academic' interest. 36 37 38 39 42 42 43 44 The piece sacrifice (3 2 ttlxc5 dxc5 3 3 'i'xe5 :es 34 �d7 1%£8) would have merely led to a sharpening of the play, for which, of course, White has no need. 32 33 34 35 a4 This manoeuvre was found in analysis. Now if 42 ...'i'h7, in order to attack the e4 pawn by . . . g6-g5, there follows 43 'i'd.2. As for the b6 pawn, it is invulnerable� as will become apparent. - 29· 30 11c3 31 lha5 32 lhb2 41 hxg4 li)f6 hxg5 'ird2 g5 11bf8 llf4 An attempt to complicate the play at any cost. Alas! White pays no attention . 24 insufficiently well-prepared for them. Since White must not allow ... lbe4, and 8 lLxi2 and 8 tbel are less· forcing, his next move is fairly obvious. . 8 lLlb4 .i.e6 In the event of 8 . .i.d7 both Uhlmann's recommendation of9 f4, and 9 e4 are- possible. . 9 10 52 53 �c3 l:txg3 d5 cxd5 . cxd5 .td7 lDxgJ 1ih2 But suppose the opponent does not notice the 'threat' of 54 ... 'iixg3 55 fxg3 f2 ... 54 'tiel Black resigns. He noticed it! Game 253 Black's position could be considered quite satisfactory, were it not for the energetic plan available to . White involving the advance of his e-pawn. However, first White establishes his bishop at d4, exploiting the fact that this important square cannot be controlled by the black knights. Botvinnik-Smyslov World Championship Match Moscow 1957, 9th game King's Indian Defence 1 2 3 4 5 6 c4 ti)cJ gJ .i.g2 d4 lDfJ lDf6 g6 J..g7 0-0 d6 c6 11 12 13 0-0 lDa6 'iia5 lDc5 13 . . .:tfc8 came into consideration. In Game 252 6 ... �g4 was played. 7 .i.e3 .i.d4 l:tel . 14 .itf5 e4 Thus, White has successfully advanced this central pawn. In order to reduce somewhat the pressure of the enemy pieces, Smysldv tries to simplify the game. 7 . 'i'a5 or 7 ... a6 is more usual, inunediately beginning preparations for ... b7-b5. Contin�tions such as the one chosen here by Black have only one point the opponent may be . . 25 14 15 16 thxa4 b3 19 20 21 itla4 if xa4 White. naturally, avoids the exchange of queens, since in an ending Black's objective of gaining a draw would not have been so difficult. 16 17 • • • f4 • • . e5 ifa3 21 22 23 dxe6 :act fxe6 l£if6 1i'h5 23 . . . 'i'xd4+ 24 itlxd4 would have led to the loss of a pawn (24 . . . .i.a6 25 :c7). Now too Smyslov loses a pawn, but at least he avoids the exchange of queens, hoping to complicate the situation in his opponent's time trouble. .ilb5 itld7 White faces the dilemma of whether or not to advance his e-pawn. After all, this will involve a simplification of the position. 24 l£ig5 With the threats of 25 .i.xb7 and 25 :c7. Black immediately gives up a pawn, but he achieves the exchange of bishops, after which he may be able to e:\.-ploit the weakening of the white king's position. 24 25 26 27 19 .i.xd4+ 1lc5 White advantageously brings into play his knight, which for 12 moves has been id.le on the edge of the board. It unexpectedly transpires that the ex change of queens will involve loss of material for Black. Preparing a further pawn advance in the centre. Here Black should definitely have played 17 . . . l:.fc8, in order after 1 8 e 5 to retreat the latlght to e8. At d7 it will be less well placed. 17 18 1i'xd4 itlf.J! .i.xc6 lhc6 h4 .i.c6 bxc6 itlg4 It would have been safer, and hence more sensible when short of time, to choose 27 h3 . Then after 27 . . . ttJe5 28 l:!c3 the exchange of knights is in evitable, and it would have been very hard for Black to find counter-chances. e6 Now, after the exchange of the "fl pawn for the e6 or d5 pawn, White's positional advantage will be obvious. Smyslov had pinned his hopes on the exchange of que�ns, but apparently he had not antici ted his opponent's clever reply. 27 28 • • • tht7 h6 'it>h 7 Here White had at least two ways to win. The most obvious was 29 itlxd6. Now 29 . . . :ads is met by 30 'ii'xa7, and 29 . . . fl.fd8 by 30 1k7, while if 29 . . . 1i'a5, i}a 26 then 30 !te2 (it was this move that I did not find during my short think). Black is forced to play 30 . . . exd6 (or 30 . . . l:lad8 3 .l lDb7 2), when there follows 3 1 :xd6 :ctfd8 (if 3 1 ...lDf'6 32 e7) 32 l:.d7+ :xd7 3 3 'i'xd7+ <it>h8 34 e7. does nevertheless have one advantage it safeguards White against losing on time. 31 32 • • • 11xa7+ exd6 <it>b8 Of course, not 3 2 . . . <it>g8 in view of 3 3 'i'f7+ <it>h8 34 e7. 33 1ld7 Or 33 'i'd4+ �h7 34 e7 l!e8 3 5 'i'xd6. 'ib 5 3 6 li'e6 h5, and Black i s out of danger. 33 . . l1c3 Creating mating threats and forcing White to conclude the game with perpetual check. • 34 35 36 37 38 39 40 The second way begins prosaically: 29 :tc7 'ifa5 30 �ee l , but here it did not occur to me that if 30 . . . :fc8 White wins by 3 1 lDg5+! After 30 . . . 'ii'xa2 3 1 :xe7 'i'h2+ White's king has to embark on a lengthy journey (3 2 'Et>fl 'i'hl + 33 c:te2 'ilg2+ 34 <&t>e l 'i'xg3+ 35 <&t>d2 'i'g2+ 36 <t>c3 l:fc8+ 37 <it>b4). Of course, in time trouble it was not easy to judge how safe it was, and so I decided to prepare this last variation, after first taking control of the a5 square. 29 30 b4 :.xc8 Ganie 254 Botvinnik-Smyslov World Championship Match Afoscow 1957, 1 1 th game Grtinfeld Defence l:.tac8 After 30 '11c3 l:txc6 3 1 'ifixc6 White's winning chances are highly problematic: with 3 1 . . . 'i'fS 32 ltJxd6 'l'd3 3 3 �e4 'i'd4+ Black acquires real counterplay. 30 31 • • • 1'd8+ <ifi>h7 'it>h8 'ifd7+ ifd8+ 'it>h7 ci>h8 ife7+ ifeS+ <it>h7 1ie7+ 'it>h8 ci>h7 1fe8+ Draw agreed 1 2 3 4 5 6 �f3 gJ i.g2 c4 d4 cxd5 �f6 g6 �g7 c6 d5 6 ... cxd5 At that time this manoeuvre was considered to be quite dangerous for Black. l:xc8 tbxd6 Alas, this sacrifice is too late and now leads only to a draw. However, it 27 7 8 thcJ the5 0-0 11 12 An. essential move. After 8 0-0 tDe4 9 'i'b3 tDc6 ! , as occurred (with a transposition of moves) ten years later in the game Donner-Botvinnik (No.3 58), Black equalises easily. In the last round of the 1 93 5 international tournament in Moscow, playing Black against Ilya Rabinovich, I failed to find this manoeuvre. 8 • • . 12 i.g5 or 12 .i.f4 was also possible. White chooses a different plan. By threatening to play ..i.a3, he forces Black to advance . . . e7-e5, after which the weakness of the c6 and d5 pawns, and also the c5 and d4 squares, becomes obvious. If after this White also sue.: ceeds in playing e2-e4, his advantage will be undisputed. 12 /l)c6 . • e5 • White also has an advantage after 1 2 . . . c5 13 i.b2 cxd4 14 i.xd4 e6 1 5 l'.1c l . In the 2 1 st game of the same match, Smyslov played more accurately 8 . . . i.f5, and after 9 0-0 tl:\e4 the position became equal 9 �d7 tha4 bl 13 14 0-0 Jl.xe5 lle8 dxe5 .i.h6 In the event of 1 4 . . . i.xal 15 'ifxal 'ii'f6 16 'i'xf6 tiJxf6 17 i.xf8 <.t?xf8 1 8 :i c 1 White has the better ending. 15 16 17 The 'solid' 9 . . . e6 also came into consideration. On the other hand, it can not be recommended that Black should try to simplify by 9 . . . tDxe5 10 dxe5 ltJg4 1 1 ltlxd5 ltJxe5 (completely bad is l l . . .e6 12 tDf6+ ll'ixf6 13 ·'i'xd8 :.xd8 14 i.g5, Barcz.a-Trifunovic, 1 948), when White gains the advantage by both 12 ifb3 , and 12 i.g5 lDc6 13 l:k l ! 10 �xc6 l:t.c8 1!fe7 Already 1 8 e4 is threatened. i.d6 17 18 1fd4 By threatening mate, White seems to win a pawn. but although Black has only one defence, it is sufficient. 18 1ff6 Now if 1 9 1fxa7 there follows 19 . . . g5, and White loses his bishop. i.f5 9 Itel 1!fd2 llfel 19 20 • • • • • • ifxf6 t[)c5 �xf6 thd7 It was here that White · could have carried out the advance in the centre, prepared long ago. After 2 1 e4 iDxc5 (2 1 . . . dxe4 22 liJxe4) 22 exf5 ltxe 1 + 23 :xe 1 gxf5 24 i.h3 the endgame is in White 's favour. Why then didn't I play this? I imagined that in the variation · bxc6 28 admitted that it is not so easy for Black to achieve complete equality. 2 1 .. .ltJxc5 22 �xf5 Black would reply 22 . . . tiJQ3 , and that White loses the exchange! 6 .i.xc3+ a3 6 . . . i.e7 occurred and was analysed in detail in Games 1 93, 1 96 and 2 1 9. The last of these was played in my previous match with Smyslov and ended badly for him. And so he decided to readjust. Returning to Game 2 1 9, I should mention "that after 6 . . . !il.e7 7 ttJf4 d5 8 cxd5 .txfl 9 @xfl it was suggested that it was unfavourable for Black to play 9 . . . ttJxd5 10 ttJcxd5 exd5 1 1 ilh5 c6 12 ttJe6 g6 1 3 'ii'e5 .i.f6 1 4 liJxd8+ .i.xe5 1 5 lfJxf7 when White wins a pawn. However, in the game Hiibner-Tinunan ( 1 979) there followed 1 5 . . . rJ;xf7 1 6 dxe5 �d7 17 f4 t[)c5 1 8 b 4 llJe4 1 9 .i.b2 c;lre6, and Black gained a great positional advantage. It should be mentioned that in the event of 1 8 .i.d2 llJe4 1 9 <ifi>e2 h5 20 ..i.e l ! �e6 2 1 h4 ! @f5 22 @f3 followed by l:thgl and g2-g4+ it would not have been easy for Black to gain a draw. 7 tLlxcJ d5 � 21 22 ltJxd7 e4 .ilxd7 Too late. After the exchange of knights, Black's bishops have become active. 22 23 .i.aJ .i.g4 l:.cdl Draw agreed Gaine 255 Botvinnik-Smyslov World Championship Match Moscow 1957, 13th game Nimzo-Indian Defence 1 c4 2 lllc3 3 d4 4 e3 5 . lhge2 ttJf6 e6 .i.b4 b6 .i.a6 A variation in which Black hopes to compensate for the opponent's two bishops by the comfortable mobilisation of his forces. Even so, it has to be 8 b3 This move essentially signifies that ,' ! . 29 Now 1 4 . . . a6 is threatened. White is giving up the opening battle, which is transferred to tl1e iniddlegame, where �e will hope to exploit the strength of his two bishops. However, here other continuations too would not have given anything real: 8 cxd5 .txfl 9 �xfl exd5.. 8 b4 .i.xc4 9 .i.xc4 dxc4 10 'i'e2 a5 l 1 b5 4-Jd5, or 8 'i'f3 0-0 9 .te2 c5 IO dxc5 lbbd7 , 8 9 . . 0-0 • 14 .i.f3 15 16 f3 : lhg5 This is the cause of all Black' s misfortunes. The resulting position with opposite-colour bishops will be advan tageous to White, since his bishop is more active than its opposite number. 14 . . . lba5 was advisable, when Black has a comfortable position. a4 This advance is perhaps slightly premature. 9 i..e 2. as I played in the 15th game of the same match, is more accurate. 9 10 .i.a3 c5 dxc4 11 bxc4 lbc6 · Black is aumng to exploit the weakness of the c-pawn, but. of course, in the battle for the centre the d5 pawn was playing a greater role than the white b3 pawn, for which it is exchanged. A subtle move. The loss of the c5 pawn would be temporary, · since Black would easily regain the sacrificed material. by exploiting , his lead in development. White, however, has a satisfactory way of defending his central pawn. 12 16 i..e2 .. a6 lhcJ f5 • This makes it easier for White to advance d4-d5, which is more dangerous for Black than the loss of the c5 pawn. Therefore the immediate 16 . . .fS was more correct. 17 . 13 .t.xc6 It was more logical to castle immediately, but it was very tempting to shut the enemy knight out of the game. .i.b7 lhb5 Smyslov quite rightly avoids the ex change sacrifice. For example: 12 . . . cxd4 1 3 Jl.xf8 ifxf8 ( 1 3 . . @xf8 14 lDxd4) 1 4 exd4 :ds 1 5 'ib l lbe4 16 i..d 3, and Black's initiative comes to a standstill. Even so, tl1e manoeuvre with the retreat of the bishop and . . . a7-a6 involves a loss of time, thanks to which White is able to complete his development. .i.xc6! The sacrifice would not have achieved anything: l 7 . . ttJxf3+ 1 8 gxf3 . 'i'h4+ 1 9 'it>e2. . 18 19 0-0 11fd3 1ff6 llfd8 Now White e�tablishes his pawns on d5 and e4, restricting the mobility of the lhe4 30 28 enemy pieces (in particular the bishop), and also · securing him control of the centre and of the. long diagonal. At t11e same time, Black�s passed pawns on the queenside are easily blockaded. 1:'%:1, •.,! �;fe1 ,,,.,i.t; ;,� ��� 1fj%.� 20 21 22 23 d5 e4 cxd5 .i.b2 1110&! ife2 t'j)fi exd5 .i.d7 t'i)e5 29 23 . . . b5 there axb5 axb5 25 'i'c3 ) 26 'i'e2 white pawns ll'ldl lbf2 29 30 31 32 33 f4 axb5 .tcJ <it>hl l:[a2 ?:.cal :eS l:c7 llb7 h5 Now the threat of an invasion on the a-file is so unpleasant, that Black tries to complicate t11e play somewhat by sacrificing one of his passive pawns. b5 c4 33 34 35 36 The threat of 27 llld3 could also have been parried by 26 . . . 'i'd6, but then 27 �Jc 1 , and 2 7 . . . ltJc4 is met by the exchange sacrifice with a decisive advantage. 27 �fcl This move and the following one emphasise best of all the helplessness of Black, who is completely deprived of any counterplay. Black temporarily keeps his blockading knight on its centralised position, but subsequently it cannot be maintained there, since he is unable to defend the long diago·nal. 25 26 l:tac8 �� IF9' ;.� fi$M � �� After the immediate would have followed 24 ll'ia4 ll'ie5 (25 ... 'ii'd6 26 bxa4 27 f4, and the become irresistible. 3 24 'li'd2 Black tries to retain both rooks; after the exchange of one pair, White's pressure on the knight at e5 would have become even more intense. But even this subtlety is unable to prevent the further intensifying of the positional pressure by the white pieces. .txb4 .i.cJ ll'ldl b4 'ifb6 ife3 Let it be an endgame. After all, this can no longer change anything! axb5 31 36 37 38 39 40 41 :xd2 .i.d4 li)cJ h4 l:ta8+ 11xd2 liJd3 :eb8 h!b3 �8b7 in 1936 I played this as Black against Ragozin (see Training Games). Then many years later in my previous match with Smyslov (Game 228) I replied 9 . lhc6. On thi.s occasion, as in Game 229 (and even earlier in the 7th game of my 1 954 match with Smyslov), Black chooses an accurate move order. . This move was sealed� but Black resigned without resuming. What can . 9 10 he do to oppose the inevitable advance of the white d- and e-pawns? A game in the style of Capablanca himself. Of course� it was pleasant to play a game in tl1e style of the great Cuban, especially against Smyslov! ltg6 1'e3 In Game 229 White retreated· his queen to d2, which proved less success ful. 10 11 • • • lL\c6 .i.d2 In the afore-mentioned 7th · Smyslov played 1 1 dxc5, when could have followed 1 1 . . 'i'a5 good prospects for Black. 1 1 .i.d2 to more interesting play. Game 256 . Smyslov-Botvinnik "ftVorld Championship Match .Moscow 1957, 18th game 11 . • . game there .with leads lL\e7 French Defence 1 2 3 4 e4 d4 li)cJ aJ 4 5 6 7 8 9 bxc3 Vg4 ifxg7 1ih6 ll)e2 e6 d5 �t.b4 The favourable score � the match for my opponent (91h- 71h) influenced the opening of this game. Smyslov avoids the more complicated 4 e5 and chooses a variation that is quite safe for White. in which I still have to fight for equality. .ixc3+ dxe4 lL\f6 l1.g8 c5 Black needs to resolve the problem of the development of his queen's bishop, and he vacates the c6 square for it. Since from here it will defend the e4 pawn, White immediately attacks this pawn, in order to try and frustrate his opponent's plan. This opening and the given specific position were well known to me. Back 32 12 ' ltlg3 i.. d7 It transpires that Black 's central pawn is invulnerable. After 1 3 tLixe4 lDxe4 14 'ii'xe4 llc6 he achieves an excellent position. White decides to win another pawn - at c5. This leads to the tripling of his pawns, which somewhat devalues his material advantage, but even so. these pawns control important squares. 13 14 15 dxc5 c4 .ll.e2 1lc7 .i.c6 ltJg4 19 This practically forces the e xchange of White's king's bishop, after which Black can consider himself to be out of danger. The point is that after 16 'i'd4 :ds 17 'i'h8+ :gs the threats of 1 8 . . . ltJxf2, 18 . . . �xd2. or. finally, . . . e4-e3, are highly unpleasant. And after 16 'W'c3 0--0--0 Black 's position is perfectly sound. 16 17 18 .ixg4 h3 ltJxe4 l:txg4 �g6 20 21 22 The capture of this pawn cannot be delayed, since if Black were able to play . . . f7-f5, he would obtain an easily convertible pawn majority in the centre and on the kingside. And at the same time White's two extra pawns on the queenside would not play any sig nificant role. However, now the game transposes by force into an ending with opposite-colour bishops that is favourable to Black. 18 19 ·• • • . . • 'i'xd6! The basic tactical subtlety of Black 's entire plan. Perhaps White had only reckoned with 1 9 . ..tiJ xd6 20 c xd6 'i'xd6, but then he would probably have gained a winning position - 21 i.b4 'ilic7 22 0-0--0. Now, however, tlrings reduce to an ending in which, despite being a pawn down, Black's chances are better. Smyslov immediately realised this, and offered a draw, but the game continued . . . cxd6 .i.xeJ l:gl �xeJ .i.xg2 'ifild7 After 22 ... i.. xh3 23 %!xg6 hxg6 (23 . . . fxg6 24 0-0--0 �d7 25 %:thl ) Black would have lost all winning chances due to the exchange of one pair of rooks. 23 h4 White avoids 23 c5 i.xh3 24 l:txg6 fxg6. since the passed h-pawn could have caused him considerable trouble (in this variation the e xchange of rooks no longer eases his position). ltlf5 l£ld6+ The only move. If 1 9 'ife2 Q:...0--0 , with the extremely c:langerous threats of 20 . . Jlxg2 and 20 .. ltJd4. 23 • • • h5 Avoiding the capture oi1 d6, which allows White to activate his pieces. . 33 24 25 c5 llbt :ag8 i.f3 after which the f2 pawn will become hopelessly weak. Therefore Smyslov decides himself to advance his f-pawn,'· thereby forcing the opponent to switch to the calculation of concrete variations� A sensible decision, considering that for the nine moves to the time control Black had very little time left. White was threatening 26 l:xg2 l:.xg2 27 lbb7+, when it is he who has winning chances. Now, after the exchange of one black rook, the other will be very active. 26 27 28 :xg6 'it>d2 <i>d3 lbg6 e5 f6 32 29 i.d2 :g2 �e3 .tel llxc2 33 'it?d3 D.g2 Naturally, the g-file must not be conceded . After the immediat� 29 . . . l:tg4 White can reply '30 l:tb4. 30 31 f4 Black avoids continuations involving the capture of the f-pawn, and chooses a course such that the e- and f-files remain blockaded, and the white rook remains passive. After 28 . . . f5 29 �g5 there could not be any question of an advantage for Black. Now� however, the potential threat of . . . lilg4 forces White to take urgent measures. .tc6 �e6 34 fxe5 f5 Thus White is again a pawn up, but all of his pawns, arranged on dark squares, are immobile. 35 36 White has readjusted. Not wishing to return with his bishop to e3 he has put his king on this square and his bishop on c3 , which has its advantages. Now . . . a7-a5 is not possible, and hence the b4 square is. secured for his rook. But the new position has its defects. There is the very strong threat of . . . f6-f5-f4+, ltb4 <i>d4 i.e4+ � 36 ltg4 The lack of time has its effect. The not so complicated continuation pointed out by Stahlberg was· stronger: 36 ... l:h2 34 37 .iel l:!h3, and White's position is critical (38 !ta4 a6, or 3 8 a4 i.c6 etc.). 37 38 39 /Let llb2 �dJ a5 .i.d5+ l:ta4 @e2 lld2 .i.xdJ .i.b5 <&t>e6 49 50 51 .tb2 .i.al i.d7 �d5 .i.c8 52 i.ct ! Now Black has to reckon all the time with the possibility of the enemy king advancing further. Winning the a3 paWll, which, it would appear, should have led to the win of the game. 40 41 46 llxd3 47 i.ct 48 <&t>e3 49 <it>f4 l:Xa3 Here the game was adjourned. 42 :txd5 or 42 d7 is threatened, and after 4 1 . . .i.c6 42 d7! .ixd7 43 l:td6+ <i;;e7 44 .i.d2 the initiative passes to White. Therefore the sealed move is more or less obvious. 41 42 43 .i.c4+ <&t>d7 cli>f2 lld4 During my home analysis I was mitially afraid of 43 c6+ bxc6 44 l:tb2, but then I established that 44 . . . ilb5 45 �d2 (with the threat of 46 e6+) 45 . . . i.d3 or 45 . . . c5 leaves Black with real winning chances. Smyslov' s decision to exchange rooks came as something of a surprise to me, since I considered the ending with opposite-colour bishops to be won: Alas, things turned out to be much more complicated . . . 43 44 45 iid2 i.g5 Now 52 . . . <it>xc5 is dangerous for Black, e.g. 53 <it>g5 @d5 54 @xh5 @xe5 55 @gs <ittxd6 56 h5 iLe6 57 cM6 i.g8 58 'i;g7, and White wins. • • • - .i.aJ i.ct .i.e6 .i.d7 <&t>d4 55 56 .taJ i.b4 .le6 i.c8 Black's objective is to drive the white bishop off the short c l -a3 diagonal, but such that during this time the white king is unable to come to the aid of its passed pawns. Alas, in the game Black's king did not in fact manage to advance any further. ..i.b5 a4 With the unpleasant tl;rreat of 46 e6+ �xe6 4 7 d7 i.xd7 48 .J:d6+ and 49 l:.xd7. But, as I have already said, the exchange of rooks seemed to me to be advantageous, and therefore my reply was also obvious. 45 52 53 54 l:ld3 35 57 .i.aJ 58 . .i.b4 .li.e6 I found the paradoxical move 67 . . . 'itb3 ! , when I thought that Black was winning: 68 <it>g5 @c4 69 .i.b2 @d5 ! The results of a lengthy analysis were embodied in a study, which was published in 1 958. Later Euwe and Konstantinopolsky found an improve ment for the defence, and it turned out that the study was incorrect. 4 However, it is hard to say how the play would have gone, if Black had continued this way at the board. On this diagonal the bishop is ideally placed, fQ.�fllling two functions: it con trols the a3 square and simultaneously defends the c5 pawn. 58 59 60 61 62 .taJ .i.cl i.aJ i.cl .i.d7 ..t.c8 'it>d5 .i.d7 Here too 62 . . . 'it>xc5 is unavailing due to 63 'it>g5 'it>d5, and then as indicated in the note to White's 52nd move. At the board · I also examined 62 . . . .te6 63 .i.a3 i.c8 64 i.c l 'it>d4 65 i..a3 'it>d3 66 i.b2 <iiitc 2 67 i.d4 a3 68 lt.1g5 <it>d3 69 i.a l !, and White saves the game. Not finding any possibility of strengthening my position, I offered a draw, which was accepted. It was annoying not to win this game, in which. I had invested a number of interesting ideas. Returning to this ending a few months later, in the position after 67 i.d4 Game 257 Smyslov-Botvinnik World Championship Return Match Afoscow 1958, 1st game · Caro-Kann Defence 1 2 e4 c6 lhcJ My opponent made this move almost without thinking. Some theoreticians (perhaps not without justification) op pose an early d2-d4, as easing Black's defence. 2 3 • • • tLlfJ d5 .i.g4 In the 17th game of this return match I played 3 . . . dxe4 (the continuation was 4 t'bxe4 tiJf6 5 tiJxf6+ gxf6 6 .i.c4 ), which can hardly be recommended. 4 5 h3 1fxf3 .i.xfJ Regarding 5 gxf3, see Game 2 74 (Tal-Botvinnik, 3rd match game, 1 960). 5 • • • /£if6 This move is perhaps more accurate than 5 . . . e6, although then too ·the 36 1'.,]1c_vclopaedia* considers that Black equalises . 6 d3 After 6· d4 dxe4 7 'i'e3, as, for example, in · the game Fischer-Keres ( 196 1) the play is more lively. e6 6 7 .i.e2 This system of development is not dangerous for Black. Stronger is 7 a3 (as in the 19th game of the return match) or 7 .i.d2 followed by g2-g4, in both cases with the fianchetto of White's king's bishop (it is well known that the immediate 7 g3 is advantage ously answered by 7 . . . i..b4. Therefore, during the game I natur ally concluded that my opponent was insufficiently well-prepared for the Caro-Kann Defence. ti)bd7 7 8 1Wg3 A 'harassing' move - White prevents the obvious development of the black bishop on the f8-a3 diagonal. but, on the other hand, at g7 the bishop will also be quite well placed. 8 g6 Jl. g7 9 0-0 10 .i.f4 I was expecting the more energetic, in my view, 10 f4 , but White tries to solve his problems with piece play alone. This cannot prove successful, of course. since Black's position is sufficiently solid, and White's tactical two-move threats are easily parried. • . • . • 10 . . . 'i'b6 Of course, not 10 ... 0-0 on account of 1 1 .i.d6 l:.e8 12 e5, when Black has no satisfactory reply. 0-0 1 1 liabl 12 il..c7 Continuing the same tactics, in the hope that the opponent will move his queen to the wrong square: 12 . . .'ib4 1 3 .i.d6 or 12 . . . 'i'a6 1 3 d4 ! Gligoric' s recommendation also came into consideration: 12 e5 tDe8 1 3 l:!fe l followed b y h3 -h4, but at that time it was not known! 1fd4 12 . 13 li.f3 e5 Now the bishop at c7 is cut off from its main forces, and White must urgently move it to a safer place. The exchange in the centre is not un favourable for Black: if 14 exd5 he can play either 14 . . . �xd5, or 14 . . . cxdS (15 ltJb5 '11>4 16 c4 dxc4 17 tlld6 1Wc5). Ilfe8 14 .i.d6 15 .i.aJ On this occasion White creates a well camouflaged trap: 1 6 ' exd5 cxd5 17 liJxd5! t!Dxd5 1 8 c3 · ·(18 . . . 1ixd3 19 • . * Here and subsequently this refers to the Encyclopaedia of Chess Openings, published in Yugoslavia. 37 • :lfd 1 r This variation could have followed, for example, after 1 5 . . . a6. Black, naturally, avoids this contin uation. 19 • • • dxe4 • • • 11.fdl b3 W'b6 22 23 The natural reply. 18 li.d6 was dubious on account of 1 8 . . . %le6 ( 1 9 :gd3 c 5 2 0 l:.bdl c4 2 1 tiJd5 lDxd5 2 2 �xd5 lDf6 23 .:t5d2 .i.h6). 18 19 ... i.cl 'i'c7 ii.el a4 b4 ti)e6 a6 Despite his great time trouble, White acts logically: by preventing . . . b5-b4 he ensures that he can play i.e2 and then f2-f3 , after which his queen will be able to return to the centre. However, b3 -b4 involves a weakening of his queenside pawns. b5 TI1e threat of . . . b5-b4 is highly unpleasant. Incidentall�, White already had little time left on the clock! 17 18 • dxe4 White imperceptibly makes a serious error and ends up in a difficult position. It was essential to play 16 lDxe4 ! lDxe4 1 7 .ixe4 with approximate equality. My opponent evidently assumed that he had nothing to fear, and on general grounds he avoided simplifying the position. 16 • .. 20 21 22 15 ' 16 • Black incorrectly avoids 1 9 .b4 20 lDa4 lDxa4 2 1 bxa4 1i'c5, when to 22 .te3 there is the reply 22 . . . 'i'c4 ! with an obvious advantage. However, he preferred not to clarify the position, in order to make things more difficult for White, whose time trouble by this point had intensified . . . ... i.e2 .:ad8 1fe7 The e5 pawn is immune, since if 24 'i'xe5 the knight moves from f6, and the knight at c3 is lost. lDc5 24 axb5 It was hardly correct for White to hurry with this exchange, since due to the weakness of the b4 pawn, and the The manoeuvre of the bishop to e3 is inadvisable, since it will be badly placed there. 1 9 i.b2 was preferable. 38 35 need to defend it, he has to concede the open rook's file to Black. 24 25 26 l:txd8 .i.b6 axb5 :xd8 36 lta8 f3 'lfel hla3 36 37 38 Of course� not 28 tiJd I because of 28 . lhxe4. . . 28 29 . • i.h6 • Ji.fl Another lost tempo� and yet . . . one less move to be made in time trouble! 29 30 31 .*.c5 .i.d3 cxd3 11ff3 lt:\xd3 llxd3 lld2 . 39 llfl White also gives up a second pawn, to gain at least some counterplay. lbd4 11'e6 lDd7 39 40 41 e5 tElgJ 'it'xb4 ilc4 D.c2 The sealed move. The exchange of rooks is unavoidable, and a prosaic ending with a material advantage is reached by force. 42 43 Now the exchange of White's dark square bishop is unavoidable, and his game becomes strategically "lost. ii.xd4 f5 e6 l:.cl fxe6 43 . . . l:.xfl+ 44 liJxfl .i.g5 45 fxg6 'i'xe6 46 gxh7+ <it>xh7 suggested itself, but my second Grigory Goldberg rightly pointed out that then White would gain some chances with 47 liJg3 ! By control ling the f5 square, the knight might yet prove dangerous. Therefore it was decided to take play into a queen ending. 44 fxg6 l:xfl+ �i1J0 : ti � %� exd4 i.eJ+ tDe2 ltle5 �bl On this occasion Black too does not disregard some two-move threats (35 . . lhxf.3 and 35 lhxd3). . f4 39 e5 is easily parried by 39 . . 'i'd5. .'{:.:>..:1 32 33 34 'lld6 White decides to give up a pawn. Hardly any better was passive defence 36 'i'd l � when Black, naturally, would have continued the attack with 36 . . . l2Jc4. A pointless move. At b6 the bishop is worse placed than at e3 . 26 27 28 "i'fl 35 . . . ltJxd3 36 cxd3 l:xd3·was prema ture on account of 3 7 ttJf4, but now Black has renewed the attack on d3. 45 46 l£lxf1 1ff6 hxg6 b4 It is hard to find an antidote to the advance of this pawn; · The capture of the g6 pawn would merely lead to the black ... 39 king slipping away to the queenside. Also bad was 47 1Wd8+ @f7 48 'i'd7+ 'iti>r6 49, ,'ifd8+ 'it>e5 50 'i'h8+ �e4, and nothing would be changed by 47 lDg3 g5 48 ft)h5 'ifcl+ 49 <i>h2 Jlf4+. Therefore White's next move is the strongest. 55 1if6 1fd5 The queen must be centralised! 56 57 58 59 60 1ff3+ 1id1+ 1'e2+ 1fa6+ 1i'a7+ Cit>d4 ci;;e5 <ifld6 ci;;e7 ct>f6 And again there are no more checks. 61 62 1fh7 ci;;b l 1fe5+ b2 Only one more step needs to be made, and the black king can .always avoid possible checks. White resi gn s . Game 258 Botvinnik-Smyslov 47 �h2 g5! Even so the laright is WUlble to come into play, . since if 48 lhg3 there follows · 48 . . . J..f4. The exchange of minor pieces is forced. 48 49 50 51 lDxeJ 11'xg5+ 'ifxe3 1ie5 dxe3 1fc7+ 'lfb8 bJ 4 5 . .• • • d4 c4 tLlcJ lDf6 g6 J..g7 e4 f3 d6 The Samisch Variation is one of th: most active continuations: Black has t• counter definite problems. cs �g6 5 6 . . • .tel 0-0 a6 Smyslov quite often carried out the · flank attack involving . . . b7-b5, an idea that also occurred in the games of other · players. The checks would also come to an end after 5 3 1i' g3 + <it?f5 . 53_ 54 1ff8+ 1 2 3 In the previous matches Smyslov was content to employ the Griinfeld Defence, but here he goes in for a more complicated opening. rbf7 White refrains from advancing his pawns; in this case his king would merely be exposed. But equally, his lone queen is unable to stop Black's passed pawns. 51 52 53 World Championship Return Match Moscow 1958, 2nd game King's Indian Defence �f5 <ifle4 7 40 .i.dJ' lLlc6 The usual continuation here is 7 . . . c6 followed by . . . b7-b5, as, for example, was played against me by Smyslov . in the 6th game of the match (No.259) and by Larsen (No.380). The move in the game leads to more complicated play, and would appear not to have occurred previously. 8 tLlge2 Now Black has to do something active. Waiting tactics will lead to an obvious advantage for White. 10 • • tLla5 • Smyslov decides to play his knight from c6 to c4. It would hardly have been any better to use the d7 knight for this, e.g. 10 . . . b5 1 1 cxb5 axb5 12 b4 lbb6 13 ..ta2 ltJc4 ( 1 3 . . il.d7 14 'i'd3) 14 .i.xc4 bxc4 15 'i'a4. llb8 . 11 .la2 b5 Essential: any other move would have been answered by 1 2 b4, driving back the enemy knight. 12 13 14 9 cxb5 b4 clxc4 axb5 lDc4 bxc4 a3 White falls into a slight psycholog ical trap. He prevents 9 .t[)b4 10 ii.bl b5 . Meanwhile, the exchange of the c6 knight for the d3 bishop is not dangerous for White, and for Black it involves a loss of time. 9 0-0 or 9 1fd2 was simpler. However, the further course of the game )\ shows that Black cam1ot extract any 1� gains from the weakening of the b3 .. Here White had to stop and think. He would appear to have a clear advantage: on the queenside he has a passed pawn, whereas tl1e black c4 pawn is securely blockaded. But Black' s two bishops are a latent strength, which could be revealed if he were able to play . . . e7e5xd4 followed by . . . ti)e5-d3 and . . . c7c5 . Then not a trace of White's advai1tage would ·remain. And if (after 1 5 . . . e5) White chooses d4-d.5, there square. 9 10 tLld7 il.bl The bishop had to retreat, in order to defend the d4 square (in view of the threat of 10 . . . e5), and to a square from where it will be . transferred to a favourable post - a2. 41 follows another manoeuvre: . . . f7-f5-f4 followed by . . . g6-g5-g4 with a dangerous att.ack for Black. Finally I managed to find the correct pl� bnt I was careless in the way that I carried it out 15 compensated by the weakness of the c4 pawn, and also by the fact that the game is opened up, and the position of the black king proves insecure: 0-0 I was sure that Black · would play l 5 . . . e5, and had prepared 16 b5, when after 16 . . . exd4 17 lDxd4 ttJe5 1 8 'ifd2 he does not have 18 . . . c5. However, my opponent finds a stronger continuation. Therefore White should have played 15 b5 inunediately. 15 16 .d2 c6 ttlb6 Black, in tum, misses the strongest move. and aft.er the exchange of the dark-square bishops his position goes sharply downhill. 16 . . . 1le8 was essential. 17 18 .i.h6 1rxh6 23 24 24 . . . ltf5 looks more subtle. Indeed, after 25 tiJd4 .:es 26 'iff2 .i.b7 27 ltJc6 ..txc6 28 dxc6 d5 Black would have gained definite counterplay. However, in the variation 25 it'd4 .ib7 26 ttJf4 (26 . . . g5 27 �5 or 26 . . . 'i'f8 27 l:tfl) the d5 pawn is defended. and the c4 pawn remains under attack. 25 1%.fl And now, since the d5 pawn is indirectly defended (25 . .. ttJxd5 26 Ve6+, or 25 . . . i.xd5 26 tDxdS lDxd5 27 'i'e6+), White prevents the activation of the black rook. .i.xh6 f6 A useful move., guaranteeing the safety of the black king. 19 20 a4 I:tfbl tl)a8 Now, when White's 'king's rook has been diverted to the queenside, Black tries to seize the initiative on the opposite wing. 20 21 22 1fe3 fxe4 f5 fxe4 Cjjc7 22 . . . 'ifb6 was more circwnspect. for the moment preventing White from creating two connected passed pawns. 23 cxd5 .i.b7 exd5 25 26 • • 'ii d7 • 11d4 26 'i'a7 was tempting, with the threat of 27 l:txf8+, but White preferred a simpler continuation. d5 . This move . seems to give Black counterplay due· to the weakness of the white pawn at d5, but this is more than 26 27 42 . dxe6 e6 ltlxe6 It is now too late for 27 'ifxe6 28 �f4 'i'e5 . on account of 29 l:ad1, when Black has everything 'hanging' . ... 33 34 11'xc4+ 'ifxc6 d5 . l1d8 35 36 37 1ib6 1fd4 llfel 1le7 11fd6 Black could not leave the eighth rank unguarded (34 .. J�xb4) in view of 3 5 ctJxd5. In the end White allows . t'De6 and the advance of the passed d-pawn, but the white pawn proves to be quicker. For the moment it is useful to reduce the number of pieces on the board. . 37 38 39 Ilxe5 b5 . llde8 lixe5 lhe6 2 8 1ig4! This is, of course, stronger than 28 'i'xc4 d5, when Black has some com pensation for the sacrificed pawn. 2s ... 30 nadl :res Black goes in for all sorts of tactical tricks, to try and eliminate the block ading knight at c3 (as will be seen from what follows) . 1!fg7 29 ltld4 This is simpler than 30 ttJxe6 'i'xc3 . 30 • • • liJc7 Smyslov did not like the endgame position arising after 30 . . . li:Jxd4 3 1 'iWxd4 'i'xd4+ 3 2 %!xd4 d5 3 3 b5 . 40 'lra7! Avoiding a trap - 40 ltle4? l:txe4 4 1 'iixe4 'ifc5+. In a difficult position Smyslov over looks the loss of a pawn. Or 40. . . !lhS 41 g3 (4 1 . . . li:Jf4 42 l:fl ). 41 ltJe4 When this sealed move was revealed, Black resigned. The reason: 4 1 'i'f8 42 b6 ttJ!4 (or 42 . . . �d8 43 b7 lDc6 44 'iib6) 43 b7 Jil.e8 44 b81i' lbb8 45 ltJgS . 31 32 1lf4 ltlc6 :es 40 .ixc6 The mutual pins after 3 2 li:Je6 3 3 'iYxc4 l:k8 34 l:.�dq i are no danger to White, and two pawns have already been won. ... • . • d4 . . . 43 Game 259 Botvinnik-Smyslov c4 The simplest way of avoiding the Griinfeld Defence! 1 • • • g6 Smyslov also played this in the 1 9th game of our 1957 match. However, this move has the drawback that White can now transpose into the King's Indian Defence. ll.g7 2 e4! 3 4 d4 ltlc3 .i.e3 f3 8 9 10 11 .i.xc4 JlbJ ttlxe4 b5 bxc4 d5 dxe4 d6 a6 Black was evidently not very happy with the way the opening developed in the 2nd game (No.258), and he does everything possible to deviate from the Samisch Variation. However, after White's ne�i move it cannot be avoided. 5 6 .i.dJ 1id2 In a similar situatidn· Sokolsky played b2-b3, which is dubious in view of the reply . . . c6-c5 . Modem theory recom mends 8 e5 and then 9 f4, which leads to a sharper game. World Championship Return A1atch A1oscow 1958, 6th game King's Indian Defence 1 7 8 A mistake, which was prepared . . . in my home analysis ! There I established that in the event of 1 1 fxe4 e5 1 2 dxe5 White did not achieve anything, and the obvious move 12 CD:f3 ! , after which White has a clear lead in development, was not even considered. Now Black has a reasonable game. liJf6 c6 In the 2nd game Smyslov did not play this. but preferred to develop his , queen's knight at c6. However� it has to be said that the set-up with . . . c7-c6, . . . a7-a6 and . . . b7-b5 is more soundly based. It can be mentioned, for example, that this plan had occurred in games by Taimanov, and also in the game Sokolsky-Petrosian ( 1 957). And, as already mentioned, Larsen played this against me (No.380). White basically follows Sokolsky' s plan: he does not prevent . . . b7-b5xc4 followed by . . . d6-d5, but aims for tl1e rapid development of hi� pieces. 11 12 . .. ltJe2 0-0 a5! Well played. 13 0-0 Here castling is a Joss of time, which hands Black the initiative. 1 3 tlJ2c3 CZJxe4 14 fXe4 .i.a6 1 5 0-0--0 was logical, although . it would have led to double-edged play. However, it is not so 44 easy to decide on such a continuation in a match for the World Championship, when you already have an advantage in points . . . 13 14 15 i.c4 l:t.acl a4 /t)bd7 l:tb8 Black's position is the more active. The weakness of the d4 pawn and the d5 square are balanced by the similar defects of the c6 pawn · and the c5 square, but what compensation does \Vhite have for the weakness of the b2 pawn and the e3 square? So that the e3 square should not be soon attacked, White exchanges one pair of knights. It is hard battling against such a subtle positional player as Smyslov, but subsequently I was able to surpass myself 16 lDxf6+ .ltxf6! 17 18 19 lDcJ i.e2 l:lfdl lDb6 .ie6 The white knight is tied to the defence of the a2 pawn, and a2�a3 cannot be played, since this wouid weaken the light squares. There would follow . . . i.e6-b3, when the d4 pawn would be in danger, to say nothing about the possibility of continuing the manoeuvre with . . . .ib3-c4, fixing the weakness of the light squares. Therefore 19 . 'i'd7 ! and 20 . . . :fds, intensifying the pressure, suggests itself, after which White's position would have remained difficult. .. 19 • • • .i.g7 It really would have been better to give the opponent the move with any non-committal manoeuvre, than to re treat the bishop to g7. White, of course, could only dream of exchanging his passive bishop for the opponent's active dark-square bishop, and only this mis take by Black allows this dream to be realised! Incidentally, now the weakness of the c5 square will be more sensitive. It is curious that Smyslov made the same mistake in a similar position in the 2nd game of the return match. The remaining knight is needed on the queenside. 20 21 i.h6 'ifxh6 .i.xh6 f6 22 23 ltd2 h4 .i.f7 Smyslov used the same method to defend his castled position in the afore mentioned game. Preventing . . . g6-g5, just in case. 23 24 • • • al! W'd7 This move proves possible, since 24 . . . .i.c4 25 thxa4 i.xe2 26 thc5 'i'd5 2 7 1he2 'i'xd4+ 28 · �h 1 is clearly in White's favour. After all, the black Black _ has played very subtly, and now his position is clearly preferable. 45 During the game I thought that after 30 . . . l:td6 (it was this move that was later suggested by the commentators) 3 1 l:.dc2 �e8 3 2 i.b5 or 3 I ..J:lb8 3 2 'ittf2 (32 . . . i.b3 33 :xc6) Black also had a difficult position. However, analysis showed the possibility of a defence in the variation 30 . . . l:td6 3 1 ztdc2 �e8 3 2 .ib5 by 32 . . . �d7 ! 3 3 tiJxd5 tiJxd5 3 4 ..txa4 tiJb6 ! 3 5 .i.b3 :xd4. pawns are fixed on light-squares, the same colour as the bishops, which is a highly unpleasant indication for the endgame. As a result of 24 a3 ! the white knight is freed from having to defend the a2 pawn, and the complex of weak squares (b4, c5) and pawns (a4, c6, e7) is in jeo pardy. All this is the direct consequence of the unfortunate move 1 9 . . . Ji..g7. 24 25 26 � i.fl :fd8 11'e8 .i.d5 A natural continuation, since other wise Black cannot consolidate his position. 26 . . . J:td5, for example, was hardly any stronger in view of 27 1if4, with the unpleasant threat of 28 '&c7. 27 lDc5 'iff8 In the endgame, at least, Black will easily be able to cover his weaknesses on the e-file, but White too, true to his match tactics, does not avoid a favour able ending! It was not so easy to decide on such tactics for the match, since Smyslov is a supreme expert in the field of the endgame . . . 28 29 'ilxf8+ lDa6 31 32 33 The alternative 3 3 tiJb4 looks less well-founded, since after 33 . . . e5 34 d5 rJJe7 (if 34 . . . ii.xd5 or 34 . . . lDxd5, then 3 5 ..ib5 ! ) 3 5 i.b5 f5 Black, by retaining control of the b3 square and blockading the d5 pawn, would have had better chances of a successful defence. <ifi>xf8 Beginning pressure on the c6 pawn. 29 30 • • . lDb4 1bc6 �xc6 �a5 :bc8 i.b3 In this way Black prevents the doub ling of the white rooks on the c-file and e:\.'J)loits the undefended rook at d2 to create tactical threats. Even so� this counterplay is insufficient compensation for the pawn that he sacrifices. But was there anything better that he could have done? . 33 . • . .i.a2 For the moment 33 . . . e5 did not achieve anything for Black: 34 tDxb3 axb3 3 5 l!d3 ! , and White, by gaining a tempo for the defence of the d4 pawn, retains the initiative. After 3 3 . . . ..i.d5 Black would have . altogether given up his last counter-chance - advancing 46 40 41 42 . . . e7-e5. Now he could have hoped for 34 @f2 e5 3 5 �e3 ttJd5+, but because of White's knight manoeuvre the situation becomes less favourable. 34 35 36 lDb7! lDc5 llle4 :.xd4 tLlxf6 lDxh7 42 43 44 lDg5+ llle4 <it>e6 <il>d5 If now Black avoids the exchange, he will be forced to play a prosaic ending two pawns down, whereas after the exchange White's material advantage is reduced. Even so. it is possible that, had Smyslov foreseen White' s 46th move, he would have rejected the win of the pawn. However. it is appropriate here to once again remember Tarrasch' s saying: in a bad position all moves are bad. It is interesting to note the except ional activity of White's king's knight, which has made 15 moves following the route g l -e2-c3 -e4-c5-a6-b4xc6-a5-b7c5-e4xf6xh7-g5-e4. In the end it peri shes on the field of battle, but not with out first securing victory for its army! l:txd4 exd4 cl;e7 i.bt 44 45 With the unpleasant threat (for example, after 40 �) of 40 . . . d3 4 1 lllg 5 (4 1 �el lllc4) 4 1 . . .d2 42 �e2 .td3+! And White still has to make one more move in severe time trouble. This is Black' s main "trump'. 40 .ie2 . l:td5 e5 Black gives up a second pawn, but diverts the knight further from the centre and activates his d-pawn. 39 /i)d5 llleJ Since the d4 pawn can no; longer advance (the knight is left undefended), this move is very useful. With two threats - 37 ltJc3 and 37 CDxf6. Black exchanges rooks, in order to remove the blockader of the passed d-pawn that he now creates. 36 37 38 <i>f2 .ta6! Now the c4 square will be under control even afteL . . . d4-d3, and as a result Black's counterattack is repulsed. 47 fxe4+ i.xe4 '&t>xe4 �6 g4! This fearlully strong move was foWld during analysis by my second Grigory Goldberg. After 46 . . . d3 4 7 .i.f3+ <it>f'4 48 h5 gxh5 49 gxh5 d2 50 h6 the h-pawn cannot be stopped. 9 h5 d5 46 �4 47 h5 gxh5 <"bg5 48 gxh5 49 w Black resigns, since the a4 and d4 pawns are doomed. A technically diffi cult game! Smyslov-Botvinnik World Championship Return Afatch Afoscow 1 958, 7th game Sicilian Defence 1 2 3 4 5 6 7 e4 lLlfJ d4 li)xd4 �cJ .i.e2 ' c5 lLlc6 cxd4 tLlf6 d6 g6 .t.eJ Yefim Geller rightly considered that this quiet continuation gives White no advantage. 7 8 h4 10 hxg6 10 11 12 exd5 lDxd5 The afore-mentioned 5th game con tinued 10 . . . hxg6 1 1 exd5 tLlxd5 12 t2Jxc6 bxc6 1 3 lLlxd5 if'xd5 14 'i'xd5 cxd5 . White, s advantage is so insig nificant that, not expecting a repetition of this variation, I somewhat casually prepared for this game. But at the board it became clear to me that 10 . . . hxg6 . 1 1 tLlxc6 bxc6 1 2 e5 favours White, e.g. 12 . . . t2Je4 1 3 ltlxe4 dxe4 14 i.d4 'i'as+ 15 i.c3 'ifd5 16 ifc 1 . Therefore I had to 'change tune', in order to avoid a plan that had most probably been prepared by my opponent in the quiet of his study. Game 260 .i.g7 Smyslov repeats the opening from the 5th game of the retum match, and from what follows the reader will understand what, in all probability, my opponent was hoping for. 8 . .. 0-0 Stronger here is 8 . . . h5, as I played in the ne�1 odd-numbered game (No.26 1 ). fxg6 li)xd5 The surprise effect operates. White makes an error, after which he runs into difficulties, whereas Black has an easy game. Meanwhile, 1 2 i.c4! e6 13 t2Jxd5 exd5 14 .i.b3 liJxd4 1 5 .i.xd4 :es+ 16 <it>fl would . have definitely been in White's favour! . 12 48 • . . 'ifxd5 variation 1 9 l:lc l 1i'e6 20 b3 l:tf8 is also in Black's favour. How did I mi ss this po·ssibility? What Threatening by 1 5 'ilb3 W'xb3 16 told here was my old 'illness'"" """ weak ltJxb3 to obtain a favourable ending, so ness of combinative vision. During the that Black has no choice . game · I considered the exchange sacri thxd4 14 fice only after the preparatory exchange 15 cxd4 of queens, overlooking after 16 . l:xf3 This reply is also forced: after 1 5 17 'i'xc4 tl1e zwischenzug 17 l:xe3+ . .itxd4 ltd8 it i s hard for White to find a good response. 13 14 · 1fc4 .i.f3 c3 . . . • • · • . . . . . 15 • • • i.e6 Now 1 5 . . . :d8 16 Wcl ! 'i'xcl+ 17 �xc 1 .i.xd4 18 i..xd4 Jlxd4 19 :c? was not dangerous for White. Here my opponent offered a draw, but, in accordance with the rules then in force, I asked him first to make a move Game 261 Smyslov-Botvinnik World Championship Return Match Moscow 1958. 9(h game Sicilian Defence . 16 11'b3 1 e4 2 tllf3 3 d4 4 · l0xd4 5 �c3 6 .i.e2 cs lLlc6 cxd4 lLlf6 d6 6 7 .i.e3 8 h4 g6 JL g7 .· I think, · nevertheless, that the development· of this bishop at c4 is more promising for White. Since analysis of the 5th and 7th games (the latter No.260) had shown that White could have gained a clear advantage from the opening, Smyslov aims to play this same variation again. - Black accepted the offer, and, as soon transpired. incorrectly. From my old friend Abram Model I learned that by 16 . l:txf3 17 gxf3 'i'c6 1 8 'i'dl i.d5 1 9 :h3 'i'e6 Black would have won the a2 pawn for the exchange. and then, by creating a second passed pawn on the queenside and a:ttacking the poorly protected enemy king, he would have gained excellent winning chances. The 8 . . . . • b5 Obligatory! 8 0--0 9 h5 d5 10 hxg6 proved to be not in Black's favour: either the h-file or the a2-g8 diagonal is opened. . . . 9 f3 10 1id2 49 . 0-0 Of course, the. . , opponent's attack could have been eli.minated by 1 3 . . . f5, but then, thanks t · hls · control of c5, White would have had an obvious advantage on the queenside,' 'without any counterplay for Black. It would be interesting to test 1 0 tl)b3 . . .io . .. � d5 Black has no other active counter play. 11 12 thxc6! bxc6 e5 Now White's basic plan is revealed: it is to play f3-f4, hindering the opening of the centre, and then to make the bold pawn thrust g2-g4, attacking Black's castled position that has been weakened by . . . h7-h5. 12 . • • �e8 After 1 2. . . t£xi7 13 f4 J6 Black would have to reckon with the further advance of the e-pawn. Therefore I chose an apparently less active retreat. Besides, in this case it is easier for Black to calculate variations: the plan of counter play against the centre by . . . f7-f6 is obvious. 13 f4 . . • • • 0-0-0 14 15 16 fxe5 g4 17 18 .i.xg4 hS fxe5 .i.xe5 .i.xg4 Black accepts all the sacrifices, but in the process he aims not to fall behind in development, and for this reason he takes the pawn with the bishop. The e-pawn cannot be given up, since even 13 .i.116 would have been met by 1 3 . .i.xe5 ! . 14 .i.xf8 'iti>xf'S with an easy game for Black. 13 14 During the game this seemed to me to be the most subtle and energetic continuation, but in fact it allows Black to seize the initiative. As for the alternative 14 exf6 trucf6, it seemed harmless for Black. A definitive evaluation of the position can be given only by a practical testing of the variation 14 g4 ! hxg4 1 5 0--0-0 ; which may prove favourable to White. It is interesting that this recommen dation, which I published soon after the return match, was tested a quarter of a century later in a correspondence game Omelchenko-Heemsoth. There followed 15 . . . fxe5 16 fxe5 'iia5 17 hS i:xe5 18 hxg6 .i.xc3 19 l:hS+ ! .i.xhS (after 1 9 . . . <ii>xhS 20 .i.d4+, in view of the inevitable 2 1 'i'h6+ it all ends in mate) 20 Wxa5, and White won. After 14 g4 Black should probably reply 14 . . . fxe5 1 5 fxe5 .txe5, with unclear complications. . hxg4 g5! Black must preserve at any cost the white h-pawn, since_ �n it will become f6 . 50 22 Jbg4+ �h8 This is where the h5 p�wn comes in · useful ! the only pawn covering . . . the black king! After . t�mporarily strengthening in 1 his way his ·castled position� and with his control of the centre and also his extra pawn, Black will laWlch a counter offensive. Now it appears that, in view of the inevitable (immediately or later) . . .i.f4 Black will simplify the game, retaining his material advantage. 19 .i. xg5 it'd6 23 . . . 20 �bl Accurately played - Smyslov is in his element. Of course, the immediate 23 l:tg6 was bad in view of 23 . .t.f4, but now even after 23 . . . .tlab8 it is possible to play 24 l:g6 ! After 24 . . . llxb2+ White does nof accept the sacrifice (25 'iti>xb2? .i.xc3+ 26 1fxc3 l:tb8+), but wins by 25 tlra l ! Therefore Black must gain control of g6. ltg8 23 Now Black could have hoped for the better endgame after the obvious 24 l:r.g6 .tlxg6 25 hxg6 r:i;;g7 26 llh l %lh8 27 :xhs �xhs 2s 'i'h6+ <it>gs 29 'ilh7+ @fS, but here too Smyslov finds a remarkable method of defence. :h4 • After lengthy reflection White finds a plan that saves the game. First of all, it transpires that he has no reason to fear the exchange of the bishops and even the queens. After 20 . . . .if4 2 1 .txf4 'i'xf4 22 1ixf4 (or even 22 :el ) 22 :xf4 23 �e2 l:a4 2 4 b 3 :xa2 2 5 �bl :a6 26 1=txg4+ �h7 27 l:tg5 or 27 �dg l Black has no advantage. How ever, this variation was not part of my intentions. I was aiming to return the extra pawn, but in the process to force the advantageous exchange of bishop for knight. • • ... 20 21 . • • .i.xf6 24 :lb4! White defends against the combined attack on his king along · the b-file and the al-h8 diagonal (24 l:ab8 25 l:.b3 ), the b2 square is ·securely defended, and he obtains a good gam�. It should not be forgotten that the position of the black ... ti)f6 1'xf6 51 king is less secure, and knight and queen (after the exchange of rooks) are nonnally more dangerous than bishop and queen . . . Therefore Black takes the logical decision to exchange the minor pieces, to spoil the opponent's queenside pawns, as well as one pair of rooks, to gain play on the b-file. It is extremely doubtful whether the alternative plan - 24 ... ltg3 25 �2 ltg2 26 1ie3 - was more justified. 24 • • • a5! By creating the possibility of the further advance . . . a5-a4, Black does not allow the enemy rook to establish itself at b3, and White will be forced to concede the b-file. 25 ltb6 26 bxc3 .i.xc3 29 30 •eJ a3 ltg4 The immediate 30 'ifilb2 could have been met by 30 . . . ltb4+. l:e4 JO 31 1i'd3 32 lifi>b2 lteJ 33 34 •d4 cxd4 'ffxd4 • • • .e5 26 tLlxdS was tlrreatened, but this merely hastens the implementation of Black's plan. Also after 3 1 . . . ifd6 32 <i>b2 lta4 33 Ila 1 it is not apparent how Black can strengthen his position. The .match situation forced White to avoid simplification if possible, which is why he did not play 26 'iixc3 . The threat of 3 3 :g 1 is unpleasant, and Black forces the exchange . of queens. 26 27 28 . . • ltxb8 <i'at! ltab8 ' lbb8+ ci>g7 34 . . . e5 was simpler and stronger. 35 %1.gl+ t:Rf7 Here the king is safer than at c 1 . Now Black's possession of the b-file no l:th3 36 h6 longer gives him anything. White is 37 llg7+ �6 J:lb4 threatening to play 29 :g1 with 38 J:lh7 counterplay, which Black prevents. White now stands slightly more 28 llg8 actively, and Black goes for a repetition This position, with its insecure kings · of moves. 39 'iftcJ and mutual pawn weaknesses, must be considered roughly equal. Before ex If 3 9 c3 there would have followed changing queens it is dangerous for 39 . . . a4, while if 39 :th8 %Ixd4 40 h7 Black to advance his e7 pawn, as this 'l;g7. J:lh3 39 would expose the seventh rank. · • • • • • • : 52 ilh4 40 'it?b2 Draw agreed:' A good game! Game 262 Botvinnik-Smyslov World Championship Return Match Moscow 1958, 12th game Reti Opening 1 2 c4 gJ 2 3 4 thfJ bJ 11 1fc2 11 12 .i.c3 Following the example of Reti, here White could have employed the manoeuvre 1 1 l:kl-c2 followed:by Wal (�a2 has also been played, with the same idea of allowing the queen into the comer). But he decided to follow the example of. . . Smyslov, perhaps a rather naive psychological ploy. As a result, Black handles the opening very confidently! /i)f6 . White avoids 2 �c3 , to give some variety to the play. Without much hesitation, Smyslov transposes into a well-known variation of the Reti c >pcning, since theory quite justifiably considers it to be safe for Black. .i.h7 c6! d5 .i.f5 This was played by Emanuel Lasker w i th Black against Reti back in 1924. .i.g2 .i.b2 7 ·0-0 8 d3 5 6 e6 lhbd7 h6 i.e7 A similar plan (White intends b3-b4 or 'ib2) was carried out by Smyslov in a game with Durasevic ( 1956), but there the plan was fully justified, since Black had developed his queen at b6 (instead of 1 1 . . . .i.h7). Here, on the other hand, Black has the possibility of advancing his b-pawn and obtaining satisfactory play. However, the other continuations that have occurred (12 tLle5, 12 cxd5, 12 l::tadl, 1 2 J:acl) have also not achieved anything significant for White. In the afore-mentioned game the hi s hop was developed at d6, but subsequently it was shown that the 1 1 1 1dcfended bishop on this square merely aids White's e2-e4. Therefore fo r a long time now . Black has been pl aying 8 .. .i.e7, or occasionally x �c5, to provoke 9 d4. . . . .. 9 10 �bd2 a3 0-0 a5 Too direct. A more subtle contin uation occurred in a game Panno Karpov ( 1 973) - l'O ... .i.h7 1 1 b4 a5, and B l m.:k seized the initiative. 12 • • • b5·? It should be mentioned that in a game with Hartoch ( 1970) Polugayevsky 53 played 12 . . . 'iibs , and after 1 3 ifb2 1 3 . . . li.d6, also achieving a satisfactory position. · . 13 14 cxb5 b4 • • . 9b2 li)bJ! 17 18 19 axb4 llxal i.e5 21 22 l:bct+ 22 li)xcl • • • li)e8 A careless move, all the more surprising for the fact that Black spent a. long time considering it. Simple and good was 22 . . . 'i'cS (or 22 . . . 'i'b7) followed by . . . ttlf6-d7-b8-a6 (or c6). In this case White's queen's bishop would have been driven from e5 without any weakening of Black's pawns, and the pressure on the b4 pawn would have intensified. Now, however, it is rather White who will have some initiative. Wc7 Black gains an important tempo by threatening i5 . . . l:tfc8. 15 16 . 17 llc8 Essential, although it involves a significant loss of time. If 2 1 .i.h3 , then 2 1 . . . 'i'b7 ! (with the threat of 22 . . . ttld7) is highly unpleasan� and since Black controls the c-file, the b4 pawn will be more difficult to defend. cxb5 A serious error. By preventing . . . b5b4 in the most primitive way, White is saddled with a weak pawn at b4 without any compensation, and .the initiative passes to Black. It is surprising that, although I expressed this opinion long ago, Psakhis, annotating a game of his with Ubilava in lnformator No. 37, attaches an exclamation mark to 14 b4 without any substantiation, and eval uates the resulting position in favour of White. This episode once again demon strates that modem grandmasters are not interested in games from the past. It is also clear that chess information should be created with the help of computer technology. Meanwhile, by continuing 14 i.d4 (in order to answer 14 . ..b4 with 15 a4 !) and if 14 . . . ttJe8 - 1 5 i.h3, White would possibly have retained some advantage. 14 20 . ifd2 21 l:lcl li)b6 W'd7 The best way out. With the help of a tactical trick, White forces his opponent to clarify the position. 23 axb4 l:lxal li)a4 li)d4! This move proves possible in view of the fact that 23 . . . f6 24 i.h3 is extremely dangerous for Black. He should have reconciled himself to the fact that it was not possible · 10 drive the bishop at e5 If 1 9 . .. i.xb4 White would not have played 20 i.xf6, but 20 'i'd4 !, winning a piece. 54 from its centralised position, and played, for example, 23 . . . /£Jc7, when all would have turned out well. But Smyslov is so canied away by the plan of driving away the bishop by . . . il-f6, that he undertakes artificial manoeuvres, merely in order to achieve . . . a serious · weakening of his own position ! 23 <i>f8 Renewing the threat of 24 . . . f6. • Now parried. .i.hJ 25 26 lLldb3 .i.al! . . . ti-f6 .i.g8 can no longer be f6 26 .i.d4 was worse because of 26 . . . 'ifc7 ! The truth is that there are dt�fects in Black' s pawn formation. After the inevitable d3 -d4 White's control of c5 will be highly unpleasant for Black on account of the weakness of his e6 pawn, whereas the occupation of c4 by a black knight is not dangerous for White. Incidentally, when White establishes a knight at c5, this will also automatically solve the problem of the defence of his b4 pawn. . . • d4 1fa7 lLld6 Many commentators criticised Black for this move and suggested playing 27 . . . 1:£Jb6. This recommendation would not have changed the evaluation of the position. As the reader already knows, due to the. weakness of the e6 pawn and the inevitable invasion of his knight at c5, White stands better. 28 29 adv�tage. This knight move is prema ture. I was afraid that 29 t'Dd3 would be met by 29 . . . li'a6 !, but then 30 .i.c3 followed by .i.el would have retained all the advantages of White's position. • • 24 26 27 t o rise to the occasion and lets slip his 'lfa2 lLlc5 29 30 • • • .i.xc5 dxc5 Alas ! Initially White had been intending to play 30 bxc5, but then he had to reject this in view of 30 . . . 'i'a5 ! 3 1 � tDd2+ 3 2 �g2 t'Dc4 (or 32 . . . tDe4) 33 t'Dd3 'ifd2, when the white bishop at a l is in danger. But now . . . e6-e5 is llllavoidable, and the game becomes equal. JO . . • e5 Over-hasty. Simpler was 30 ... 'i'fi! (3 1 'i'c2 'ifb5 32 .ig2 e5), since the black queen is already needed on the kingside. Now White again seizes the initiative. 31 t!lbl The queen will take up an active position at f5 . lLlc4 31 After his error on move 1 4 White has achieved much, but now he again fails • • • d4 It would have been more prudent to refrain from opening the hl -a8 diagonal. 55 , 1fc7 32 1lf5 Of course, not 32 ... lllxc5 in view of 33 'if c8+ c3ilf7 3 4 bxc5 'ifxal 3 5 .i.e6+ <it>g6 36 We8+ with inevitable mate. 33 was still unfavourable in view of 3 9 .i.xc3 dxc3 40 'i'al c2. .4. 1. J1i'c l . . But this entire variation is an obvious delusion. This became · known to me a few months after the game, when in November of that year I was playing in the Dutch town of Wageningen. In a conversation with Lodewijk Prins, the Dutch master told me that back on 1 7th May he had published a newspaper article entitled 'The height of chess skill', in which he had given an analysis of this highly interesting position. �d3 Bringing the last piece into play (not counting the bishop at a l , which, like the knight at a4, is firmly shut out of the game). 33 'i'd7 'i'xd7 34 ..i.xd7 would have been a mistake on account of 34 . . . lDa3 3 5 c6 �e7. 33 • • • .i.f7 3 3 . . .'i'c6 appears to be stronger, but this could have become dangerous after 34 f4 (34 . . . t'De3 35 t'Dxe5 !). 34 1t'h7 Of course, the black king must not be allowed to move to a safer position at g8 or h8. 34 · .• .i.g8 In time trouble Black plays un inventively, but it is already hard to offer him any good advice. Also after 34 . . . 'ile7 White would have occupied the long diagonal with his queen. 'it'e4! 35 Now the white queen breaks into the enemy position. 35 36 37 11a8+ .i.g2 However, first it should be mentioned that 3 8 f4 is necessary, since otherwise it is impossible to open up the position of the enemy king and to bring into play the bishop at a 1 . But at the same time f2-f4 weakens the a7-g l diagonal, and to exploit this Black should have immediately played 3 8 . . . t'Dc3 ! It transpires that, because of the weakening of this diagonal, the variation 3 9 fxe5 fxe5 (but not 3 9 . . . tlJxe5 40 tlJf4) 40 il.xc3 dxc3 4 1 1lfa l c2 42 'ii'c l i s completely harmless for Black in view of 4L . .i.c6 (or 42 . . . 'i'd7) 43 .i.xc6 'ifxc6 44 ttJe 1 ife4. .ll.f7 .i.e8 The h 1 -a8 diagonal is completely under White's control ! 37 @e7 37 . . . t'Dc3 is hopeless in view of 3 8 i.xc3 dxc3 3 9 11'al. - . 38 • • • f4! I must confess to the readers that initially I made a quite different com ment on this move · than the one that I am now giving. I was sure that 38 . .. t'Dc3 56 II And yet 3 8. f4 is not a mistake! Only, after 3 8 . . . ltJc3 39 fxe5 fxe5 White must exploit the latent pe�sibility 40 c6 ! ! 40 ltld6 41 e3 (it is important to retain this pawn) 41 e4 42 lDc5 1fxc6. If 42 . . . ttJe2+, then 43 'it>fl d3 44 .i.xe4 ! ttJxe4 45 'i'b7 5, or 42 . .. ii.xc6 43 'i'g8. 43 1fa7+ <t>f6 44 h4 (44 . 'i'd5 45 <itih2) - in all variations Black is unable to save the game. In addition, I should add that this is also the outcome of 40 . . . e4 41 'i'b7 �d6 42 t£lc5 .i.xc6 43 'ii'xc7+ 'it>xc7 44 the6+ and 45 lDxd4. Thus 38 f4 was both necessary, and sufficient, and 38 . . . t£lc3 would not have saved Black in view of the remarkable plan involving 40 c6 ! ! , found by Prins. .•. ••• - . . The point is that 'i'b7 is threatened, when the queen at c7 is attacked, and it will only be possible to defend it with the king from d6 or d8, which entails new problems. Prins gives the following variations: 38 . • • ltJe3 In great time trouble Black merely accelerates his defeat; his threats prove to be illusory. 39 40 I 40 ltlxe2+ 41 <i!i>f2 ltlc3 42 1!fb7 �d6 43 .i.xc3 dxcJ 44 ltlc5 .i.xc6 If 44.: . i.g6, then not Prins' s 45 .i.e4 fxe5 1fe4 fxe5 thxg2 Black also stands quite badly after 40 . . . ltJc2 4 1 li)xe5 ttJxal 42 ltJg6+ 'it>d8 43 'i'h4+ <it>c8 44 liJe7+ 'it>b8 45 'i'e4. .•. c2 46 i.xg6 on account of the perpetual check pointed out by Konstantinopolsky (46 . . . c 1 'if 4 7 ltle4+ �d5 48 ifxc7 'i'c2+ 49 @f3 'iid l +), but 45 ifxb5 ! 'i'f7+ 46 .tD ! e4 47 l2Jxe4+ i.xe4 48 'i'c5+, or 44 . . . 'i'f7+ 45 �e2 ! (but not immediately 45 'it>e l c2 46 l2Je4+ �e6 47 .i.h3+ 1if5 ! or even 47. . . �d5 !) 45 ... 1i'h5+ 46 'it>el c2 47 ifb8+ 'it>e7 48 'iic7+ <ittf6 49 ifd8+, and things are bad for Black. 45 .i.xc6 1fxc6 46 tlle4+ <sfi>d5 47 lllxc3+ �d6 48 thxb5+ <bd5 49 iff7+! , and White wins. 57 Here the game was · adjourned, and White sealed his move. 41 · l£lxe5 Game 263 Botvinnik-Smyslov World ChampionshipReturn Match Moscow 1958, Nth game Less strong was 4 1 .i.xd4 .i.c6 42 'i'xe5+ 'i'xe5 43 tLlxe5 .i.d5 44 � g5. But now after 41 tl\xe5 ttJc3 42 Jlxc3 dxc3 43 <it>xg2 or 4 1 . ..CDe3 42 Jlxd4 CDd 1 43 ltJc4+ followed by CDd6 things are completely bad for Black. Therefore no one was surprised at Black's decision to curtail the struggle and resign the game. It was only the fol lowing circumstances that were strange. When the second game of the match was adjourned in a lost position for Smyslov, he turned up for the reswnp tion merely in order to see if the· correct move had been sealed. On this occasion Smyslov acted differently, and, without even seeing the sealed move, he resigned the game. Why did this happen? If it was because Black's position appeared to my opponent to be so hopeless, then why did he have to announce his resignation so late, that the arbiters, organisers and the other participant with his seconds had already set off to the resumption in the chess club? Naturally, they were all able to learn of Black's decision, only when they arrived fo r the resumption. It seemed to the author of these lines that all the persons associated with the match were somewhat disappointed by the fact that in the given situation the World Champion restricted himself to a telephone call, and did not consider it necessary, like all the rest, to come to the match venue. English Opening 1 2 3 4 5 6 c4 �c3 cxd5 g3 .i.g2 bxc3 llbl 8 c4 l£lf6 d5 l£lxd5 g6 l£lxc3 .i.g7 l£ld7 7 In this way the b7 pawn is indirectly defended (8 .i.xb7 .i.xb7 9 l:txb7 l'Llb6), but now the black knight is badly placed. This plan cannot give White any advantage. In a later game from the same match (No.264) I played 8 00 with the idea of subsequently creating a pawn centre: d2-d4 and e2-e4. 0-0 8 . 9 • • lt)fJ ltb8 Black should not have neglected the possibility ·of 'J>reventing d2-d4 by 9 e5 . ... 58 10 0-0 1 1 :)d4 b6 e5 Sooner or iatq · this has to be played, so that White . should not have free control of the centre. 12 13 .t.aJ dxe5 18 l:td5 18 19 l:txe5 20 :dt White had prepared in advance this way of countering the. iinpleasant 18 . . . l:th5. ile8 'ife6 i.xe5 Carelessly played. I was expectinf , when it is hard to imagine anything better for White, but I overlooked my opponent's more than convincing reply. However, also after 13 ltJxe5 ltJxe5 14 dxe5 'i'�dl 15 �.fxdl i..f5 16 e4 .i.g4 I 7 f3 .i.e6 Black would have easily equalised. l 3 ltJxe5 14 ttJxe5 .ixe5. 15 .ic6 . . . The conclusion of the manoeuvre begun with 18 :d5. Now 20 'iVc6+ 2 1 l!d5 b 5 i s not dangerous for White, since he advantageously replies 22 'i'e4 ... . The e5 pawn will not run away, and for the moment Black completes his tkvelopment. The variation 14 'i'a4 1/ \xc5 1 5 �xe5 i..xe5 16 l:fdl 'i'c8 is not dangerous.for him. 14 15 16 17 11c2 ltfdl �xe5 . @xg2 17 ' lhe5 21 22 23 "flxe6 <it>CJ i.f6 l:t.xe6 l:tc6 In an equal ending, Smyslov endeavours to win a dangerous course. Although White's queenside pawns are slightly weak, the activity of his king fully compensates for this defect. lL'lxe5 'i'c8 .txg2 I f 1 7 ltJxg6 Black replies 17 1i'e4 White precisely aims for exchanges (with an advantage of three points, playing for a draw was psychologically fully justified, since it was my opponent who had to aim for a win). .i.b7! 13 lle8 20 21 - . . . .ie4. 24 25 59 :ct ; eJ ild4 ·..tcs 26 ..i.b2 26 27 28 <it;e2 h3 It transpires that in the event of 3 l . . .fxg4 32 hxg4 Black�s passed pawn is easily blockaded,: wherea.s the white pawns will avalanche, •· . forward. But avoiding the exchange leads to difficulties for Black, since it is White who will choose a moment that is convenient for him to exchange on f5 . ltc5 31 Of course, with the bishops on the board the black rook is somewhat restricted, and it is easier for White to defend his isolated pawns. f5 @f7 Given the opportunity, White may be able to make use of his pawn majority, for which he prepares g3 -g4. Initially Smyslov sensibly reckons with this threat. 28 29 • • • . i..e7 33 :gt! 33 34 35 36 �c2 gxf5 .i.xf6 38 �CJ :c6 Now the white rook breaks through on the g-file. a4 . @dJ • il.cJ It was now essential to play 32 . . . .tf6, in order after the exchange of bishops to retain the threat of . . . l:a5 . Now the white pawns will be quite safely defended, e.g. 29 . .lk5 30 �d3 :as 3 1 llal i.f6 32 ..i.c3 ilxc3 3 3 <tixc3 'it>e6 3 4 �b4. 29 30 . 32 h5 h4 :.d6+ .i.f6 g�-f5 �xf6 Or 36 . . . llxf6 3 7 ltg5, with a sig nificant advantage to White. l:c6 37 ltg8 38 @b3 seems stronger, so that the c-pawn should not be pinned, but then Black has the reply 38 . . . l:td(). 38 • • . a6! But now, naturally, there is the threat of 39 . . . b5. 39 40 41 klh8 �g5 �6 When I was sealing this move, I realised that White stood better: Black's king is tied to the defence of the weak h4 pawn and is therefore more passive than the white king. White can create a passed pawn in the centre, where he has a pawn majority, whereas it is hard for Black to do this on U.ie queenside. As a When you want without fail to win a drawn position, it is easy imperceptibly to make a bad move. Black assumed that the reply 3 1 g4 was impossible, since he would create a passed h-pawn, but this was a delusion. 31 lih8 l:g8+ g4! 60 result of a painstaking analysis, I drew up a very subtle positional plan, with which I arrived for the resumption. Tiris plan was so well-camouflaged, that it was overlooked even by Smyslov, a supreme specialist in the endgame and . in positional play in general (although, · perhaps he was simply lazy in his analysis?). �g5 41 42 �d4 For the moment White has to avoid . . . b6-b5, and it was less advisable to do this by 42 <iitb4 on account of 42 . . . l:td6, when the black rook may also cause some problems. But now 42 ... :td6+ is unfavourable for Black, since after 43 �es the white king is too active. :cs 42 . • • But how can Black be forced to play . . . a6-a5 ? This cannot be done by 43 l:.a8: there follows 43 . . . %:.aS . This means that White must first occupy the d-file, to prevent counterplay by Black along this file, and the rook must go to d4, so that the h4 pawn is attacked. 43 l:th7 �g6 The correct reply was 43 ... l:tc6f, and after �4 .:.d7 'ifiif6 45 l:d5 <ifi>e6 46 �c3 :d6 Black forces the exchange of rooks, . transposing into a drawn pawn ending. But it turns out that Black has not yet guessed his opponent's plan. As for the pawn sacrifice offered, White, of course, declines it (44 .:.xh4 l:a5). 44 l:td7 , �6 1lc6 45 :ds Black hurries to correct his mistake and to play . . . �e6 and . . . :d6, but he does not succeed. 46 �CJ The king makes way for the rook, 46 . . . 'iSre6 is not now possible due to 47 l:ld4, and in the given situation the rook exchange loses: 46 . . . lld6 47 ltxd6+ cxd6 48 'it>d4 'iti>e6 49 f3 @e7 50 'ittd5 ltd7 5 1 f4 a5 52 e4 with an easy win. · 46 47 48 %td4 lld7 Jle6 �g5 ltc6 49 �b4 Now this is possible, since Black cannot reply 49 ... .:.d6. It only remains for White to transfer his rook to a8, which will finally force . . . a6-a5. 49 �6 50 :d4 Diverting the king to the flank. If immediately 50 l:.d8, then . . . �e5-e4 is possible. Now about my plan. Since all the time White has to reckon with . . . b6-b5, his main objective is to provoke . . . a6-a5. If, for example, this move had already been made, 43 f4+ �g6 44 llxh4 would lead to a win. But for the moment this is inadvisable in view of the reply 44 . %la5 45 :b8 :xa4 46 l:c8 c5+! 47 �d5 :a3 . • • • . . 61 50 · 'ifi>gS 51 l:d8 l:e6 Now it has all become clear to Smyslov, ',and he, naturally, decides to • • to avoid mistakes. It can happen, how ever, that precisely at this moment a reaction sets in after the tension of time trouble and fatigue tells. Thus here this led to a mistake: instead of 57 l:xa6 bxc4 58 l:tc6 c3+ 59 <Zi>bHii?xf2 60 l:xc3 followed by playing his rook at a3 and • try and complicate matters. 52 llc8 f4 If Black's rook were not on the e-file, 53 e4 would be possible, whereas now advancing his passed a-pawn, White decides to retain his c-pawn. the exchange of pawns is forced. 53 54 55 56 exf4+ lbc7 lth7! lth6 <li>xf4 57 58 59 � :.e4 • • • axb5 axb5 <i>g2 When I played 57 axb5, I expected that now there would follow 59 . . . �xh3 60 cs �g2 6 1 c6 h3 62 c7 :es 63 .:h6 h2 64 �c6 hl 1li' 65 :txh l 'ilrxhl 66 �d7, and Black would resign. But now I saw something else; How can White win after 59 . . . :e l ? If 60 c5 �bl + 61 <itia6 �xh3, he .does not achieve anything with either 62 l:f4 :al+ 63 <iPb7 rlb l+ 64 'l;c7 llhl l ! (a fantastic move) 65 c6 �g2 , or 62 c6 .tial + 6 3 ;.b7 l:bl + 64 �c8 'iftg2 65 f4 (or 65 c7 h3 66 <iii?d7 l'tdl+ 67 ltd6 l:.xd6+ and �8 . . .<ii?xf2) 65 . . . h3 66 l'tg6+ � 67 lnt§.�g� 68 fs h2 69 f6 :n 10 J1g6+ ci>h4 71 c7 lfi>h5 72 l:g8 l:xf6. 56 . . . �g2 57 l1xb6 �xh3 58 llxa6 �g2 59 as h3 60 :tg6+ � 6 1 l:h6 <1tg2 62 a6 ltel 63 c 5 :al was analysed by Averbakh. He showed that White wins: 64 �b5 :b l+ 65 q.,c6 h2 66 a7 D.al 67 �b7 :b l+ 68 @as l:.cl 69 c6 h l 'i' (69 . . . l:xc6 70 ltxh2+ and 71 <lrb7) 70 l:xhl �xhl 71 �b7 :lb 1 + 72 lt>a6 lta l+ 73 �b6 :lbl + 74 �cs %:.c l + 75 q;,b4 :lb l+ 76 �c3 :al 77 c7. 56 57 l:if6+ 'iflxb5 b5 It would seem that, after reaching the nex1 time control at move 56, there should be more opportunity to think and 62 5 My opponent thought for a long time, then, alas,. picked up his rook, but . advanced it �i:tly·as far as e2. From there it can no tonger be played to hl (as in the preceding variation), . which means that it is all ends quickly. As often hap pens, a mistake by one side provoked an error in. reply by the other. 6 . . 59 60 61 c5 <iti>a6 Or 6 1 . . .<li>xh3 62 :f4. 62 63 <lib7 llb6 64 65 66 67 68 c6 7 8 bxc3 l:tbl ll}f3 ll}xc3 .i.g7 · �d7 The strongest move (in Gune 263 I played 8 c4). For the moment White does not advance his c-pawn, so that after . . . e7-e5 he has the possibility of attacking Black's central pawn with d2-d4. :e2 :b2+ :a2+ 8 lJ.b2+ Ac2 10 9 0-0 d4! 0-0 e5 If 63 . . Jhf2� then first 64 :b3 . [ c7 �xhJ <i>g2 lib2+ l:c2 1k6 llb6 f4 Black resigns One of the deepest positional plans that I found during the analysis of an · adjourned position. Game 264 Since the further advance of the e5 pawn would be advantageous only to White, Black will be forced to exchange on d4, in order to release his knight at d7 from its defensive duties. Now 1 1 .i.g5 is tbreatened, and so Black has little choice. Botvinnik-Smyslov World Championship Return Match Moscow 1958, 16th game English Opening 1 2 3 4 5 c4 t2:\c3 cxd5 g3 lhf6 d5 'Dxd5 g6 .i. g2 When the previous _game from a match, in which the players had the same colours, . is exactly repeated, there is heigbte�ed interest in the question of · who will be the first to deviate. 63 10 11 • • e4 • c6 1ia5 The only possibility of counterplay. With this demonstration on the queen side, Black tries to force his opponent to switch to defence. If · 1 1 . . .b6 White would advantageously · · reply 1 2 .i.gS, and then 'i'd2. 12 11'c2 cxd4 ' 14 -*.d2 13. remains tied to the defence of the b7 pawn. 18 e5 would have been · wrong, since� by replying J8 . . . :t7 ! , Black would have freed his queen's bishop, and White would have been left with nothing but weaknesses! exd4 tClb6 This move is not bad, but 14 Jle3 ! was perhaps stronger and more subtle, in order to answer 14 . . . i.e6 with 1 5 a4 ! ( 1 5 . . . .t.c8 · 16 :al, or 15 . . .ttlxa4 16 :xb7 £Db6 1 7 ifxc6), when Black has a difficult position. Now, however, he succeeds in exchanging queens, when the worst for him is over. l4 ... •a4 Playing for complications by 14 'abs would. have been inapprop riate: after 15 a4 .lg4 16 l'.De5 .ixe5 17 dxe5 i.:f3 1 8 .if4 (or 18 a5 ttld7 19 i.xf3 'i'xf3 20 :xb7 ttJxe5 21 .tc3 ) White's position is preferable. ... 15 16 11xa4 l%fc1 /l)xa4 18 19 White needs to be careful. 1 6 :b4, for example, suggests its�lf. hoping for 16 . . . ttlb6 17 a4, but after 16 . . .bS Black gets rid. of his weakness at b7, retains his. knight at c4, and µis . queenside pawns begin advancing. Now 17 :b4 is threatened. 16 17 • • . • . · . fxe4 .i.f5 . 20 21 .ixf5 g4! . l:xf5 • An important interposition. White succeeds in driving the rook off the fifth rank, and after this the a2-a4 advance looks quite different. f5 The correct decision: it is true that subsequently Black reaches a difficult ending, but he does manage to gain a draw . . . 18 • . .ie3 White arrives at that very idea which he did not find at the proper moment on move 14. He waits for Black's bishop to move from c8, and then plays a2-a4. 17 • .i.xe4 The whole point is that after 20 .i.xf5 :xrs 2 1 a4 Black forces simplification (2 1 . . . .i.xe5 22 dxe5 l:xe5 23 hb6 axb6 24 J:xb6 :e7) and can hope for a draw. White� however, finds a way of gaining a dangerous initiative I should merely add that it is doubtful whether Black had anything else. ttlb6 • • 21 22 . . a4 • .:ff8 l:ae8 The only way to defend the b7 pawn was by 22 . . . .lxe5 23 dxe5 :ti, but then after 24 .i.xb6 axb6 there would have followed 25 J!tc4 :a6 · 26 f4, and the thes! TI1e only active possibility - White takes control of n, and the · bishop at c8 64 active placing of his pieces, White takes measures to ensure that the black king remains . on the back rank (incidentally, after 28 llxf7 <t;xf7 29 ltc7+ it is also possible to play 29 . . . ci>g8 30 %%xa7 l:.xe3 3 1 a6 l%.e6 and then . . . g6-g5, when · the white king is shut in on the king side). white pawns become extremely danger ous. Therefore Black gives up a pawn, but brings ·his�1ast piece into play. tnd5 23 .ias " 24 :.l:xb7 25 26 fxe3 dxeS lhxe3 .txe5 �xe5 28 29 :xeJ Itel+ a6 Jlcc7 An elegant solution. another way to draw pointed out in the 29 . . . llxc7 30 %txc7 l:te6 30 31 32 33 34 35 �g2 'ifilg3 <ili>g2 <i>gJ <i>h4 lbc7 After the game was found as previous note: 3 1 Iha? g5f lteZ+ l:te3+ l:e2+ :tel+ l:bc7 Despite its apparent simplicity� the · position is not without its subtleties. After 27 :xa7 .:te4 ! 28 h3 l:xe3 29 itxc6 g5 White does not aclrieve any thing - perpetual check is threatened. But when I was considering my 2 1 st move, I had prepared a different move. ltfi! ! 2 7 lbc6 A splendid defence, thanks to which Black prevents the doubling of rooks on I he seventh rank. I have to admit overlooked this move when that I I played 2 1 g4. The tempting 27, . .J:e4 would have lost to 28 .b3 l:txe3 29 :cc7 l:xh3 30 nxa7 �h4 31 �g2 ! (3 1 . . . 1txg4+ 32 .JJh3 ). But now, since after 28 ltxf7 �xf7 n �c7+ �6 30 l:xa7 llxe3 31 :xh7 lla3 32 :a7 'iti>g5 33 .lla6 :a2 Black's material lo sses are . made up for by the h6! 35 Here too tile white king is shut in on the kingside. h5 36 g5 D raw agreed, on White's proposal. Smyslov saved the e with truly . i;: , virtuoso play ! gaib ·. 65 Game 265 · Smyslov-Botvinnik World Championship Return Match Moscow 1958, 21st game Sicilian Defence 1 2 3 4 5 e4 lllfJ d4 tllxd4 ltlc3 c5 ltlc6 cxd4 tht'6 g6 When White needs to win without fail, such tactics are justified. The simplification and exchange of pieces 9 tllxc6 dxc6 10 such tactics back in 1 93 5 by Salo Flohr e4 pawn, and, in addition, Black need not fear the doubling of pawns on the e-file� since one of the doubled pawns 'i\>xd8 can always be exchanged. 10 f3 was correct. Not only to attack the f7 pawn, but also in order to prevent 8 . .i.e6. The Encyclopaedia also recommends 8 i.f4. 10 . . 8 9 • • • f4 positional mistake, since it weakens the International Tournament 8 e5! This reckless move is essentially a in our game from the second Moscow 11xd8+ .i.c4 • out of the game, whereas now the development of the bishop at e6 cannot be avoided. Naturally, Black does not object to the exchange of queens. I was taught 7 • . for example 6 i.c4. 6 • 9 .. i.e6 would have been an irre parable mistake in view of 1 O .i.xe6 fxe6 1 1 e5, when the bishop at f8 is shut would seem to favour White> but this subsequently demands more accurate play! Perhaps for this reason Smyslov should have chosen a different move, . . • .i.e6! As a result Black's fl is defended, and the opening of the f-file cannot give White anything. ci>e8 a4 A natural, but� apparently, inaccurate move: it was hardly worth preventing . . . b7-b5. It was more important to 11 12 .txe6 J:(f1 fxe6 Since the a7-gl diagonal is open, prevent the activation of Black, s central kingside castling is dubious ( 12 0-0 quently this continuation was analysed in detail by Boleslavsky). After the exchange .of bishops the .i.c5+ 1 3 <it'h l t'.Dg4 14 h3 file3). pawn by either 9 li.f'4, or 9 es (subse 12 In the game White's opening advan • • • .i.h6! black king will occupy a secure position at e7. Weaker was 12 . . .i.b4 1 3 fxe5 tage now disappears. 66 {,i)xe4 1 4 l:lf4 .li.xc3+ 15 bxc3 when the white bishop becomes very active. 13 f5 After 13 :i:xe5 .i.xc l 1 4 .r:.xcl lDg4 1 he black knight would have occupied a n excellent position at e5. .i.xcl 13 14 :xct <t>e7! Now it would appear that White can effectively gain an extra pawn, but this can only be exploited in a pawn ending. !tad8 exf5 Jbd8 %lxd8 · 1s b3 :gS.. b5 ;. 19 g3 1 9 tl'ig4 would seem to be more .dangerous for White, but then he could have responded actively: 20 tiJe4 exf5 (20 . . . ttlxh2 2 1 f6+ and 22 %th l) 2 1 Jlxf5 <li>e6 22 :Its. <i>xe6 20 fxe6 h4 21 :n 2 1 .. .l:.g4 22 ttJ<i l b5 would perhaps have led to more interesting play. :hs 22 gxh4 16 17 � • ... . • The weakening of the e4 pawn tells: White cannot play 15 fxg6 hxg6 16 h3 i n view of 16 J lh4. After 15 fxe6 Xhd8 !, with the threat of 16 'at>xe6 and 1 7 . :d4, Black's initiative would have grown. However, Smyslov finds a s��nsible defence. 15 lldl! In this way White gets ri d of his weak e4 pawn. Black is obliged to occupy the d-file� but it is dubious to do this inunediately: after 15 ... l:ad8 16 Zlxd8 :xd8 1 7 fxg6 ! hxg6 18 %tf3 lDg4 I 9 :g3 White would have gained a pawn majority on .the kingside. There fore he must fir$$. ex�hange on f5. 15 gxf5 . . Here Smyslov offered a draw, and after, in accordance with the rules then in force, I had asked my opponent to make a move� he played: 23 li)dt Now Black does not achieve any thing with the natural 23 . . . lDg4 24 :g2 l%xh4 25 h3 ! �6 26 %lg7 (26 . . . l:xh3 27 lhb7), but by 23 . .. %lxh4 24 lDe3 l:h3 he would have maintained some pressure. However, here too White, of course, has every chance of gaining a draw. But how could Black reject the opportunity to take his score to 12 ... . . . • • 67 points, leaving himself with the task of adding just half a point in the remaining three gaJlles? And Black accepted the offer without much hesitation. Game 266 World Championship Return Match Moscow 1958. 23rd game Reti Opening q)fJ gJ b4 ltlf6 g6 It is well known that this opening, which occurred back in the game Reti Capablanca (1 924), and which caused a sensation at the time, does not bring White any particular gains. Smyslov chose it at a decisive moment in the match, hoping that the author of these lines would have forgotten the above game. However, my opponent was un lucky: during the Return Match I made a thorough study of 3 b4, in the hope of employing it as White. It need hardly be added that in the opening Black felt quite confident! 3 4 5 ..lb2 thaJ .i.g2 0-0 c4 ltlc2 d3 e4 t£le3 •c7 lhbd7 · e6 Smyslov considers himself obliged to prevent . . . d6-d5, but perhaps there was no need for this. Moreover, he apparently wanted to sacrifice the b4 pawn, although the capture of this pawn would probably have been not a bad plan for Black. However, the conse quences of the refusal of the sacrifice also had to be envisaged! 12 13 14 .i.xg7 fxeJ thg4 lhxeJ ciixg7 b6 !llb7 The position has simplified, and the doubled e4 and e3 pawns restrict White's offensive possibilities; it is now Black who has a positional advantage. Reti developed his queen's knight at d2. The move· in the game has its point, since .from a3 the knight can easily be transferred to a strong position - at e3. 5 6 7 8 9 10 11 12 Smyslov-Botvinnik 1 2 3 then advanced his central pawn ( . . . e7-e5); the text move is, of course, more circumspect. 15 ..lg7 0-0 d6 c5 q)gS This two-move threat (16 :xt7+ and 1 7 tllxe6+) is easily panied. · l:lae8 15 · 16 Capablanca first played . . . illbd7 and 68 a3 dS! As is well known, doubled pawns are bad not only because it is hard to attack with the� but also because they are easy to attack. In the given case White cannot exchange in the centre - then the l!-file is opened. 17 bxc5 bxc5 Such an occurrence (and Black made t h is move after 20 minutes' thought! ) is one that could be seen only . . . in the last ga me of a match! Black spent a long t i me studying the variation 17 . .. ilxc5 I x d4 ! 'fle7 (18 . . . 'ifxc4 19 I:.c l and 20 tl:c7) 1 9 h4, and failed to notice that w ith 1 9 . . . h6 he could have won the i mportant central e4 pawn . . . 18 11i'b3 exd4 c5 20 21 22 lCif3 ltact h6 e5 .i.c6 Even now White's position is dif ficult: he can play only with his pieces, his centre is fixed, and his d3 and c5 pawns are weak. Black' s position, meanwhile, is without any defects. Cleverly played. White wants to acti va te his game somewhat, using the t h reat of 1 9 exd5 exd5 20 cxd5 followed hy d5-d6 (for example, after 20 . . . l:.xe3). Therefore Black blocks the centre, after w hich White's pawns in the centre and his bishop become passive, an attack by h i m on the king is not possible, and any 1..� 1 1dgame will favour Black. 18 19 20 If the black knight were at c5, block ading White's c4 pawn, he would be in a desperate position. Therefore the pawn sacrifice is sensible, especially since its . acceptance after 20 . . . tDxc5 2 1 :tac 1 (or 2 1 'i'c4) 2 1 . . .'i'eS 22 'ifb5 would allow the white pieces to become active. Black, naturally, avoids the opening of the posjtion. 23 24 ltlh4 ifdl llb8 1i'd8! After playing his queen onto the d8-h4 diagonal, Black can always parry 'i'g4 with . . . 'ifg5. 25 26 27 28 d4! cxd4 28 69 li.hJ ltlf3 1ic2 ltld2 thf6 1!fe7 :lb7 llc7! Of course, not 28 . . . llfb8 because of 29 tDc4 ! 'i'xG5 30 tlJxe5. Now, how ever, 29 l2Jc4.-. is not possible, and it is inevitable that White will have prob lems over the defence of his c5 pawn. 29 30 31 tJJo ttlh7 :bs �g5 32 ttlxg5 hxg5 1i c4 l:f2 This exchange is also in Black's favour, since one of White's few active pieces disappears from the board. 32 . . . 'ii'xg5 would perhaps have been stronger. 33 34 a4 .li.g4 Here White sealed his move; his position is difficult, but on this occasion Black did not repeat the mistake made in the 1 5th game, when he took a light hearted approach to the analysis of the adjourned position. I worked almost right through the night and for several hours during the day - it was demon strated that White's position was lost. My second found the most crafty move for White: 4 1 h3, in order to exchange the bishops by .i.g4. How ever, in this case Goldberg rightly thought that Black would gain a decisive attack on the king: 4 1 . . . .i.xh3 42 g4 l:lxc5 43 'i'xc5 ltxc5 44 l:lxc5 'i'f6 ! (threatening 45 . . . 'i'f4) 45 'i;h2 i.fl . And after 4 1 'ib3 (4 1 'i'a6 :l.xc5 42 'i'xc8+ ltxc8) 4 1 . . � (41 . . .l:txc5 42 'i'a3) 42 'i'c4 (if 42 'ib6 there follows 42 . . . i.xa4, while after 4� 11i'a3 i.e6 ! and then . . . l:tb8-b3 White is helpless) 42 . . . ii'e6 ! (with the threat of 43 . . . 'i'xc4 44 lhc4 .i.e6) 43 ifxe6 .i.xe6 44 c6 <l;e7 45 :cs @d6 46 l:xa5 !txc6 47 :xc6+ %:txc6 48 � .l:cl 49 .te2 g4! 50 l:ta6+ (with: the rooks on, ihe i.e8 a5 A new weakness has appeared in White's position - the a4 pawn. 35 .i.dl On the one hand, White should not have lifted his control of the d7 and c8 squares, since Black's bishop and rooks become even more active, but, on the other hand, 3 5 'iia6 l:a7 was by no means any better. And what else could White do? 35 36 • • • l:lfc2 l:lbc8 Ad7! Of course, there was no point in Black exchanging two rooks for the queen (36 . . . %lxc5). White now does not have many moves: he has three pieces tied to the defence of his c5 pawn. 37 38 39 40 i.e2 �g2 'i;gt i.dt . <ii?g8 'i;g7 'i;f8 'i;g8 The last move in time trouble. As will be ·evident from the following com ments, 40 ... 'ii'e6 ! would have immed iately forced a won ending. 70 manoeuvre of thebishop to c6 and then . . . fi-f5 is even wore quickly decisive) 50 :c6 (50, .. 'iti>c7 5 1 :a5) 5 1 l:xc6+ c,;?;xc6 52 <i1tel <it>b6 53 ®d2 ctia:s 54 .�.dl �b4 55 .1'.c2 f5 White is unable to save the bishop ending. Therefore it came as a great surprise when, 50 minutes before the game was due to be resumed, the arbiter Harry Golombek informed me by telephone lhat Smyslov had offered a draw. Not without some pressure from my friends ( in which I was 'reminded' of the adjournment session of the 1 5th game) I accepted the offer. . . . 41 'i'b3 Draw agreed With this concluded. the Return 6 . . . a6 7 'i'xc4 c5 was essential, so as not to carry out the advance of the c-pawn in two stages. 7 8 9 'if xc4 ..ie2 0-0 Jl..e7 0-0 Match lDb6 9 9 c5 10 dxc5 would have led to roughly the same situation as in the game. After I O e5 ttlb6 1 1 exf6 lDxc4 12 -fx.e7 'i'xe7 1 3 .i.xc4 cxd4, on the other hand, Black does not stand badly, since 14 itJxd4 'i'c5 is wlfavourable for White. ... Game 267 Botvinnik-Raizman Olympiad, Munich 1958 Queen's Gambit Accepted 1 2 3 4 5 c4 li)cJ �f3 d4 'i'a4+ lDf6 e6 d5 dxc4 After 5 e3 the normal variation is reached, while 5 e4 leads to a more complicated battle, but. since my opponent was largely a tactical player, I avoided this sharp continuation, pre rcrring play · in which positional errors hy Black might accumulate. . 5 6 e4 10 'i'bJ c5 11 12 13 14 15 dxc5 ilg5 i.h4 l:Ifdl tbb5 il..xc5 h6 JJ..e7 1fe8 1fc6 16 e5 lDfd5 It is hard to offer Black any good advice, but now the position is opened, and White, exploiting his lead in development, seizes more space. Otherwise 16 li:Jc7 cannot be prevented, but here the queen soon comes under attack. ©bd7 c6 71 lla8 23 .*.c4 7 1fg5 24 &f)d4 There is nothing else with which to defend the e6 pawn. 25 l:lxf8+ 'i>xf8 26 l:tdl �g8 27 .ixe6+ <ii?b 7 Black's king has at last found safety, but he is already one pawn down. Now White avoids the win of a second pawn (28 .i.xc8 l:.xc8 29 'i'xb7) on account of 29 . . Jk l . 2 8 .ib3 &f)g6 29 &jjtJ 1!f'e7 lla6 30 1ib5 31 bl .i.e6 32 .i.xe6 1ixe6 If 32 . . . ltxe6, then 33 l:d7 'ife8 34 'i'xb7. 33 'ifxb7 llb6 34 111c7 &fjf4 The trouble for Black • is that if 34 ... :xb2 White wins as follows: 35 ::td7 'ifg8 36 lLld4 l:ld2 37 .tld8 1fa2 38 tLlxf5 8 . He has to play on two pawns down. llxd6 35 . · lld6 36 1ixd6 , 1ib3 17 .i.xe7! 17 l:tacl 'i'e8 18 i..xe7 iixe7 was more favourable for Black, when his knight would have remained on its central post at d5. 17· &f)xe7 1le4 18 l:lact 1fg6 19 l:ld4 After 1 9 . . . 1!fxe2 20 l:tel Black would have been forced to give up his queen for rook and bishop� but perhaps this was not worse for him than the game continuation. 20 li)xa7 A simple exchanging combination, allowing the white pieces to invade the enemy rear. llxa7 20 llxa2 21 1fxb6 22 :ds f5 The only possibility of avoiding major loss of material. If 22 ... lLlc6 there would have followed 23 l:txf8+ <it>xf8 24 iic7 (24 l:Xc6 is weaker on account of 24 . . . :a 1 + 25 i..fl l:lxfl + 26 <it>xfl iid3+ 27 <it>el We4+ 28 <it>d2 'i'xc6) 24 ... :as 25 :xc6 (25 . . . �al+ 26 l:.c l). • . • 72 opponent decided to see how I would defend with Black. 3 4 5 tLlfJ .i.f4 cxd5 tLlf6 White normally plays 5 t[)c3, but if he had wanted, things could have merely led to a transposition of moves. 5 6 37 Thus 6 t'.Dc3 would have led to a stan dard position, whereas now Black also gains the opportunity for a different decision. 1fd2 6 White has a sufficient material ad ,· antage, and so, naturally, he avoids complications. 37 38 �h2 l£ic6 e3 lDh5 l£ixh3+ g5 After 3 8. . . t[)g5 3 9 &Llxg5+ hxg5 40 'ii'xg5 White would have easily won the ending, but now the position of the black king is exposed. 39 1fd7+ <it>g8 40 'i'c8+ �g7 41 liJd4 1ixb2 42 l£ixf5+ Black resigns. Mate is inevitable: (42 . . . <it>g6 43 'i'e6+ <it>h5 44 lDg7+ <&t>h4 45 g3 mate). I carried out a similar idea back in 1 952 against Tigran Petrosian (cf. Volume 2, Training Games), and later in 1966 in the USSR Team Championship in a game with Birbrager. In the latter game e2-e3 had not yet been played, and White could have retreated his bishop to d2. Here he could have played either 7 .i.g5, as Petrosian did, or 7 i.e5, provoking 7 . . . f6, which weakens Black's position. But after the unpreten tious retreat of the bislwp. to g3 Black has no difficulties. Game 268 Citrone-Botvinnik O�vmpiad, Munich 1958 Slav Defence 1 2 3 d4 c4 cxd5 d5 c6 Usually I myself used to employ this exchange. Therefore, most probably, my 73 ' .." 8 7 il.gJ hxgJ lhxg3 g6 9 10 �cl �e2 .i.g7 e6 11 12 13 0-0 :ct �a4 0-0 i..d 7 Sooner or later Black will have to safeguard his kingside in this way. In the middlegame this set-up allows Black to open up the game by ad vancing his pawn to e5, but before completing his development this would be premature. Forcing a weakening of Black' s position ( . . . b7-b6), which, however, cannot be e�']>loited. Because of his mistake on move 7, White has no active plan. 13 14 15 b3 �c3 16 ll.\bl 1fb8 17 18 19 'ii'd 2 i.. b5 .i.xc6 ifb7 l:t.ec8 b6 l:e8 J.f8 . i..b5 l:t.xc8 f6 23 24 l:txc8 1fxc8 25 �d3 g5 26 27 28 a4 li'id2 �fJ 29 lhb2 W'b7 1i'c3 Avoiding further simplification, but if Black wishes he can always exchange queens. For the moment he will con tinue to strengthen his position. Nothing would have been achieved by 24 . . . 'ifxc3 25 t'L\xc3, when the bishop at b5 is attacked. The queen is transferred to b 7, to gain control of the a6 square. Otherwise Black cannot begin operating with his rooks on the c-file. Further restricting the mobility of the white pieces. White plays for simplification, hop ing that in the endgame it will be easier for him to main the balance. The position is indeed simplified, but it is in the endgame .:that the two black bishops will be particularly effective. ... �e5 lhc8 %tel In the given situation this advance is quite safe for Black, since the passive placing of White' s pieces does not allow him to mount an attack on the enemy king. After the weakening of the a3 and b4 squares in White's position, it makes sense for Black to move his bishop onto this diagonal. 19 20 21 22 The black king is the first to come into play - a good omen for the coming endgame. In the ending .too White has no active �xc6 74 continuations, and all he can trust in is . · passive play. ·- 35 • • • e5! Even further restricting and tying down the enemy pieces. 29 30 31 • • • �d3 ltlb2 "1e7 'ifi>d8 1!fc7 1ixc7+ 'it>f1 • . . �et �ed3 38 ltl3b2 i.h5 bishops are attacking the helpless white knights, neither of which can now move. <l;xc7 39 40 41 b5 Clearing the way for the approach of Black's king to the white pawns on the queenside. 34 ... .i.a3 is threatened, and after 34 axb5 i.xb5+ 3 5 <ii?e l i.b4+ 36 @dl .i.fl White would also have lost a pawn. 34 35 bxa4 e4 A picturesque position - Black's Earlier, in view of Black's active pieces and with the queens on, it would have been dangerous for the white king to move towards the centre. 33 ltldt bxa4 But not 37 . . ..i.xa4 because of 38 ttJc3 . The king has approached, and the exchange of queens has become timely. Subsequently the activity of the king on the queenside will be decisive. 32 33 36 37 <i&>el .lb4+ 'it>ft <i&>b6 f3 White resigns Game 269 Botvinnik-Pomar Olympiad, Munich 1958 Caro-Kann Defence .i.a3 1· 2 3 4 5 It was easy for Black to act - such positions were played in masterly fashion by Akiba Rubinstein. 75 c4 e4 exd5 d4 lhc3 c6 d5 cxd5 ltlf6 e6 6 ttlfJ resulted from 1 1 h4 with the threat of 12 ltJxb6 t'.Llxb6 1 3 i.xh,7+, and if l l . . .f5 1 2 ltJg5, or l l . . .h6 1 2 fl.h3 . !i.e7 · Later, by the efforts of many grand masters, in particular Tigran Petrosian, it was · shown that the simplest is 6 i.b4 (b.y analogy with the Nimzo Indian Defence), when Black has a comfortable game. 11 12 ... 7 After 12 . .txc5 (Black loses immed iately in the variation 12 . ltJxc5? 1 3 t2Jxc5 .ilxc5 1 4 .i.xh7+ 'it>xh7 1 5 liJg5+) 1 3 .i.xh7+ lttxh7 14 liJg5+ lttg 8 15 ltlxc5 White has good prospects of an attack. . • • • 13 14 15 0-0 1O b4 ltJa4 e4 ltlxd7 exf3 c6 cxd7 0-0 In the afore-mentioned game Sokolsky played 1 5 . . i..f6, after which White retained · some advantage. Apparently for this reason Pomar chooses a different way. 16 • xf3 lLle5 il.d3 In the event of 8 b4 Black had the reply 8 . . . t'.Lle4, but even so this was the most logical continuation. 8 ilf4 also came into consideration. Now, however, Black gains the opportunity to under mine the opponent' s pawn chain. 8 9 . .. c5 A committing move. Perhaps White should prefer 7 cxd5 exd5 (if 7 . . . tiJxd5 8 j/_d3) 8 i.b5+ .i.d7 9 .ixd7+ t'.Llbxd7 1 0 'itb3, when he has a slight, but enduring advantage. 7 8 .. ' . bxc5 e5 dxc5 • . 17 1i'g3 17 18 1fxd3 Of course, not 17 i.xh7+ lttxh7 18 'i'b5+ lttg8 1 9 1fxe5 on account of 1 9 . i.f6. b6 a5 liJfd7 . . lLlxdJ .i.d6 With the prosaic threat of l 9 . .i.xh2+ ( 1 9 . 'i'h.4 20 f4) 20 �xh2 'i'h4+ 2 1 . . .. lttg l 'i'xa4. However, ·several years later in a correspondence game Sokolsky Simagin the more interesting 1 8 . . . d4! was played, with approximate equality. 19 i.aJ 19 20 21 22 .i.xd6 11fd1 ltlcJ1; Tying the bishop at d6 to the defence of the rook at f8. 11 ' b5 A continuation which was lrnown from the gam� Kopaev-Sokolsky ( 1.950). Complicated play would have llb8 1fxd6 .le6 Naturally, the ..white knight should occupy the d4 square, but for the 76 go in for the exchange of the queenside pawns and to pin his hopes on the strong position of his lalighi in comQination with his active rooks. At this point Pomar, evidently taking account of his opponent's shortage of time, offered a draw. But the game continued . . . moment White wants to consolidate his queenside r;dvantage with · a2-a4. It stands to reason that Pomar prevents this. · 29 30 31 22 23 24 25 ltabt l1b3 ttle2 28 ttld4 lta3 11re3 i.d7 i..xb5 1f a3 ltfc8 1lc5 Wc2 If 25 . . . d4 White could have replied 26 :b2. If White's knight is able to take up a dominating position in the centre with material equal, and with him also having a protected passed b5 pawn, tltlngs will be hopeless for Black. Therefore he takes the opportunity to eliminate the a2 pawn, since 26 a4 loses to 26 . . . .i.f5 ! Nevertheless White is prepared to sacrifice a pawn 26 27 1fd2 lh:a5 bl 31 32 lta7 .li.e6 It is hard to suggest anything better. For example, 32 ... :eds is not possible 9xa2 1i'c4 l:te8 on account of 3 3 :cl Wb4 34 'i'xb4 lixb4 3 5 l:xd7. By skilfully manoeuvring, Black has retained his material advantage. The obvious 29 :c1 leads after 29 . i..d7 30 iid2 11b4 to. a worsening of White's position (he also has to reckon with the threat of 29. . . i..g4). Therefore he has to . i.. d 7 A serious mistake, as a result of which White regains the .pawn and invades by force with his rooks onto the seventh rank. 3 L.:b7 3 2 :c 1 1id3 33 'lb4 .i.c6 etc. was correct. 33 34 35 . llcl 11'xb4 lhxe6 11b4 ·:xb4 fxe6 llcc7 :h4 Black is forced to give up the g7 pawn. 36 77 Finding the only way of defending the h-pawn. 37 38 39 40 41 lbg7+ �h2 .D.ge7 f3 %lad7 'it>h8 litf8 lth6 sacrificing a pawn, . by answering 4 1 .. JUI6 with 42 g4 :xf3 .43 �g2 1:.f8 44 !it>g3 followed by llg7+ and g4-g5, but it soon transpired �t with 43 lUu6 ! 44 %tg7+ � 45 ltxh7 :r? ! Black can defend satisfactorily. Closer to midnight I found something approaching zugzwang, but my analysis companions Keres, Kotov and Flohr quickly dispelled all my illusions. When it became clear that, with the four rooks on the board, White could not achieve anything real, positions with one pair of rooks were studied. It was only around four o'clock in the morning 1hat I finally ' suspected' that one of these positions was won. An hour later everything had become ' clear', and I was able to go off to sleep, but a consultation with Flohr was set for seven o'clock in the morning . . . The analysis was approved, and all that we had to decide was the practical aspect: how to conduct the resumption psychologically, in order to win in the most certain way this point that was so important for the team? It was decided to divide the resumption into three stages: 1) mark time initially, so that the opponent should gain the impression that no winning plan had been found; 2) then imperceptibly try to create a position of fictitious zugzwang, and finally, if nothing positive was achievect, then 3) as if despairing of success, exchange one pair of rooks and switch 'to the real plan. Such tactics would cause Black the maximum disorientation, and a mistake on his part was probable. When at ten o' clock ·the game was resumed, the Spanish master (now .. ... <ii>g8 It seemed to me that White had a great advantage, but Black has good defensive possibilities. This last move was sealed at the adjournment. The first impression is that White should win easily: his rooks have complete control of the seventh rank, whereas Black's rooks are tied down. One of them has to keep watch on the h-file, guarding the h7 pawn, while the other has to safeguard the king against mate on the back rank. In addition the e6 pawn is weak. However, analysis showed that in the event of passive defence by Black ( . . . l:f8-f6-f8-f6) it is not so easy for White to strengthen his position. The whole problem is that the white pawns cannot advance . . It is true that as soon as the analysis of, the adjourned position began, Flohr . suggested immediately 78 grandmaster) had p.o suspicion, of course, of our cunning plot He played quickly and cqnfidently. 41 42 43 :tr6 . ltg7+ l:lge7 <ith8 · Thus a repetition of moves - the first part of the plan - is carried out. 43 44 45 46 47 <ifi>g8 �gl l:lf8 l:lff6 llf8 <i>f2 l:td8+ Ad6 With this move White imperceptibly switches to the second part of his plan. lUJ6 47 <it>f8 .48 ltb6 49 50 ltd7 l%a6 White would finally have had to. reveal his hand and to switch to the main plan. However, the talented Spaniard was insufficiently experienced, �d he sue. cumbed to the clever tactics devised by two players, whose combined length of service as grandmasters already exceed ed half a century . . . Pomar thought for a long time, but, ' lulled' by White's unconvincing manoeuvres, he was unable to overcome all the dangers. 50 51 52 53 :as+ ltxf8+ llxh7 53 54 f4 <ifi>e8 d4 55 56 :a7 'it>gJ llf6 llf5 l:tb g6 :rs �xf8 Black can no longer save the game. '&t>g8 The advance of this pawn forces White in the end to release the black king from the back rank. · Intending 57 .. Jid5. 57 58 59 60 Thus, White has managed to create a position in which Black can no longer routinely ·continue 50 :t'S in view of 5 1 l1xd5, and 50 ... lthg6 is also bad because of 5 1 :as+. However, the zugzwang is, alas, only apparent: by continuing 50 . .l:t�5 ! (pointed out by Keres) Black wowp have maintained a defensible · position: After this reply ... . 79 Ital ltd2 � <i>e4 <l;e7 lld5 q.,d6 �c5 Or 44 ... r!hg6 45 g4. 45 ll.g7+! (the g2· pawn has to be defended) 45 <i>h8· 46 �d4 l:l g8 (otherwise White strengthens with impunity the position of his king) 47 lbg8+ �xg8 48 <i&>eS llg6 49 g4 h5. If 49 . . .l:Ch6, then 50 f4 llxh3 51 <it>xe6. 61 llc2+ <it>d6 Now Black,s counterplay with his passed d-pawn is neutralised� and White can begin advancing his kingside pawns. 62 g4 .•• l:ta5 The only active possibility, but now Black loses another pawn. 63 lld2 <lre7 64 65 66 67 68 69 lixd4 ltd3 � :eJ g5+ cli>g4 50 l:te7 lla3 lla4+ CiW6 llal I am sorry to say that during the analysis I thought that 50 lla7 was also good enough to wi� but after the game Tal found that in this case White does not achieve anything in view of 50 . . . hxg4 5 1 hxg4 llg7 ! 50 bxg4 51 hxg4. 'ifa>f7 This would evidently have been the most convenient moment for Black to tenninate his resistance. 69 70 71 72 73 h4 lle5 l%a5 <i>hS ••• 1:a4 xtb4 l%bl ltgt+ The black king cannot avoid return ing to the eighth rank, and then even reducing the material deficit will do nothing to help. 73 74 75 76 77 78 lta7+ <i;g6 e5 h5 e4 h6 h7 Black resigns Despite the material equality and the limited number of pieces, Black's position is difficul�. If 5 1. . .�hS White wins by 52 l:lxe6! l:Xe6+ 53 'ati>x� d4 54 </;f7 ! ! d3 55 g5 d2 56 g6 dl W (Black has obtained a new queen two moves before White, but even this 4does not save him) 57 g7+ �h7 58 g8'1'+ �h6 59 \ig6 mate. However, there was also another variation: 51 � 52 :h7 'itg8 53 llh5 But what would have happened if Black had defended correctly? Let us return to the adjourned position� in order to demonstrate the third stage of the plan, which in fact remained 'off-stage,. 41 l:Z.tT6 42 �gt ltf8 43 @fl ltff6 44 <la>e3 lif8. ••• ••• '1tf7 80 54 g5 <ia>g8 55 f4 <lilg7 56 llhl ci>g8 57 :at � 58 �d6 �. case i n the King's fudian Defence, the knight would be badly placed. But not 58 . . . ®g7 59 <:/Je7 �h7 60 rJ;n, which had seemed inevitable to me during our night-time analysis. 59 :.a7 �e8 (or 59 . . . <it>g8)� and a draw is nevertheless unavoidable ! It is probable that, if it had not been a team event, White would have immed iately gone into this ending1 but in an Olympiad� at a point when the tourna ment situation was very tense, even a chance opportunity could not be neglected! 9 f3 It is doubtful whether 9;, exf5 was stronger: 9 . . . gxf5 (of course, not 9 . . . i.xf5, as will be evident from the further course of the game) 10 .i.h5+ @f'S, and it is not so easy for White to find a good plan. 9 10 • • l£if6 • 0-0 11fd2 Black should himself exchange on e4, when the open f-file allows him to hope for simplification. 11 exf5 Game 270 Botvinnik-Alexander Olympiad, Munich 1958 King's Indian Defence 1 2 3 4 5 g6 i.g7 d6 ltlc6 e5 d4 e4 c4 lDcJ i.e3 Black simplifies the game, but this involves a loss of time. 6 d5 This is more energetic than 6 tbge2 . 6 7 . . • tDge2 11 If 7 . . il.g4 White could play either 8 f3 . forcing a problematic piece sacrifice (8 . . . i..xf3 9 gxf3 ll)xf3+ 1 0 ct>f2 iff6 1 1 �g3)� or simply 8 'i'd2 Jtxe2 9 t'.bxe2 ttJxe2 10 i.xe2 tiJf6 1 1 f3 with a slight advantage. · .i.xe2 .. i.:d·s - . 8 . A typical strategic mistake in the King's Indian Defence. On no account should the central base the e4 square have been left under White's control. l 1 .. .gxf5 was essential, retaining chances of counterplay. Now, however, White ' s plan is clear: an attack on the kingside. l£id4 lLlxe2 12 13 f5 Practically forced: after 8 . . . ttJf6 the advance . f7-f5 · would have been hindered, while at. .e7, as is usually the g4 hJ - .*.d7 There is no reason to hurry . First White must neutralise the advance . . 81 . . . e5-e4, which would allow Black to activate his bishop at g7. . · 13 • • • 22 23 24 25 a6 Alexander tries for activity on the queenside, but here he has insufficient forces to create any real threats. 14 0-0-0 h5 hxg6 'ifdJ . .i.c8 ..tf5 .txg6 b5 This essentially concludes the game, e.g. 25 . 1i'f7 26 :xh7 .i.xe4 27 1Wxe4 Wxf3 28 'i'g6 iif7 29 ::th8+ ! , or 28 . . . 'i'xe3+ 29 <it>bl :n 30 l:dhl . Black finds a clever reply, but even this is no longer able to halt the nonnal ...development of events. .. 15 c5 Now Black cailllot avoid c5-c6� which will split his position into two poorly connected parts. 15 16 17 18 19 ttle4 .i.dJ c6 h4 b4 a5 'l'e7 .i.c8 ..ta6 28 1fe4 l:txf3 bxg6 The simplest. By temporarily block ing the bishop at g7� White takes play into an endgame� and when his rook subsequently reaches f7, -it.- wi �l all be over. 28 flf7 . · ttlxf6+ g5 .i.e4 .i.xg6 1lxg6 'iftf8 Or 27 . . Jhe3 28 l:.dfl with inevitable mate. Black is hoping to weaken the pressure on his position by exchanging the light-square bishops, but White exchanges knights� and the vacated central e4 square will be occupied by his bishop. 20 21 22 25 26 27 .txf6 .i.g7 29 30 31 The position is blockaded, and in view of the threat of h4-h5 Black must urgently return his bishop to f5. g6 11'xf5+ 11df1r. 11f5 ltxf5 Now the rook · penetrates by force onto the seventh rank. 82 variation is of considerable significan�e in the opening, and therefore his prospects are better. 6 7 tiJfJ al 7 8 d3 9 b4 tLlc6 White is annmg to play b2-b4 as soon as possible. 31 32 33 34 -· lb.ft+ :n <ia>c2 l:bfl+ 'it?g8 :cs e4 The bishop gains its freedom too late when it can no longer be of any help. 35 b3 36 <ii>d l 37 @e2 38 i..a7 Black resigns: the �e7 0-0 Black has thought up an interesting counter with the sacrifice of a pawn, and therefore he does not prevent 9 b5. It would have been more circumspect, however, to eliminate this threat by 8 . . . a5, as was played against me in later games by Flohr (Wageningen 1958) and Portisch (No. 363). f5 By continuing 1 O . . jlf6, Black intends to obtain a perfectly good position; of course. White has no reason to reject the win of a pawn. The alternative was 9 . . . a6. . .lc3 i..e5 Ac3 white king will win the e4 pawn and then continue its victorious advance. [ Game 27 1 Botvinnik-Duckstein O�ympiad. Munich 1 958 English Opening �=------i 1 2 3 4 5 c4 lbc3 g3 cxd5 i.g2 e5 lbf6 d5 lbxd5 lDb6 10 b5 1 1 ·lDxe5 12 f4 This is essentially a variation of the Sicilian Defence with reversed colours. The extra tempo that. White has in this Threatening 1 3 . . . .i.b3 . 83 both lDd4 .i.f6 li.e6 1 3 . .. ltJb3, and 13 14 llbl fxe5 .i.xe5 f4! 18 Cleverly played. In the event of 1 5 i.xf4 l:l xf4 16 gxf4 'B.h4+ 1 7 <it>d2 1i'xf4+ 1 8 · e3 Wf2+ 1 9 �e2 'ifxg2 20 exd4 .i.d5 2 1 l:lgl 'ii'xh2 the position of the white king would have given serious cause for alarm. Therefore castling is forced. 15 0-0 18 d4 20 21 22 ltb4 lle2 d5! 11fxc3 11c4 ... f3 .a2 1fa1 Another tactical subtlety. This move proves possible, since due to the position of his king at g8 Black cannot take the pawn. And after the d4 pawn has advanced, White has both an extra pawn, and the better position. . exfJ lbf3 • • • 19 19 'i'aS is also bad, if only because of 20 .id2. Black continues to 'tangle' with his opponent, but in the end it is he who becomes entangled. Simpler was 15 . . . fxg3 16 hxg3 (or 16 l:lxf8+ 'i'xf8 17 hxg3 Wc5) 16 . Jlxfl + 17 'i'xfl ttJd5 with some initiative for the pawn. 16 17 %le3!! Titis move leads to a won position. Now Black's queen's rook remains out of play for some time, and his queen is driven to a l . �xf3+ 1f d4+ In a fit of emotion Duckstein im perceptibly commits a decisive error. Essential was 1 7 . . . l:lxf3 18 'i'xf3, and only then 1 8 . . . 'ii'd4+. After the ex change of rooks Black could have been confident about the safety of his queen, and his queen's rook would have quickly come into play. In avoiding the simplification� Black overlooks White ' s reply and ends up i n a hopeless position. 22 23 . • . lld2 llad8 23 l:ld4 was weaker, since on this square the rook would have come under attack. 23 24 25 -*.b2 l:lxdl .i.f5 'l'xdl+ The endgame is, of course, easily won. 84 25 life8 25 ... c6 was S(ftheWhat better. 26 .i.fJ Defending the rook at dl. ttld7 26 ttlc5 27 e6 28 llf4 .i.g6 h5 29 h4 Or 29 . ttJd3 30 lixd3 i.xd3 3 1 l:.f7. ttld3 30 lk4 31 -*.d4 3 1 l:txd3 .i.xd3 3 2 :xc7 was also good enough for a win. lie7 31 c5 32 l:tc3 33 bxc6 bxc6 34 lldxd3 Of course, 34 J:xc6 could also have been played, but the forcing variation in the game wins most simply. .i.xd3 34 l:tee8 35 .i.c5 cxd5 36 :xd3 l:.d7 37 e7 38 .ixh5 3 8 :xd5 would have led to the same result. ltexe7 38 lbe7 39 Jl.xe7 40 llxd5 It is not often that one has to play an ending with an extra piece and a pawn. Black's aim of winning back at least a pawn leads to mate. lleJ 40 11.xaJ 41 <it1'2 cli>f8 42 i.g6 43 :e5 Black resigns • • Game 272 Uhlmann-Botvinnik • Olympiad, Munich 1958 ' Nimzo-Indian Defence 1 d4 e6 2 c4 ttlf6 3 ttlcJ ii.b4 4 e3 b6 5 i.dJ The variation employed by White as though implies 5 t'.Dge2. Now the worst for Black is already over. 5 i.b7 6 ttlf3 ttle4 7 0-0 . . f5 7 Was it not possible to win a pawn? After 7 ... ti)xc3 8 bxc3 .ixc3 9 :b 1 Black loses too much time and ceases to control the centre, which is :fraught with danger. 7 ... -*.xc3 8 bxc3 tt:lxc3 9 'ifc2 i.xf3 10 gxf3 'l'g5+ 1 1 'it>hl 'i'h5 12 l:tgl 'i'xf3+ 13 l:tg2 f5 is also risky, as White has not only a draw ( 14 'i'xc3), but also the possibility of an attack:· 14 i.a3 ti)e4 15 :n l:t.g8·· 16 i.e2 'i'h3 1 7 f3 ti)f6 18 d5 (Keres-Spassky, 1 965). __ 85 8 the game is quite . sensible, but there was no reason to avoid the exchange of pawns on b6 . 1fc2 Here Svetozar Gligoric used to make the pawn sacrifice 8 d5 . The acceptance of ii involves some danger (e. g. 8 . ltJx:c3 9 bxc3 i..xc3 10 l:!bl Ji.f6 1 1 e4 0--0 1 2 l:e l fxe4 1 3 Jlxe4, as in a game Gligoric-Hecht, 1 969), but it can also be declined (say, by 8 . . . .i.xc3 9 bxc3 ltJc5). . . 8 9 10 bxc3 l:tbl • • • i.. xd2 :b2 16 :at /i)xd2 /i)d7 And now avoiding the capture on b6 is a serious mistake, allowing Black unexpected tactical possibilities . 15 bxa5 .i.xc3 0-0 . • • Otherwise 16 . . . ltJb6 and 17 . . . a4, creating a securely defended passed pawn. White is threatening c4-c5, but Black could have gained good counterplay by immediately switching his king's rook via f6 to the h-file. Incidentally, back in 1 924, playing against Abramovich, in a Dutch Defence, I carried out a similar victorious raid with my rook along the route f8-f6-g6. 10 13 14 15 16 17 • . • /i)b6 lba5 As the further course of the game shows, it would have been better to allow . . . a5-a4. c5 This decision to block the c4 pawn leads merely to a weakening of the b6 pawn, and White intensifies the pressure. 11 a4 1ic7 Halting the further advance of the rook's pawn by 1 1 . . . ttJc6 would have allowed the possible 12 i..xe4 fxe4 1 3 'i'xe4. d6 12 a5 After 12 . . . bxa5 White would soon have won back the a5 pawn, and Black would have been left with a weak pawn at a7. 13 17 Ji.e4! ! Of course, Black avoids 17 . . .ltJxc4 1 8 .i.xc4 if xa5 1 9 ..ixe6-f!. and 20 �xb7 ' but, by moving the b7 bishop out of the line of fire and pinning the bishop at d3 , h e creates the irresistible threat o f taking with the knight on c4. ltJd2 Initially 1 3 tDel seemed very danger ous, but after finding the variation 1 3 . . ltJd7 14 f3 ltJef6 1 5 d5 g6, I came to the concl�iOn that Black's position was defensible. The idea of the move in . 18 19 86 .txe4 ·' 'ii'b J';f· . fxe4 corrections into this evaluation. Donner knew about this, of ,.co�se, but he decided to test an idea of his own. After 1 9 'i'xe4 ltlxc4 20 iYxe6+ Vif7 the fork remains in force, and Black wins the exchange. 19 20 21 22 23 'Ifxc4 ifxe6+ :a2 1Wxe4 tDxc4 •xa5 <i>h8 'Ifc7 s 9 tDas c5 In the notes to the afore-m�ntioned game it was shown that 9 . . . c6 loses by force after 10 b4. In a hopeless position, such over sights tend to happen. 23 d5 lDd2 ... ifti White resigns This game is not typical of the play of Wolfgang Uhhnann (born 1 93 5). In later years he was one of the world' s leading grandmasters (he was a World Championship Candidate in 1 97 1 ), and he had a significant influence on the development of chess in East Gennany. His play is distinguished by great energy, and he is especially dangerous in attack and counterattack, as I myself was to experience at the following Olympiad, in Varna. 10 11 12 'i'c2 b3 .ilb2 13 bxc4 llb8 b5 bxc4 A necessary preparation for Black' s nex1: move. He creates a pawn weakness for his opponent at c4. Game 273 Botvinnik-Donner 13 Wageningen 1 958 1 2 3 4 5 6 7 c4 tl)f3 tDf6 g6 i.g7 gJ ..tg2 0-0 ttlc3 d4 d6 a6 lDc6 0-0 • • • ..th6 This is Donner' s idea, which he had already tried in practice. In order to understand the chara�ter of this game, it should be borne in mind that it took place in the last round, and only a win would secure White victory in the tournament. Therefore it has to be acknowledged that Black's choice of opening was psychologically subtle. He has now created the threat of 14 . . . .txd2 and 1 5 . . . ttJxc4, and if J4 e3 he has the unpleasant reply 1 4 . . . .ilf5 . King's Indian Defence For a long time this variation was considered safe for1Black, but the game Botvinnik-Geller (No. 2 1 3 ) - introduced 87 White chooses the most active con tinuation, although it also assists the development of the opponent's counter play. 14 16 17 f4 ... e5 15 16 �ael gxf4 exf4 l:le8 Subsequently, instead of this, Robert Wade recommended 1 7 e3 ! 17 14. i'Llcb 1 followed by .i.c3 also came into consideration. 14 • • • lLldl • • • J:Cb7 As soon as White transfers his bishop to c3, Black's position deteriorates, since tl1e basic defect of the entire variation is revealed - the poor position of the knight at a5 . Therefore Black was obliged at any cost to continue playing actively, and, as Boleslavsky rightly pointed out, by first playing l 7 . . . ltJh5, and only if 18 e3 - 18 . . . :b?. Then after 1 9 .i.c3 :be? 20 ii'a4 lDb7 Black succeeds in gaining counterplay in the centre and on the kingside. If 14 . . . ttlg4 White would have par ried the immediate threats by 15 ttldl . 18 .i.cJ 19 eJ .i.d7 Here the threat of 1 9 ifa4 already had to be parried. Now that . . . .i.f5 is no longer a tlrreat (the bishop has to guard the a4 square), White need not hurry. 19 20 After the game I suggested that 16 . . . ttlh5 would have been the most ' dangerous for White. forcing the reply 1 7 e3 . Eight years later this recommend ation was tried by Petrosian in a game with Portisc]\ but after 1 7 . . . :es 1 8 liJce4 ! .i.f5 1 9 .i.c3 White consolidated his advantage, since the e4 square remained under his control. The whole point is that the correct way for .Black to exploit this idea is a different one: 17 . . . Jtg? 18 liJdl .ifs . For example, the game Ree-Sax ( 197 6) continued 19 .i.e4 .i.xe4 20 liJxe4 .i.xb2 2 1 ttJxb2-ite�t . with equal chances. Donner, however, played differently. . • • l£1h5 ltJf2 It only remains to play this lrnight to d3 � and everything will be ready for the offensive. 20 21 22 .i.xg7 "ifcJ 22 23 .i.fJ .*.g7 lhxg7 Here the queen successfully fulfils those functions that were previously carried out by the bishop, .and the weakening of the f6 square assists the development of White' s attack and the advance of his central pawns. It is the latter that Donner tries to prevent. 88 f5 A delay at the critical moment. White wastes time on taking control of the h5 square. Meanwhile, in the event of 23 e4 fxe4 24 tt1fxe4 Black would · have been unable to defend simultaneously his f6 and d6 squares. And if he replies 23 . lLJhS, then after 24 tiJd3 fxe4 25 t2Jxe4 J.f5 26 th'Cd6 l:txe 1 27 ltxe 1 'i'xd6 28 :es+ <1;;n 29 1i'h8 White's threats are irresistible. Now, however, Black is able to parry the advance of the e-pawn, and the worst for him is over. . His only chance is to exchange queens, after which he will be able to switch his pieces to the queenside am.d exploit Black's numerous weaknesse� (a6, b6, c6 and d6). Therefore White provocatively moves his knight from d2 to fl , which in the event of the exchange of queens allows Black, and not White, to gain temporary control of the b-file. But this, as soon transpires, is of no significance. . 23 24 25 tild3 �fl .i.c8 :n 25 e4 is no longer a danger for Black, in view of the fact that, after the ex change of pawns on e4, his knight at g7 could go via f5 to occupy the important d4 square. As a result, for the moment it is not clear how White can strengthen his position. 25 26 27 lEifl 1i'a5 28 29 30 31 'l'xa5 ltc2 l£id2 l:t.cJ! lhxa5 llb7 lhe8 31 32 33 34 l:taJ l[bl thxbl This reaction came immediately - the bait has worked. The start of the decisive manoeuvre. �b6 thb7 l:txbl+ lbb7 l:t.fe2 llef8 Black's only real fighting piece has been exchanged, and the remainder are quite incapable of any activity. It cannot be ruled out that soon Black will go onto the offensive with . . . g6-g5. White resorts to a psychological trick. 34 35 89 &/jc7 &/jd2 � 36 37 38 <bf2 h4 1'.dl 38 39 40 llb3 :b6 cl;e1 h6 My opponent, ·like many other players. thought that my ...main trump was my ability to prepare. . In so doing, they did not take ac(a;ount ·. of another, possibly more important component of my chess strength - positional under standing. It was this that enabled me to find good plans in original situations. Now White is threatening to play his bishop to c6 . To prevent this, Black is forced to weaken the defence of his b7 and a6. .i.d7 l£la5 :as 5 6 An oversight in a hopeless position. 41 42 e6 d4 6 d3 followed by f3-f4 came into consideration. l£lxc5 dxc5 <"llf7 d6+ Black resigns 6 7 • . • l£ld7 ilf4 In a small tournament every game is worth its weight in gold. It was this one that enabled me to win first prize. Carelessly played, since as a result of the subsequent manoeuvre . . . lhg8-f6-h5 Black gains a tempo. 7 J.e3 was stronger. Game 274 This assists the above manoeuvre, and, in addition, here the bishop feels · safe, since a2-a3 and b2-b4 would make it impossible for White to castle queen side. 7 Tal-Botvinnik ftVorld Championship Match Moscow 1960, 3rd game Caro-Kann Defence 1 2 3 4 5 e4 l£lc3 t[)fJ h3 gxf3 • • . 8 h4 9 e5 J.b4 lbgf6 c6 d5 .i.g4 ..txfJ After 5 'i'xf3 White has some advan tage, so that there is no point in inventing such an artificial course. Both in this, and in other games of the match, such a method of playing by Tai was aimed at creating an original situation in the opening, in which the first concern was not the correctness of his conceptio� but to force his oppon ent to spend more time, time which would be so needed . . . in time trouble. Unexpectedly White changes plans. If he had been · intending to block the 90 advantage. Therefore Black chooses the prosaic manoeuvre of his knight to a strong post at f5 . centre, he could have done this without 7 �f4 He should have played 9 'i'd3. . 9 10 . • • .i.g5 tl'lh5 1ia5 15 16 Already in this match I was trying to employ a strategy that was highly un pleasant for Tal - to avoid the opening of the position, after which his cal culating skill would give · him a big advantage. It was for this reason that I refrained from 10 . . f6. However, I employed this method insufficiently consistently. and, in addition, my poor form prevented me from utilising the benefits of this way of combatting Tal. ..ad2 al Jl..eJ 1t'b6 Jl..e7 g6 tl'la4 'ilfd8 tl'lg7 h6 The inevitable opening of the h-file will be to Black's advantage, since the weakness of the f-pawns becomes real. 17 18 .i.xh6 .i.f4 tl'lf5 nxh4 19 20 l:lxh4 tl'lxh4 If Black takes with one of his minor pieces, it will be pinned on the h-file. . 11 12 13 ... i.g5 0-0-0 Of course. not 1 3 . . . 'i'xb2 14 etJa4. 14 . 20 15 • • • b5! At the appropriate moment - now, after White has castled - this thrust is fully justified. In the event of 2 1 ctJc3 lLlb6 Black's attack is very strong. The other possibility - taking the f3 pawn was inadvisable, since it would have handed the initiative to the opponent. Tai employs an idea that is typical of his play: he aims to open the position at any cost, in order to make the calculation of variations harder for his opponent. W'd2 Tempting Black into 1 5 . . . .ixh4 1 6 Xtxh4 ii'xh4 17 Jl..g5 Vh2 1 8 1Wb4. When you are playing Tai, it is waste of time considering such variations. Even if they are objectively unfavourable for him� subjectively they are to his 21 91 tLlc5 lhxc5 22 23 dxc5 .i.e2 ilxc5 il.e7 Thus, in a quiet position Black is a pawn up, and the outcome of the game would appear to be decided. Alas, this was not the case: he still had 17 moves to make before the control, and already had little time left on the clock. 24 25 26 ©bl l:thl .tgJ ilc7 0-0-0 tl)f5 27 �h7 :f8 The invasion of the seventh rank by the white rook is not dangerous. It has merely created a threat, which would have been most simply parried by 27 . . . .tc5, when if necessary the f7 pawn can be protected by the rook from d7. The move in the game is more passive, although it does not yet spoil anything. 28 29 ii.f4 .i.d3 30 lhh8 Here Black could have won by 3 1 . . .'iitb8 32 .ltx:f5 (32 .ixb5 ild8) 32 . . . 'i'hl + 33 �a2 'i'xf3 34 .ie3 d4 (35 i.xd4 'i'd5+). Unfortunately, in time trouble I missed 34 . . . d4 (it was later pointed out by Petrosian). True, Tal later tried to show that White could nevertheless have gained a draw, by playing 32 a4 (instead of 3 2 .txfS). I will not try to cast doubts on the nwnerous and complicated varia tions that he gives, but will mention that Black can simply force the transition into a favourable endgame: 32 . . . 1i'd8 3 3 ilxd8+ .i.xd8 34 axb5 ltld4 3 5 bxc6 tbxf3 , with the threats of 36 ... i.c7 and 36 . . . ltJd4. 1fd8 'lh8 The further course of the garne and subsequent analyses confirmed that this move is quite logical. It parries the threat of 30 i.x:f5 gxf5 3 1 .i.h6 and leads to simplification. 29 . . . g5 30 .i.h2 rlh8 was weaker in view of 3 1 ..ixf5 exf5 32 'i'd3, when White is out of danger. 31 Tai rightly pointed out in his commentary that after 30 l:txf7 'i'e8 3 1 iVa5 'iixf7 . 3 2 ii'xa7 nh7 3 3 a4 .id8 Black would have parried all the threats, while in the event of 3 1 �xe7 White has insufficient compensation for the sacrificed exchange. 30 31 11'a5 l!fhl+ Now White has finally achieved his aim: the position of the black king is exposed, and things end in perpetual check. 32 33 34 35 '1i>a2 11a6+ 1fxc6 .txb5 Wxfl ci>b8 'iWxf4 Already here White could haye gained a draw - by 35 'ifxb5+ cj;c7 36 "i'xh8 92 10 ifa5+, but he tries to go fishing in troubled waters. 35 36 37 'ife8+ 1fc6+ 1i'xe5 ¢>b7 .. Tal's hopes of gaining an advantage in this variation of the Caro-Kann Defence were not realised. In time trouble White might have tried 37 .ta6+ 'it>xa6 38 'i'c6+ <it>a5 39 10 11 c3 , since here only one move 3 9 . . . 'i'e2 leads to a draw (40 'i'c7+ �b5 41 'i'b7+). [ 37 • ..i.d6 game Keres-Olafsson World Championship �Match Afoscow 1960, 5th game suggested itself. with sufficient compen sation for the sacrificed pawn. 11 12 13 Caro-Kann Defence _______, c6 d5 dxe4 .tf5 .i.g6 lhgh5 tllxh5 ll'lbd7 il'lxh5 It is easy to understand that Tal considered this position to be in his favour : after the natural reply 1 3 . . . 0-0 White would have gained chances of an attack on the black king. But Black finds another possibility. More often this knight is played to f4 which, of course, does not change anything. 6 ... 7 h4 e6 A subtle point of this variation is that if 7 lDf4 is played immediately, and if 7 . . . ..t d6 - 8 h4. then after 8 . . . Vc7 the position of the knight at f4 is insuffic iently secure. Now. however, Black has time to ensure a retreat for his bishop to h7. 7 8 9 h6 �f4 i. c4 ( 1 961). Although after 12 . . . fxe6 the opening battle developed satisfactorily for Black, .i.xd7 lbxd7 instead 12 . . . 0-0 13 Game 275 via h3, . . . Tal-Botvinnik e4 d4 lbc3 lbxe4 lhgJ lhte2 . .te3 11 i. xe6 is refuted by 1 1 . 0-0, e.g. 1 2 .i b3 .r!e8 1 3 i..e3 .txf4. After 1 1 c3 tDbd7 the sacrifice 1 2 .i xe6 was made <it>b8 Draw agreed in the 1 2 3 4 5 6 1ie2 In the 9th game (No.277) Tal played 1 0 0-0, which is also not dangerous for Black. It can be definitely' stated that .i.h7 lbf6 13 93 • • • ll.g8! Starting with the present game� this became . a standard way of defending Black· s' g7 in the given situation. Black will castle queenside, where he has nothing to fear, since White too can only castle on that side. 14 15 g4 g5 20 ifc7 .i.g6 !tlgJ i.xg5 21 22 23 • . • 23 24 1i'h6 ff Black had ex ii.dJ l:hd3 'ifxh6 .il.xd3 1£lb6 • . • :o gxh6 Provoking . . . f7-f5, in order to gain counterplay. hxg5 f5 l:td6 llg4 24 25 26 27 lle1 c3 ltle2! 27 28 29 l:thl l:tg3 tL\d5 l:td8 l:lxg3 30 31 fxg3 <ii>d2 lt g4 The only possibility. Black. of course, must avoid the variation 27 :;m4 28 t2Jf4 c.fi>d7 29 ltJg6 :112 30 t2Jf8+ @e7 3 1 l:xf5 . i.f4+ i.xf4 . . . After 1 9 '.t>b 1 ii..xg5 20 hxg5 the g5 pawn would have been weak. 19 20 . Otherwise there would have followed . . . tiJd5-f4. After 1 8 hxg5 Black would have had a pleasant choice between 18 . . . i..xg3 and 18 ... :hs. 18 19 • changed queens on:-e3., White's position would have been preferable. Now, however, after the practically inevitable exchange on h6 the g-file is opened, and White's weak f- and h-pawns cause him considerable trouble. Now 16 gxh6 is not possible in view of l 6 . . . 'ii'a5+. 16 0-0-0 0-0--0 Now too the h6 pawn is indirectly defended, since if 1 7 gxh6 gxh6 1 8 Jtxh6 there follows 1 8 . . . l:th8. The knight at h5 proves to be out of play, and White is obliged to improve its position. 17 18 . A necessary fineS.$P. 1i'xf4+ 11t'e3 This improves White's pawn for mation, but Black rejected 29 . . J�dg8 30 .:t.hg 1 , since he wanted to keep one pair of rooks on the board. 1lg8 A loss of time. Black should have immediately advanced his· 'king to d6 and then played . . . b7-b6 and . . . c6-c5, intensifying the pressure on the enemy position. After the manoeuvre of . the rook to e4 it will occupy a pretty, but less active position, than at g8. 94 32 33 34 35 <it?el � l:l.el �cl ci>d7 lie4 <it>d6 If my opponent had thought for a little longer, he would undoubtedly have seen the clanger awaiting him after the natural reply 41 e4+. It was this that I examined in the 17 minutes before I sealed my move. but I did not find a win. This is what could then have happened: ••. In the knight ending too Black· s chances are better, but the exchange of rooks eases White's defence. 35 36 37 @xel 'it>e2 38 39 40 41 cxd4 <li>d3 �e2 a4 l:txel c5 cxd4 42 'it>c4 b6! It is very important to take control of the c5 square. Both in his night-time analysis before the resumption, and in his detailed comments, Tai missed some subtleties in the resulting position and gave only the variation 42 . . . a5 43 b4 axb4 44 <li>xb4 liJf6 45 tiJf4 tbd5+ 46 liJxd5 ®xd5 47 �c3 with a draw. Meanwhile. by continuing 44 . . . �d5 (instead of 44 . . . liJf6) Black would have won: 45 a5 ( 45 @c3 b6) 45 . . . e3 46 �c3 �e4 47 @c4 (4 7 d5 <iPx.d5 48 ci>d3 @es 49 @xe3 lLJxg3 50 l"Dxg3 f4+ etc. 9) 47 . . . �f3 48 �d3 ltixg3 . Therefore White must reply to 42 . . . as with 43 b3 ! 43 ...b6 44 b4 axb4 45 �xb4 liJf6 46 <iti>b5, with sufficient counter-chances in view of the weak ness of the b6 pawn. 37 . . . c4 would have created greater difficulties for White. �f6 �h5 e5 A moment typical of Tai' s match tactics. He played this quickly, merely in order to force Black to seal his move, in the hope that after five hours of play it would not be easy to find the strongest reply. I personally do not like such methods, and I prefer moves that are in keeping with the logic of the position. I th.ink that, from the �·practical viewpoint too� this is the soundest way. 95 7 . . . d6 8 e3 h6 9 .i.h4 g5 10 .i.g3 tiJh5 1 1 i.b5+ ( 1 1 . . . .td7 . 12 i.xd7+ 'i'xd7 1 3 l'he5), but now · the . manoeuvre . . . lDh5 is parried even more simply. 43 b4 a6 44 d5! There is no other possibility. 44 <t>e5 45 as bxa5 46 bxa5 ti:)f6 47. lbf4 e3 48 <.t>dJ with a draw. ••• Also taking account of the fact that Tai had not exchanged on e5, obviously fearing the centralisation of the black king, I sealed another move, thus mis sing this chance, which unexpectedly could have appeared after 4 1 . . .e4+ 42 rbc4 b6 or 42 . . . a5 . Alas! 41 42 42 43 lDf6 dxe5+ <ltxe5 b4 Draw agreed 8 This would be the result after the obvious variation 43 . . . lllli5 44 b5 f4 45 gxf4+ ltJxf4+ 46 etJxf4 <i>xf4 47 <itd4 ��g4 48 a5 @xh4 49 a6 bxa6 50 bxa6 �g3 5 1 �c5 h5 52 -.tic6 h4 53 @b7 h3 54 �xa7 etc. 9 10 8th game King's Indian Defence 1 2 3 4 5 6 7 d4 c4 tlJf3 d5 cxd5 ll'lc3 i.g5 • • • .i.e2 d6 a6 A dangerous symptom in the given variation. The threat of . . . b7-b5 is easily parried. and the weakness of the b6 square may tell. If Black is forced to make such a move, it follows that his position is difficult. Botvinnik...:...Tai World Championship Afatch 1 960, lle8 Now the black knight can no longer occupy h5, since it is controlled by the queen (in the event of . . . h7-h6 and . . . g6-g5), and later this duty will be taken over by the king's bishop. Game 276 A.foscorr1 e3 ll'ld2 9 ll'lf6 e6 c5 exd5 g6 11 a4 tLlbd7 1fc2 �b6 . Finally, after a change of inove order, the position has become fully theor etical, but things are no easier for Black as a result the evaluation is clear White stands better. 12 0-0 1lc7 .i.g7 0-0 An insignificant finesse, since sooner or later Black will be forced to play . . . d7-d.6. Evidently Tai wanted to avoid . 13 96 14 .i.f3 18 19 This caution is necessary. After 14 e4 ttlli:d : 5 1 5 exd5 .i.xc3 16 ilxc3 %he2 it would not have been easy for White to demonstrate that the weakening of · Black's kingside is worth the sacrificed pawn. 14 ... • • • 1ia2 W'd8 fS It was essential to play l 9 . . .b5, in order to bring the queenside pieces into the game. At the same time, the advance of the kingside pawns that Black under takes is not in the spirit of the position. c4 20 21 22 23 lt:\cJ lLlc4 .i.e2 llJa4 23 24 25 gJ f4 �h8 h5 i.d4 26 ifaJ :lb8 g5 g4 'ilf6 The white pieces have taken up the standard positions� typical of this variation, brought into practice by Averbakh. The opponent' s weaknesses on the queenside are fixed, and it could have been considered that the game was essentially decided. However, as in many other games of this match, my uncertain play led to 'miracles' . One gains the impression that Black overlooked the loss of this pawn. Any other move would have been better, e.g. 14 . . . .if5 1 5 e4 i.d7 followed by . . . ttJc8. 15 It was unfavourable for Black to take en passant, since after 25 . . . gxf3 26 :xn his king would have been in danger. But now he loses any chance of an attack against White's kingside. .i.xf6 In order to win the pawn. White has to exchange his good bishop. 15 a5 liJbd7 16 tiJa4 was dubious on account of 16 . . . 'i'xa5 17 tCixc4 'i'c7 with the threat of . . . b7-b5. 15 16 17 Now the defended. 17 a5 llJce4 c4 26 . . . Jtxe3+ 27 tt:Jxe3 'ii'd4 was not possible because of 28 @£2. Now White had an elementary win, for which he should have gone into an endgame: 27 exd4 'i'xd4+ (27 . . . l:xe2 28 'i'd3 :e4 29 .:.ae l ) 28 :fl Itxe2 29 Wc3 Wxc3 30 bxc3 l:t.xf2 3 1 @xf2. Instead of this he makes a move that has no particular point, which. however, is typical of supem.cial play in time trouble. •f ilxf6 l£ld7 pawn cannot be -*.e5 1 8 llxc4 After 1 8 lt:Jxc4 b5 1 9 axb6 ctJxb6 Black would have gained some compen sation for the material deficit. 97 was made on purely practical grounds: to ease the defence in · · severe time trouble. 31 32 33 34 bxc3 l:lfel :ct 11xc3 .i.b5 l£\e4 Instead of this move, which could have lost, with 34 :d4 ltlxc3 35 .id3 White would have retained his extra pawn and winning chances. 27 �ab6 Why leave the important c5 square undefended? 27 h4 White's mistake could have been exploited immediately by 27 . . . .ixb6 28 axb6 ttJcs. Instead, for the second time Black allows his opponent the oppor tmtity to force a win by exchanges (28 exd4), but White is 'consistent' . . . 28 l:tadl .i.xb6 On the second occasion Tai does not miss his chance. 29 axb6 30 gxh4 30 ... l£\c5 Black can be content: his pieces have finally come into play, there are numer ous weaknesses in White's position, and the extra pawn is of no significance. A sensible decision: White avoids the creation of a weak pawn at g3 . i.d7 In the event of 30 . . .'i'xh4 White could have replied 3 1 l:lfl or 3 1 1i'c3+ and 3 2 1'e l . 31 llbc8 34 Although Tai had thirteen minutes remaining, he made this move instantly, evidently reckoning that the most important thing was not to allow the opponent time to think. After 34 . . . :ec8 ! (not leaving the b7 pawn undefended) Black would have won the exchange, without allowing White any significant compensation. Now, however, the pic ture changes sharply. 35 36 37 tt.\a5 llxe2 lbcJ Le2 tbxcJ Of course, White 'voluntarily' gives up the exchange, wjnning an important tempo, since after 37 :ec2 (or 37 l:leel) 'lfcJ White had only three minutes left for ten moves, and the exchange of queens 98 37 . . . lDe2+ 3 8 l:xe2 .Ihcl+ his position is hopeless. 37 J:lxc3 • 38 • and so I sealed my move only after 30 minutes' thought. 41 • lLlxb7 Now it is clear what Black's thought less 34th move has cost him. Were the b7 pawn defended here, it would be time for White to resign. 38 • • • Itexe3 In his commentary Tal suggests that 38 . :bs 39 ltJxd6 l:td3 40 ltlxf5 l:txd5 4 1 e4 :xrs 42 exf5 J:xb6 would have led to a drawn ending. This seems insufficiently convincing: after 43 � White has a clear advantage. In this variation Black should not return the exchange (4 1 . . . l:t xf5), and then, perhaps, he retains chances of continuing the fight. .. 39 40 l:beJ �xd6 lt)f7+! ! The knight makes way for the d-pawn with gain of tempo. Now after 4 1 ... <it>h7 42 d6 one of the pawns queens immediately, while if 4 1 . . . r:i;;g7 (or 4 1 .. . �g8) there follows 42 b7 l:b3 43 liJd8. Here 43 . . . a5 44 d6 a4 45 d7 a3 46 ltJe6+ or 43 . . . <it>fS 44 h5 �e8 45 h6 . <it>xd8 46 h7 is possible. To complete the picture, it should be pointed out that the seemingly tempting 4 1 b7 would even have lost: 4 1 . . J lb3 42 M+ <it>h7 ! 43 ltJd8 a5 44 d6 a4 45 d7 a3 46 ltJe6 a2. Black resigns. %ixe3 lld3 Game 277 Tal-Botvinnik World Championship Match Moscow 1960, 9th game Caro-Kann Defence 1 2 3 4 5 6 e4 d4 t£ic3 tDxe4 tDg3 ltlle2 c6 d5 dxe4 .*.f5 .lg6 lhf6 The two players repeat the 5th game of the match (No.275). 7 8 9 10 Here I had to seal my move. I was able quite quickly to find the winning plan, in which Black's king and rook are unable to cope with White's knight and three passed pawns. Even so, I decided to check all the variations at the board, h4 lhf4 .tc4 0-0 h6 .i.h7 e6 In the afore-mentioned game Tal played 1 O 'i'e2 and did not gain any advantage, but after h2-h4 kingside . castling also seem unjustified. 99 10 • . . ild6 these considerations did not occur to either of the players during their home preparations. Now Black wants to castle, after which the position of the white pawn at -h4 would .give him an obvious advan tage. Therefore, when I was preparing for the game, I anticipated that Tai would without hesitation sacrifice a knight for two pawns, in addition obtaining fine piece play. 10 . lDd5, the move recommended by the Encyclopaedia. looks more risky. 13 14 /i)xe6 .i.xe6 - . 14 15 16 il.xh7 �f5 16 ... � Ibh7 fxe6 ifc7 A natural move, but perhaps not the strongest, after which there could have followed 13 �5 and then f2-f4, when it is not so easy for Black to find an acceptable plan. I should mention that the attempt by White to play differently: 1 3 ttJh5 :tf8 14 c4 (instead of 1 4 f4) 14 . . . �g6 1 5 l£lg3 liJbd7 16 c5 i.xg3 17 fxg3 liJd5 1 8 :le 1 0-0-0 would have led to an inferior position for him. Instead of the move in the game, it would have been better to reply 1 2 . liJbd7, and only after 1 3 l:el ( 13 li)h5 <it>f'S) 1 3 . . 'i'c7, not fearing the discovered check. It is amusing that . . tLlbd7 White, of course, could not allow his opponent to castle queenside, and he ex1racts the maximwn from the opposition of rook and king - he exchanges the light-square bishops, to gain the f5 square for his knight. .. 11 12 :tel .tg8+ g6! A move, as it seems to me, in the style of Capablanca. Black agrees to material equality (a piece for three pawns), but quickly completes the mob ilisation of his forces, and his pieces will be well coordinated. The following variations looked less · . convincing: 16 . . . Jth.2+ 1 1 @hl g6 t8· �xh6+ llxh6 ( 1 8 . ®g8 19 g3 gxf5 20 Ji.f4) 1 9 liJxh6 'i'f4 20 .l:.e3 (but not 20 g3 .ixg3 2 1 fxg3 'i'xg3) 20 . . 'ii'xh6 2 1 <it>xh2 'i'xh4+ 22 @gl (22 . tlJg4 2 3 ilf3+ and 24 'i'g3). 1 00 .. . . . . 17 : . .i.xh6+ 18 l£lxd6 . · <ai>g8 "i'xd6 After 1 8+l:t,di6 1 9 :e6 the storm clouds would have been gathering over the black king, e.g. 1 9 . . l:.xh4 20 10 �Yd3 . . 19 .i.g5 lite7 The exchange of rooks reduces the attacking potential of the white pieces. 1 1 is significant in this position that for the moment White does not have any passed pawn. And if. with this aim, he plays 20 lhe7 'ilxe7 2 1 h5, then Black will have good chances both in the endgame (2 1 .. .gxh5 22 'i'xh5 ltJxh5 23 .�1.xe7 CDf4), and in attack (2 1 . . . rJ;;g7 22 hxg6 �h8). 20 21 1id3 Wg3 rJ;;g7 A positional mistake, since in the endgame, on account of his doubled g pawns, White will have little chance of saving the game. It would have been better to try and maintain the position. 21 . . • l:txel+ Incidentally, 2 1 . . . 'ii'xg3 22 :xe7+ \M'8 23 fx.g3 @xe7 is also possible, but Black chooses a rather more cunning continuation. 22 23 lhel fxg3 .• • • c4 d5 cxd5 d6 27 28 29 30 :ct :c7 .lxf6 lllg4 cxd5 ltldf6 This entire pawn breakthrough is essentially a diversionary operation to gain freedom for the white king. ltfi I:td7 <ttf7 White also has to part with his bishop, in order to be able to advance his kingside pawns. 30 31 32 1fxg3 <ai>f2 llxd7 l£lxf6 �e6 Nothing is changed by 3 2 'ifi>f3 <ai>xd6 . Although Black is a piece up (for three pawns). White has a good bishop, and if he should succeed in bringing his king into play, the outcome will be unclear. 23 24 25 26 27 32 33 34 35 36 � 'iti>f4 g4 @e4 rJ;;x d7 �xd6 i;t>e6 l£ld5+ By playing 36 'it>g5 c3;Il 37 h5 White would not have achieved anything on the kingside, but after 3 7 . ltJb4 he would have lost both of his queenside pawns. l:tf8! Making it harder for the white king to come into play, whereas White's control of the e-file cann�t. .achieve anything for him. . 101 . 36 37 ' 38 39 <li>f4 �e4 a3 ll'if6+ ll'id5+ l£lb4 The white pawns would have been more diffi cult to attack after 3 9 a4, but their loss could not have been avoided. 39 40 41 h5 h6 l£ic6 g5 Tal again hurries to make his 4 lst move on the board, to force Black to seal his move. 45 46 st?d6 �d5 a5 �xh6 47 48 49 50 51 52 <ifi>c4 <ifi>b5 b3 �xa5 �b4 �CJ lDct lDdJ ibcl lDxbJ+ lDcl 52 53 54 �c2 <it>dJ <it>g6 ltle2 ltlct + 55 56 <itc2 �d3 ttle2 tLlf4+ 57 58 �6 �c4 g3 ltle2 White resigns Since Black has time for this, by no means urgent, capture, it means that on the other side of the board his lmight is successfully opposing the king. A final attempt - White pursues the knight. But such an operation normally succeeds only in studies. The reader already knows that for the resumption I had very little time left (less than 20 minutes), and now my flag was already hanging. In such circwn stances I never reject an opportunity to repeat moves. 41 • • • Here the time control was reached. 'it>f6 I spent a great deal of time consider ing this move, but I sealed it, after convincing myself that the raid by the white king on the queenside did not present any danger. Analysis, and then the resumption of the game, confirmed this. 42 43 @d5 'it>e6 �g6 World Championship}.,fatch Afoscow 1 960, 20th game Nimzo-Indian Defence Or 43 <it>d6 ttJa5 44 �c7 b5 45 �b8 lbc4 46 <3lxa7 (46 b3 a5) 46 . . . tbxb2 47 �a6 liJc4 48 'it>xb5 liJxa3+ etc. 43 44 a4 Game 278 Botvinnik-Tal 1 2 3 ll'ia5 lbb3 102 d4 c4 lDcJ · · ttlf6 e6 il.b4 4 5 a3 bxc3 �xc3+ lhe4 7 8 . • . 1ih6 g6 d6 At first sigtlt _this seems an unnatural move, whiclt�hould be easy to refute, but this impression is deceptive. The knight thrust has some positional basis: Black is aiming for . . . f7-f5, transposing into a variation of the Dutch Defence that is not Wlfavourable for him. It is true that the knight will be forced to retreat, which loses Black tempi, but if White plays 'iic2, then this loss .of time will to some degree be compensated. In the previous games of the match where this variation occurred, I did not find the correct system of development. Finally, in home analysis I was able to do this, but at a time when the fate of the match was already decided. Even so, I thought it useful to check my con clusions, reckoning that such an oppor tunity might not present itself again. If 8 ... lllxc3 there would have followed 9 f3 with the threat of IO a4, while if 8 . 'i'g5 9 �3 and then f2-f3. After a long �ink Tai prefers to continue his development. Now there is a threat to the c3 pawn. In the event of Black avoiding this main idea (for example, after 6 ... 0-0)� he would have had to reckon with the possible manoeuvre ..td3, llle2 and f2-f3, forcing the knight at e4 to abandon its central position� without provoking 'i'c2. Of course, not 10 ... fxe4 1 1 fxe4 liJxe4 12 llJf3� when Black stands badly. 6 7 e3! f5 'ifh5+ This is the point. If Black does not want immediately to give up the right to castle (7 ... c.t>f'8), he has to weaken his dark squares, which may prove especially sensitive in view of the absence of the bishop covering these squares. The play becomes more lively, and Black has to give up ideas about the c3 pawn. 103 . . 9 10 10 lhf6 f3 e4 • • • 11 12 ..tg5 i.d3 13 lhe2 e5 'iie7 llf8 The need for such moves indicates that the opening has not gone in Black's favour. The natural choice, but not the best. Tai rightly considers that 13 'i'h4 was stronger, and Llien Black would not have gained counterplay on the f-file (if 13 . . . 'i'f7 there would have followed 14 ..th6). 13 • • • iff7 The subtle point of this move is that after 14 �xf6 ifxf6 15 exf5 .i.xf5 16 i.xf5 gxf5 17 'i'xh7 ttJa6 18 'i'h5+ 'i'xc4 1 8 'ifxh7 exd4 with compli cations. There is no need to demonstrate that this was advantageous to White. Meanwhile, with 16 . . . h6 17 .td2 g5 18 'i'g3 tDc6 1 9 h3 tiJf6 Black could have achieved a good game without any complications. @d7 ! Black has sufficient compensation for the sacrificed pawn. 14 Wb4 This move is based on an oversight: White was convinced that the opening of the f-file was in his favour. Meanwhile,. after 14 0-0 the opening of the f-file would have become impossible, and if 14 . . . thg8 1 5 'ii'b.4 f4 he could have begun play on the opposite wing, e.g. 1 6 c5 'i'g7 1 7 cxd6 cxd6 1 8 'i'el h6 1 9 Sl.h4 g5 20 ii.fl, and it transpires that the black king is insecurely placed. 14 15 . • 16 fxe4 .. 1ff2+ 17 18 19 <it>d2 .i.xh4 :hfl 19 20 21 l:lxf8+ <iti>xd3 22 lDgJ 'ilfxh4 lDf2 There is no point in giving up the active dark-square bishop. fxe4 l£lg4 • . Of course, not 16 . . . tiJf2 because of 1 7 0-0. Here I checked once more the variation 16 <t>d2 h6 1 7 .i.e3 g5 18 'i'g3 ltlxe3 19 <it>xe3 (but not 19 'i'xe3 1i'f2). and realised to my regret that, when considering my 14th move, I had overlooked the clever 1 9 . . . 'i'f4+! White is therefore forced to allow th� exchange of queens, and with opposite- . colour bishops his advantage is slight. lDxdJ <it>xf8 .te6 · A poor manoeuvre, after which it can 16 h3 Tai writes that if 16 :gl he was intending to play 16 . lDfl 1 7 i.c2 . be considered that the battle is over After lDgl -f3 in the end Black would have been forced to exchange pawns on d4, and he would have had to overcome considerable difficulties in order te maintain the balance. 22 23 . 104 lDfl ti)d7 a6 24 .i.f2 . 25 ' lDd2 @g7 lif8 Now the knight can reach f3 , however, not in two moves, a s indicated earlier, but in four, and . White can no longer achieve anything With this, since his opponent has succeeded in mobilising all his forces. 26 27 �el b6 lllf6 l:lbl Draw agreed Game 279 Olympiad, Leipzig 1960 King's Indian Defence 1 2 3 4 5 6 7 d4 c4 lll cJ e4 f3 .i.e3 .i.dJ lLlf6 g6 �g7 d6 0-0 b6 a6 • • • c5 Threatening to intensify the pressure in the centre - 9 . . . liJc6. 9 d5 e6 10 11 12 exd5 lLlgJ 12 13 exd5 lllbd7 • • • 0-0 lle8 This allows Black to seize the initiative. It was essential to play 13 .i.e2, preventing the following knight thrust, then ttlce4 and Ji.f4. 1'd2 8 tllge2 c5 9 es is also played, after which Black does not succeed in forcing d4-d5. 8 lll ge2 White prepares a retreat for his bish op (after 12 . . . tlle 5), but significantly weakens his d4 square. This is wrong, since the exchange of the lmight for his bishop did not present any danger. The variation beginning with 6 . . . b6 is usually associated with . . . c7-c5. However, Black must be careful. For example, the immediate 7 ... c5 loses material after 8 e5 in view of the threat of i.e4. The move in the game vacates the a7 square for the rook, and in some cases . . . b6-b5 will be possible. 8 10 Now after the exchange on d5 White will have to recapture with the e-pa� and the opening of the central file is unfavourable for him. This could have been avoided by 1 O a4, in order to have the possibility of playing 1 1 cxd5. It should be added that after 10 dxe6 .i.xe6 Black also has no difficulties. Tamburini-Botvinnik 13 • • . lLlg4 Irrespective of whether White retains his dark-square bishop, as in the game, or chooses 14 fxg4 .:t.xe3, the advantage . is already with Black.. 105 14 15 16 i.g5 .i.f4 .i.e2 f6 /i)ge5 21 22 .i.fl r h5 h4 Routine play. As has already been mentioned, White should not have avoided the exchange of his d3 bishop. 16 . • . f5! Black has achieved a pawn formation that is ideal for the given variation. In addition his knight occupies an ex cellent post at e5. whereas the e4 square is unavailable to the white knights. It only remains for Black to complete the mobilisation of his forces before l�mnching a decisive offensive. 17 18 .i.g5 .i.h6 llfdl �abl i.eJ :a7 lbf7 This move is possible. since the d4 square is defended by a white rook (2 1 ltJg4 22 fxg4 l:txe3 23 'i'xe3 .i.d4 24 �xd4). ... 23 24 lbge2 �f4 lbh5 An oversight, leading to the loss of a piece. However, it is already doubtful whether White's position can be defended. lDf6 il.h8 In view of the weakness of the d4 square. there is no point in Black exchanging bishops. 19 20 21 The board is full of pieces, but Black is successfully playing for domination. Therefore it can be concluded that this method of play was known even before the appearance in the chess arena of Anatoly Karpov. At the time of the Olympiad in Leipzig, the future World Champion was just nine years old. 24 �xf4 25 .i.xf4 .i.d4+ White resigns: after 26 'iii>h l g5 he • • • loses a piece. The opening variation that occurred in this game used previously to be con sidered favourable for White, Thanks to the efforts of the young ·�andmasters Geller and Petrosian, a reassessment of values occurred, and on this occasion I was able to make use of their interesting ideas. 106 in the game Evans-Dorfman ( 1978), he would have retained a reasonable game. Game 280 Tarnowski-Botvinnik 9 10 Olympia'1. Leipzig 1960 Catalan Opening 1 2 3 4 5 d4 g3 .tg2 c4 cxd5 5 6 7 8 lllcJ lt:\f'J 0-0 .i.f4 h3 0--0 This move is of little use. Thanks to the tempo presented to him, Black is the - first to begin play in the centre. it)f6 d5 i.f5 c6 · 10 11 12 �cl g4 'Lle4 l::tc8 Now the point of White's 10th move becomes clear: he wants to maintain his bishop on the h2-b8 diagonal, even at the cost of weakening his kingside. In this situation the exchange in the centre practically cleprives White of any opening adva..�tage. cxd5 lllc6 e6 J.e7 12 13 • • . tDxe4 Jl.g6 ilxe4 In the event of 13 ... dxe4 14 tDe5 the d4 pawn cannot be taken in view of the loss of a piece: 14 . . . 'ii'xd4 15 tDxc6 'i'xdl 16 tDxe7+.. or 14 ... tiJxd4 15 J;.xc8 'i'xc8 16 'i'xd4 . The reader will probably already have noticed that the game has trans posed into a well-known variation of the Gliinfeld Defence with reversed colours, and hence with an ex'tra tempo for White. However, the drawback to his position is the .fact that his king's bishop is restricted by the pawn at d5, whereas Black's king's bishop has more of a future. If White had shown caution here and played 9 tDh4, as, for example, 14 tiJe5 White aims straightforwardly for simplification, hoping in this way to gain a draw. However, he fails to take account of certain subtleties in the position. 107 14 tLlxe5 15 i.xe5 15 16 17 �xg2 flb3 After tlie exchange 26 'i'xc4 dxc4 (26 ...bxc4 is also good) Black would immediately have · cr�ted a decisive pawn majority on the queep.side. 15 . .llxe4 was hardly any better, if only because of 15 . . . tbc6 and 16 ... �6. il.xg2 'Ifa5 b5! 26 27 28 Tlueatening 18 .. J:Ic4. White decides to concede the c-file, after which Black's advantage becomes obvious. 18 19 20 l:.xc8 a3 lldl l:I.d2 eJ i.gJ f3 i.f2 'ildJ • a4! 1fxd3 b4 . llxc8 l!Cc4 'lfb6 'ifc6 a5 �cl f6 3 3 l:ta8 i.e 1 he loses a piece. A textbook game. Such an ending could have been played in former times by Capablanca or Rubinstein . . . The first rank is also under Black's control. With his domination of the pos ition, he can permit himself the luxury of spending a tempo on this prophyl actic move. The blockaded white a- and b-pawns will soon come under attack. 25 26 . Forcing the exchange on b4, after which the white rook will no longer be able to return to the second rank. 28 ... !k2 was weaker on account of 29 b4! The start of a manoeuvre, aimed at securing complete control of the open file and continuing the pawn offensive on the queenside. 21 22 23 24 • e4 l:t.xd3 .i.xb4 29 axb4 exd5 30 exd5 31 l:leJ :.c2 White resigns. After 32 l:e8+ cJ;f7 Game 281 Neikirch-Botvinnik O�ympiad, Leipzig 1960 Sicilian Defence 1fc4 1 2 3 4 5 6 e4 tl)fJ d4 tL'lxd4 lllc3 i.c4 c5 l£lc6 cxd4 tL'lf6 d6 This, the Sozin Variation, is still popular, although its history dates back to the 6th USSR Championship (Odessa, 1929). · 6 • • • e6 Perhaps the soundest reply, although the more risky 6 ... 'ib6 is also popular. 7 108 i.b3 The other common continuation here is 7 .i.e3 i..e7 8 'i'e2 followed by queenside castling, which usually leads to mutual attacks on opposite wings. But my game with Padevsky (No.246) took a different course: 7 0--0 ..te7 8 ..te3 0--0 9 ilb3 l'.Da5 10 f4, which may be the strongest for White (but not for Black, who instead of 7 . . , i..e7 should have played 7 . . . a6, immediately creating counterplay on the queenside). 7 . . • if necessary be exchanged for the bishop. 9 10 11 And this is now a positional mistake . Since the move does not give White any tactical advantages, it plays into Black's hands (it opens the a8-hl diagonal). 11 0-0 il.e7 <it?hl • . lLle8 check for himself. 12 ktf3 12 fS is a move known to theory, when instead of 9 �hi White has played 9 ..t.e3 . Here, however, such an attack is unfavourable. 0-0 ::•r:,z�:z I��;::;�: 9 . If l l . ..dxe5 1 2 fxe5 �d7, then White wins by 1 3 ktxf7, as the reader can Nowadays 7 . . . a6 is more often played. for reasons explained in the previous note. 8 lLla5 b6 f4 e5 12 • • • lLlxb3 After 12 . . . i..b7 1 3 :h3 Black would have been unable either to exchange the bishop, or to consolidate with . . . g7-g6, since 1 3 . . t'Dxb3 would allow the inter position of 14 'l'h5 (14 . . . h6 1 5 t2Jxb3), and if 13 . . . g6 14 lDxe6 with a very strong attack. By his timely' elimination of the bishop, Black weakens the pressure on the a2-g8 diagonal. . A loss of time. which is not in accor dance with Sozin 's plan. White cannot now count on an opening advantage. 9 .i.e3 was better. transposing into a position from the afore-mentioned Game 246. Now the most convenient way for Black to mobilise his forces is by developing his bishop at b7, for which it is very useful for him to place his queen's knight at a5. From there it may 13 lt:\c6 13 14 15 tt:lxe7+ axbJ 1 3 axb3 i.b7 14 .1Ih3 g6 is also in Black's favour. White is counting on the opposite-colour bishops, but it is well known that this factor favours the side whose bishop is the more active. 1id7 1ixe7 f6! An important move. Black exchanges tl1e e5 pawn, and his seemingly passive knight comes into play, · coordinating excellently with his bishop. 109 16 exd6 White wrongly decides against the sacrifice of his e5 pawn. In this case his bishop would be activated, whereas in the game it remains blocked in by its own pawn at f4. 16 17 :dJ 18 lia4 20 lla5 20 21 �d6 21 22 ltxd6 20 l:ta l was more circumspect. .fl.b7 2 1 ltJc5 .i.xg2+ would have led to loss of material. lDxd6 lDf5 lDxd6 Played not for an attack, of course, but to transfer this rook to e l , i.e. for defence. 22 l:d8! With the deadly threat of 23 . . . 'i'c6. 1ffe8! 18 An imperceptible, but highly signifi cant move. Now the bishop moves onto the long diagonal, which was not possible earlier because of l:Ld7. 19 tne4 Black would also have retained an imposing positional advantage after 19 l:r.e4 .ilb7 . Now, however� he also gains a decisive attack� since� unfortunately for White, his rooks remain un connected. 19 b5! The rook has to move off the fourth rank, and it remains cut off from White's remaining forces. 23 Wd2 23 24 'ifxd6 lhd6 'i'd8 25 1fxe6+ .. 23 J:xd8 'i'xd8 24 'i'el (24 'i'xd8 l:xd8 25 .i.e3 l:td l + 26 ..tg l l:r.d2) 24 . . . Wd5 25 'i'e2 l:.d8 26 .ie3 would have led to the same won ending for Black as in the variation with 24 'i'xd8. Begi nning with this move, Black operates with two threats: mate on the back rank and the capture of the un protected rook. Soon the�e threats will be impossible to parr�y-siinultaneously. Or 25 'i'd2 'i'xd2 26 .ixd2 r!d8, which would have led to a position examined earlier. l lO 25 ltti 26 Wet l:te7 7 8 The fate of the game is decided: Black's pieces· �e; acting concertedly, · whereas White�s.are disunited. • • • 0-0 .i.d2 Here too 8 a3 is better, in order to answer 8 . . . i.d6 with 9 e4 1 8 ... •e1 · It will be recalled that in Game 236 Black decided to retain his bishop, and he immediately retreated it to d6. The two moves are more or less equivalent. ·· ' White resigns. The concluding position is interesting for the fact that, although all Black's pieces are on the edge of the board, they are dominant. From the standpoint of chess composition, the artistic effect would be stronger, if the black bishop were at a8, and the rook at e8. Game 282 Portisch-Botvinnik O�ympiad, Leipzig 1960 Queen's Gambit 1 2 3 4 5 6 7 d4 c4 ltlf3 lDc3 eJ i.d3 0-0 ltlf6 e6 d5 c6 ltlbd7 .i.b4 7 a3, as I played against Euwe (No. 175), is more energetic. 9 /l)e5 9 10 11 12 13 cxd5 al i.xc3 dxe5 Only after 9 'ife l , which leads to more complicated play, does theory consider (on the basis of a game Ivkov Kolarov, 1957) that White can retain a slight opening advantage. Now, how ever, far-reaching simplification begins. c5 exd5 .i.xc3 �xe5 llie4 If 1 3 lbg4 there could have followed 14 i.e2 (if there is nothing better) 14 . . ltJxe5 1 5 'ifxd5 with an obvious advantage to White. Now he is forced to part with one of his bishops, and he prefers to retain the more promising one at c3 . ... . 111 14 il.xe4 15 �ht dxe4 This move does nothing to improve the placing of the white pieces, and therefore it is a waste of time. The only way to fight for the retention of the initiative was by 1 5 'i'h5, controlling the f5 square and intending f2-f4. If the white pawns had been able to advance, this could have caused Black some problems. 15 • • . 1fa4 ltadl 18 19 20 exd6 ltdl %hd6 11xd6 a5! White, gaining a tempo by the attack on the e4 pawn, wants to win the battle for the open central file. However, he no longer has control of the g4 square, and the bishop will feel perfectly safe at f5. 16 17 :d6! The final subtlety. After 20 'i'xaS J:.d8 Black would have occupied the d-file. Draw agreed. Black exploits the time granted to him, and brings his queen's rook into play, in order to have the opportunity of switching it to the kingside. After 1 5 . . . .i.fS 16 f4 ! the black bishop could have become a target. However, now too it would have been better for White to play 16 f4, forcing 16 . . . exf3 . 16 18 At just the right time. White prevents the switching of the enemy rook to the kingside. .i.f5 .:a6 Game 283 Botvinnik-Schmid Olympiad, Leipzig 1960 King's Indian Defence 1 d4 c5 2 3 4 d5 e4 ltlf3 d6 .g6 The young Lothar Schmid was noted for being very enterprising, and he al ways aimed for a. lively game. Thus here Black endeavours to seize the - initiative from the very first move. After c2-c4 (now or a little later) it is harder for Black to create piece play, 1 12 but is this something that White should avoid? After weighing up everything 'for' and 'against', I was tempted by the possibility of obtaining some training in a piece battle, which I would certainly be faced with in a few months' time in my World Championship return match with Tai. Therefore the c2 pawn was left in its place. 4 5 6 .i.e2 tLlc3 Black appears to have achieved much, but the insecure position of his king in the centre allows White to land a spectacular counter-blow. 11 .i. g7 tLlf6 tLla6 Now (after 5 .te2) this move is pos sible, since if White wanted to double the enemy a-pawns, for this he would not only have to exchange his bishop for a knight, but it would also lose a tempo. And the black knight will be excellently placed at c7, supporting . . . b7-b5-b4 and attacking the white d5 pawn, which hinders e4-e5. 7 8 0-0 a4 tLlc7 a6 Fully in keeping with Schmid' s style. By saving time on castling, he at once begins active play on the queenside. 9 10 ttld2 ttlc4 e5! My opponent had not been afraid of this, thinking that after 1 1 . . .bxc4 12 exf6 .i.xf6 he was not threatened in ·any way. But when he continued the variation with 1 3 ilh6, he easily established that Black's defences would be disorganised. He had to go in for the exchange in the centre, allowing the fur ther activation of White' s king's knight. 11 12 • • • axb5 dxe5 axb5 12 . . .l£lxb5 was more circumspect, not allowing d5-d6 and d6xc7. But could the obstinate Lothar be expected to deviate from his intended plan? 13 14 l:txa8 l£ixe5 1fxa8 b4 Again (as after 10 . . . b5) it appears that Black has achieved his goal (indee� where can the knight go to?). il.d7 b5 15 d6! And again White finds an elegant reply, which unexpectedly changes the 1 13 evaluation of the position. First he left his king's .lqiight en prise. and now he does not 'move away his queen's knight.'. . : the simpler 23 'it'a4, when the black king becomes easy boGty for the white pieces. ; 15 . . • bxc3 One cannot, of course, object to the fact that the Encyclopaedia attaches a question mark to this move, but in the event of 1 5 . . . exd6 16 'i'xd6 bxc3 I thought that I would win with 17 .i.f3 'i'c8 ( 1 7 . . . 'i'a6 1 8 ti)c6) 1 8 l/_Jxd7 ti)xd7 19 .tg5. Later it was recommended that White should fir.st follow a game Larsen-Szabo ( 197 1): 17 .1ixc7 0-0 18 li:)xd7 li:)xd7. and then try by 1 9 b3 to exploit his minimal advantage. Without any evidence, it is hard to agree with this re".'evaluation of the position. I still consider that tl1e varia tion given by me, ending with 19 il.g5, is convincing enough. . 16 . dxc7 l7 .i.f4 23 24 25 26 27 28 29 30 lic8 Now the c7 pawn is invulnerable. 17 18 lDxd7 cxb2 lDxd7 19 .i.b5 J.d4 20 21 22 cJ cxd4 .txd7+ e5 exf4 1 8 . . . 1i'xd7 would also have been answered by 19 .ib5. White was threatening 20 'i'xd7+ 'i'xd7 2 1 c8'if mate. The simplest way to win. 22 23 . • • 'ilfe2+ 'iixd7 Only here, when the position is now absolutely won, can White be re proached for the fact that he did not find •e5 <if;f8 <it>g8 f6 l:.bl '&xc5 �g7 lte8 :xb2 :bt f3 gxf3 Wh3 W'c6 Black resigns It was no accident that this game appealed to Tal. He rated it highly, since it was conducted in his style. The Olympiad in Leipzig was the only open training event for the par ticipants in the coming match in the Spring of 196 1 . After this it only remained for them to prepare in the quiet of their studies for the return match, which by a FIPE · decision of 1 959 was being held for . the last time. 1 14 This move too was not obligatory. Now White advantageously places his knight at h4 and prepares f2-f4, espec ially as . . . g6-g5 is bad because of lDr5 . Game 284 Botvinnik-Pachman OlympifJld, Leipzig 1960 King·' s Indian Attack 1 2 3 4 5 6 7 gl tilf3 -*.g2 0-0 dJ tilbd2 e4 9 10 11 d5 g6 -*.g7 e5 tlle7 tllb4 f4 f5 .i.e6 tild7 White provokes a weakening of the g6 square and secures control of the h3-c8 diagonal. 0-0 11 12 13 fxg6 -*.bJ -*.f7 bxg6 tllc5 Black brings up this knight to the defence of his kingside, at which the enemy pieces are aimed. 14 15 16 A slight chance for White in this King's Indian position with colours reversed is provided not only by his extra tempo, but also by the poor position of the black knight at e7 (compare it, for example, with the position from Game 201 Botvinnik O'Kelly, where Black played . . . tiJf6 and . . . ttJc6, and . . . e7-e5 only later). Even so, if Black had quietly played 7 . . . tiJbc6, it would not have been easy for White to find a good plan. After Black's next move this problem is simplified. 7 8 a4 d4 f6 b4 tllc4 .i.xg5 lile6 tilg5 fxg5 It may appear that in this way Black eases his defence. In reality, the position with opposite-colour bishops that soon arises is advantageous to White, since Black's bishop is restricted and his pawns are weakened. 17 �f3 .lxc4 Otherwise the e5 _pawn cannot be defended. 1 15 18 19 dxc4 ..d2 i.f6 <:/;g7 25 And now White carries out a manoeuvre which I first noticed in a similar situation in a game of Capablanca (playing his knight to d3 ). 20 thel ! After 2 0 ltlxg5 Axg5 2 1 'i'xg5 ltlc6 22 W'xd8 (22 'ii'd2 a5) 22 ...llaxd8 23 llab l Z!xfl + 24 �xfl a5 (or 24 ... d3), Black, despite being a pawn up, would have quite good drawing chances. White avoids this continuation, preferring to maintain a positional advantage. 20 21 • • • lhd3 a5 b6 White was threatening ttJc5-e6. 22 c5 In this way White retains a good pawn formation. 22 23 l£.c6 b5 cxb6 This preparatory action is essential:i after 25 'ii'xb4 Black would have regained his pawn (25 . . . !Le7). 25 26 27 28 29 1fc4 �g2 .i.e6 30 31 32 ltxfl ltxf8 1fd5 33 ii.g4 1ffxb4 cxb6 ite7 Ji.cs · 1fd6 llxfl The exchange of rooks does not ease Black's position, but perhaps even makes it worse, since his rear is weakened. l:tf8 1f xf8 �h6 32 . . . d3 33 'i'xd3 'i'f2+ 34 �h3 would not have given Black anything, since the fl square is defended. But here after 33 'i'xe5 in the new situation 33 . . . d3 would have saved Black. 33 • • • 1Wd6 Black is pinning his hopes on the endgame. but they prove unjustified. 34 35 23 . • . thb4 Black sacrifices a pawn, in order to exchange knights, and with opposite colour bishops to restrict the opponent's initiative. 24 ltlxb4 axb4 1 16 <t>f3 1f xd6 @g7 .i.xd6 36 �e2 36 37 38 39 <iit>d3 �c4 <iifi> d5 The king immediately heads for the queenside, o! course, and Black cannot gain control of the d5 square, since in this case he loses his g6 pawn. i.b4 <iit>f6 .tel .i.b4 Here and later Black commits an inaccuracy, by allowing the white bishop to penetrate without difficulty to e8. He should have played 39 . . . .i.c3 (indirectly defending the e5 pawn), in order after 40 il.d7 to have the reply 40 . . . <it>e7. 40 ..id7 Grune 285 Botvinnik-Tal Ttflorld Championship Return Match Afoscow 1961, 1st game Nimzo-Indian Defence i.el 40 . . . �e7 was nevertheless better, although after 4 1 �c6 .i.a5 42 �c7 � (or 42 . . . i.b4 43 �xb6 ! �xd7 44 a5 �c8 45 a6 @b8 46 a7+ �a8 47 <it>a6, and Black loses) 43 �d8 White's bishop would have reached e8, which, as will be seen, would have ensured the win. . .l\cJ 41 i.e8 42 There could have followed 42 . . . .ta5 43 <J;c7 @g7 (or 43 . . . g4 44 i.d7 �g5 45 .te6 'iii>h 5 46 i.c4 'iii>g5 4 7 .i.e2 'iii>h5 48 h3) 44 <iti>d6 �6 45 i.xg6 ! (this is . where the weakness of the g6 pawn tells) 45 . . . @xg6 46 'itixe5 ..ic3 47 'itie6 g4 48 e5 .i.b4 49 <it>d7 <ttf7 50 e6+ � 5 1 a5, and the black bishop cannot sJop the widely-separated passed pawns. <iit>c6 White sealed this move, and Black resigned without resuming. 1 2 3 4 5 6 7 c4 lDcJ d4 eJ i.d3 a3 i.xc4 lDf6 e6 .i.b4 0-0 d5 dxc4 It is well known that 7 i.xh7+ 'iti>xh7 8 axb4 lDc6 does not give White any advantage. 7 . • • ..i.d6 As was shown by my games with Barcza (Budapest 1 952) and Furman (Moscow 1955), after 7 . . . i.xc3+ 8 bxc3 c5 9 ilb2 White has a clear advantage. 8 lDfJ 8 9 . lDb5 8 f4, which was tried in Game 346, proves unsuccessful after the energetic reply 8 . . . c5. • • lDc6 A rather naive attempt to refute the variation. Correct here is 9 b4, as played in the 3rd game of the match (No.286). 1 17 Naturally, he chooses a more compli cated continuation. 9 • • • e5 The most active continuation. 10 lDxd6 16 10 dxe5, which occurred in the game Ilivitsky-Lipnitsky ( 1 952), seems more logical. 'ifxd6! 10 The right way! After 10 . . . cxd6 1 1 dxe5 ttJxe5 ( 1 1 . . .dxeS 1 2 'i'xd8 -'lxd8 1 3 ttJg5) 12 i.e2 White has the advan tage. But now the exchange on e5, which he cannot avoid. involves the loss of castling and Of several tempi. 11 12 13 14 dxe5 @xdl <it>e2 .i.d5 'ifxdl+ lDg4 lDcxe5 Played in order to hinder . . . b7-b6 possible. 14 15 • • • .i.e4 if 17 18 19 20 c6 .i.e6 White is behind in development and, despite the two bishops, his position is clearly inferior. He could of course have played for a draw: 16 t:Dxe5 l'.bxe5 1 7 f4 .i.c4+ 18 � lt\d3+ 1 9 �.xd3 .i.xd3 20 i..d 2, but · in this variation he is completely deprived of counterplay. lDd2 llad8 White nevertheless has a pawn majority in the centre, and in order to hinder its exploitation, Black should definitely have played 1 6 . . . f5 . Then in some cases he would also have been threatening . . . f5-f4, which would have been highly unpleasant for White in view of the dangerous position of his king in the centre. Now, however, making use of the respite afforded him, White completes his development, and in the approaching endgame the position of his king is transformed from a negative factor into a positive one. h3 .i.c2 b3 .:d1 lDf6 lld7 ll.fd8 It only remains fo� i.b2 to be played, and Black will · be in difficulties. Therefore Tal is forced to exchange the bishop at c2 . . 20 21 1 18 .i.xd3 thdJ :xdJ 22 .i.b2 White, however, decided to exchange both pairs of rooks, since he assumed (not without reason) that in the ending with knight against bishop his winning chances would be considerable. :Jd7 Of course, not 22 ... �xb3, which would have lost a piece, but the move in the game also indicates that Black is underestimating the �gers of his position. 22 ... cs was essential, activ ating his queenside pawns and fixing the weakness of the b3 pawn. 23 24 .i.xf6 b4 ttlb3 · 26 27 28 <!>et ltacl f3 b6 .t.e4 :xdl+ ttxdl + 30 ..td5 The last chance was 30 ... .tfS, in order to defend the c6 pawn from d7. Now the knight breaks into Black's position and begins wreaking havoc there. ..tf5 ..td3+ A temporary and insignificant achievement. It is clear that the bishop will soon be forced to retreat. :xdl <i>xdl And so, the planned exchange of heary pieces has taken place. gxf6 How the position has changed within just a couple of moves! Black is left with weak pawns, and the ending (with or without the rooks) will be difficult for him. Now it would perhaps have been best to try and restrict the enemy knight by 24 ... �d5. 24 25 28 29 30 31 lDd4 c5 32 33 bxc5 lDb5 bxc5 a6 3 2 e4 was threatened. Or 33 ... as 34 ltJc3 ..i.c6 35 <it>d2. and. the white king penetrates unhindered to c4. 34 35 36 37 38 It is possible that 28 �xd7 :!xd7 29 'it>e2 (but not 29 f3 iii view of 29 . . . :d3) 29 . . . .i.d3+ 30 <it>f3 was even stronger. 1 19 lDc7 lDe8 h4 ttld6 gJ -*.c4 f5 @f8 .i.fl All Black' s pawns are isolated, and his bishop is unable to defend them. Before beginning the 'gathering of the harvest', · White places nearly all his pawns on dark squares. 38 39 40 41 was already at b3, thanks to the fact that the bishop had reached c4 not in two moves, but in one. <tle7 lhxf5+ <tle6 e4 <ties <it>d2 Black resigns Game 286 Botvinnik-Tal World Championship Return Match Moscow 1 961, 3rd game Nimzo-Indian Defence 1 2 3 4 5 6 7 c4 lhc3 d4 e3 li.d3 a3 .i.xc4 9 10 lElfJ b4 • e5 .i.g4 • .i.b2 After 10 . . . a6 1 1 .i.e2 White would have retained some advantage (Botvinnik-Hecht, Belgrade 1 969). lhf6 e6 .i.b4 0-0 d5 dxc4 .i.d6 11 d5 In the 5th game I chose a more logical continuation - 1 1 dxe5 thxe5 1 2 .i.e2, but here I was testing my pre match preparations (incidentally, instead of 1 1 . . . tDxe5, Konstantinopolsky re commended 1 1 ....txe5 for Black). It was no accident that Tai employed this variation of the Nimzo-Indian Defence against me. Black gives up his pawn centre, aiming for piece play. 8 9 • 11 12 /i)c6 The simplest way of developing the queen's bishop. Incidentally, White does not achieve any advantage with either 9 ltlb5 (as I played against Tai earlier in the same match - No.285), or 9 i.b5. In order to appreciate how old this variation is, it will be recalled that this latter move (9 .i.b5) occurred back in the game Capablanca-Ragozin (Mos cow 1 936). Only, there the white queen CD.e7 il.d7 h3 12 . . . i.h5 would have led to more complicated play, but, since Tai guessed that I was prepared for this variation, he chooses a less obvious move. In addi tion, Black is aiming to play his knight to g6, and in this case the position of the bishop at h5 will be unfavourable in view of g2-g4. · 13 lhg5 thg6 While the lmight was at e7, Black could have safely made the prophylactic 1 20 move 1 3 . . . h6. Now, however, White carries out a tactical operation, ex changing the· enemy light-square bishop and gaining a significant positional advantage.. 14 15 lDe6 dxe6 An important move, restricting still further the opponent's pieces. White can easily defend the f4 square, and if Black wants to transfer a knight to d4, which has been left without pawn control, an exchange on this square may merely increase White's positional advantage. fxe6 'ifi>h8 With the knight at e7, here the bishop would have been able to move. 16 17 18 exd7 0-0 'ilfxd7 llf5 ifg4 ifc2 Black, of course, avoids the exchange of queens, since in the endgame White's advantage would be obvious. 20 _21 . 22 23 'ife2 1fg4 WeZ e4 'W'f5 1fc2 11ff5 27 28 29 1fh5 1!fe2 lDeJ! ltlg8 ltl6e7 .. lDg4! lDh6 l'.Llxg4 hxg4 ltlc6 1fd7 :ad8 'ife8 lDh6 White now avoids the exchange of knights on d5, since the exchange of the same pieces on g4 is even more favourable to him. lDg8 For the same reason� Black avoids the exchange. 19 l:tadl ifg4 g3 There is no way for Black to achieve any activity. If, for example, 26 . . . c6 27 lbe3 lbf6, then 28 1i'f3 . lDd5 By offering the exchange of knights, White aims to gain control of the impor tant squares d5, e4 and g4, which will suppress the activity of the opponent's pieces. 18 23 24 25 26 29 30 • This is equivalent to suicide, since it opens the h-file. 30 . . . ifb5 3 1 . �g2 (3 1 'Dxe5 ifxh3) 3 1 . . .lLlc6 really would have been better. 31 121 32 <i!?g2 33 -*.d5! 33 ... �d4 34 35 .*.xd4 .i.c4! exd4 ile7 The exchange of rooks would have hindered White's attack on the enemy king. Tai was unable to refrain from this 'activity'. but the resulting opposite colour bishops are in White's favour, since Black has no way of opposing the activity of the light-square bishop. The afore-mentioned plan has been carried out unhindered, and now with every move the threats intensify. 41 �g7 42 43 ... c5 36 37 b5 f4 .i.f6 d3 38 39 40 41 l:lxd3 i.xdJ e5 :ht l:txd3 .i.d4 g6 Game 287 Botvinnik-Tal World Championship Return A1atch Moscow 1 961, 7th game Nimzo-lndian Defence After this Black has some slight con solation: two connected passed pawns, but they are blockaded. Black sacrifices a pawn, in order to ex.change one pair of rooks. b6 The sealed move. After 43 . . . 'ifd7 White wins by 44 'i'c6 'ii'xc6+ 45 bxc6 l'k8 46 e6, while if 43 . . . 'fie7 44 g5 l:tc8 (otherwise 45 'i'c6 and 46 'i'f6+ ! ) 45 f5 gxf5 46 l:txh7+ �xh7 47 ii'h4+ rl;g7 48 1i'h6 mate. Black resigns . The simplest. It only remains for White to play i..d3 , f2-f4, l:thl and e4-e5 . 35 'ife4 .i.c4 1 2 3 4 c4 �CJ d4 a3 �f6 e6 i.b4 In the earlier games ·or the return match I chose 4 e3 . Now I decided to check what my opponent had prepared in this variation, which had occurred 122 several times in the games from our previous match. 4 5 .. bxcJ . i:.g5 h6 9 i.h4 �xc4 10 .ixc4 dxc4 1 1 'i'a4+. 8 9 10 i.xcJ+ b6 Black avoids 5 . . . liJe4 6 e3 f5 7 'i'h5+ (No. 278). A good alternative is 5 . . . c5, as played, for example, in Game 100 and · also a training game with Kan (see this volume). 6 7 .i.xft exd5 A plan typical . of such positions, which I employed, for example, against Tal in the 4th game of our first match, although in a somewhat different situati9n. il.a6 f3 e4 �xfl i.g5 10 h6 The pawn exchange 10 . . .dxe4 1 1 fxe4 would have given White pre dominance in the centre and the possibility of conveniently developing his pieces at f3. Tai tries to complicate matters with a pawn sacrifice ( 1 1 i.xf6 1i'xf6 - of course, not l l . . .gxf6 - 12 exd5 0-0, with a comfortable game for Black). True to my match tactics, I declined the pawn sacrifice, so as not to concede the initiative to my opponent. 7 11 • • • This move can be considered a positional mistake, since it allows White to immediately get rid of his weak c4 pawn. I have to admit that already then I knew that 7 . . . liJc6 gives Black a good game, and after 8 .id3 t2Ja5 9 'i'e2 d6 ! followed by . . . 'i'd7-a4 (or c6) his position is perfectly sound. One gains the impression that, from the standpoint of opening · preparation, my opponent had not made very fruitful use of the ten months separating our second match from the first. 8 1fa4+ d5 cxd5 11 c6 Avoiding the endgame, since after l 1 . . . 'i'd7 1 2 'i'xd7+ liJbxd7 13 i.xf6 (or 1 3 i.h4 dxe4 14 �e l ) 1 3 . liJxf6 14 e5 Another plan was carried out in the game Alekhine-Eliskases ( 1 93 3/34) : 8 . 123 . White's advantage is obvious. But the course chosen by Black leads to a difficult middlegame. Now the bishop can be retained, and Black will have no time to think about occupying c4 with one of his knights. 12 J.h4 Of course, here too White avoids 12 i.xf6 'i'xf6 1 3 exd5 0-0. A dxe4 12 highly dubious decision. The opening of lines in the centre merely assists the development of White's initiative. 13 :et g5 This is now forced. 14 .tf2 The only move. In the event of the routine 14 !atg3 Black's position would not have been so bad after 14 . . . 'i'd5. But now if 14 . . . 'ii'd5 there can follow 1 5 c4 (the d4 pawn is defended) 15 . . . "i'e6 16 'ifc2 (16 fxe4 tlJxe4 17 ifc2 f5 18 g4 is unconvmcmg on account of 1 8 . . . �d8 ! 1 9 gxf5 Wxc4+ 20 'i'xc4 tlJd2+) and 17 fxe4 with a decisive advantage for White. This is the subtle point of the manoeuvre, begun with 1 2 i..h4. 14 15 • • • White aims as quickly as possible to bring into play his last reserve, the rook at h l . If now 1 7 . . . g4 1 8 fxe4, and g4 remains inaccessible to the black knight. After 17 . . . l::tg8 (as should have been played) 18 hxg5 hxg5 White's king's rook is immediately activated. Black's reply also allows the bishop to come into play with decisive effect. 17 18 19 1!fe7 lLle2 ... 'i'c2 b5 1ixa3 If 16 . . . e3, then 1 7 i.xe3 CLld5 1 8 .te l . Convinced that the positional battle is lost, Black tries at least to retain a material adv{Ultage. 17 gxh4 ttlbd7 0-0-0 On the queenside too the king will not find a quiet refuge, and loss of material for Black becomes inevitable. But not 1 5 fxe4 in view of 1 5 . . . tL1xe4 16 'ifc2 f5 17 g4 t'Dd7 1 8 gxf5 �df6. 15 16 .i.xh4 ttlgJ h4 1 24 20 tLlxe4 20 . . . ttJxe4 is not possible on account of 2 1 'i'xe4 with two threats: 22 Wxc6+ and 22 ..i.xd8. Therefore Black offers a piece sacrifice. 20 21 lthe8 �! White is not in a hurry to gain material� After 2 1 ctJxf6 J:.xe 1 + 22 ct>xel 'i'al + 23 'i'd l 'ili'xc3+ 24 <it>fl ctJe5 some complications would have arisen. It is more important to retain the initiative. 21 22 23 24 fxe4 l:lal ltxa7 ttlxe4+ f6 11fe7 'If xe4 1!f xe4 %la8+ llxe4 tLlb8 27 .igJ 28 11.hal �b7 • • • 30 31 32 33 b4 .id6 bxc3 <iiftb5 .i.c5+ lila4 Black resigns From the character of the play, one might imagine that it was Tal playing White, and Botvinnik playing Black. World Championship Return Match Moscow 1961, 9th game English Opening 1 2 3 c4 li)cJ g3 lDf6 e5 c6 4 5 6 tLlfJ lDd4 cxd5 e4 d5 1fb6 7 8 ihb3 -*.g2 cxd5 A well-known continuation of Keres. Although White can hope for an open ing advantage, the play becomes sharp, which always suits Tai. My opponent was in a fighting mood, since in the preceding game of our match he had won in good style. At that time this was almost an innovation. It had been played not long before this match by Mikenas against Taimanov ( 1 960). In that game there followed 7 ltJc2, which is more common even now, although the continuation chosen by me is sound enough. llc8 Defending against the threat of 2 9 % Ua7+ �b6 30 i.c7 mate. 29 , l:l8a7+ = Gatne 288 Of course, not 26 . .. �c7 27 li.g3+. 28 .i.xb8 Botvinnik-Tal The exchange of queens does not weaken White's attack. 25 26 30 Again threatening mate (3 1 l:t la6). Black loses a piece. �b6 125 i.xc3+ 1 5 @fl .i.xd4, sine� the queen is won at too high a cosL · · ' 'ifxb4 14 . 15 . • .i.xf6 Of course, not 1 5 dxe4 tt:Jc6. 15 16 8 • • • d3 .i.eJ llld4 12 lLlc2! a4 1fb4 al No better was l l . . . 'ifxb2, to which the simplest reply would have been 12 0-0. An obvious refutation of the rook' s pawn's swift advance. 12 13 . . • ii.d4 The diagram demonstrates clearly the failure of Black' s play. 16 17 avoids 14 :b l .i.e6 l:lcl 17 'iic l , suggested by Bronstein, would seem to be more logical, in order to use the rook on the b-file (e. g. l 7 . . . tt:Jc6 18 :b l 'fie7 19 dxe4 dxe4 20 tt:Jxe4, and the threat of 2 1 l:Xb7 can hardly be parried). However, White would have had to reckon with the 11• possible reply 1 7 . . . 'ifb2 At the same time, as will be seen from what follows, the move in the game by no means throws away the win. 17 18 19 'ilfxb2 �b4 Only in this way can the 'would-be attacker' queen be saved. 14 · /i)xb4 Naturally, White gxf6 a5 In view of Black's retarded develop ment, he had no justification for wasting two tempi on this and the further ad vance of his pawn. Such an ' initiative' can be parried without difficulty. The correct plan of development was demonstrated by Black in the game Reshevsky-Keres ( 1 963): 8 . .. .i.f5 9 d3 i.b4 10 0-0 .i.xc3 1 1 bxc3 0-0 12 .i.e3 'i'c7, but the quiet play resulting after 8 . . . ..i.f5 evidently did not suit the World Champion. 9 10 11 0-0 dxe4 lllxe4 lllc6 dxe4 After 1 9 tt:Jd5 'fid4 20 Cl1c7+ cJite7 2 1 t2Jxa8 (the outcome would also be unclear after 2 1 :xc6 'i'xd l 22 lixe6+ 1 26 23 a-pawn would have caused White con siderable problems. 19 • • • .i.xa2 The boldness of desperation Other wise Black would simply have been a pawn down and in a difficult position. � 20 lhd6+ 21 22 lhxb7 lhc5 24 25 26 22 .. J::tc 8 was completely bad in view of 23 ifd6+ <iftg8 24 ttJe6 'i'b8 25 :xc8+ 'ifxc8 26 ltJf4, creating numer ous threats (27 'i'xa3, 27 'ifxf6 etc.). In a difficult position Tal finds a clever chance. The main aim of the move is not only to move the rook away from the attack by the bishop, but also to defend the queen in the variation 23 'i'd6+ <i;g7 24 ltJe6+. lha6 • • .i.b3! ltlxb4 l:lfxdl :at .i.xdl llxb4 lhe5 l:tb8 26 23 • White cannot avoid the exchange of queens. The tempting 23 f4 ltJg4 24 e4 (or 24 tiJd7+ '3;;g7 25 ltJxb8 l:txb8 ! ) 24 ... ltJe3 2 5 'i'd6+ <:/;;g7 26 e5 did not hold any great danger for Black, since instead of 24 . . . l'De3 he would reply 24 ... h5 ! White takes the correct decision to go into an endgame, where he has every chance of winning, since the a3 pawn is doomed. • • • �b2 Was it possible to make things more difficult for White, by aiming to give up the a3 pawn at the cost of the exchange of minor pieces? In a rook ending the conversion of a material advantage is not easy when the pawns are all on one wing. It turns out that White could have avoided such simplification, for example 26 ... .:a4 27 f4 ltlc4 28 1:.d4 (but not 28 l:td7 ttJe3) 28 . . . 'it>g7 29 i.c6 :b4 30 i.e4 ! (threatening 3 1 :xa3) 30 . . . l:a4 3 1 i.c2 llb4 32 i.d3 l:tc8 33 l:txa3. Therefore Black gives up the a3 pawn without a figh� and the game enters a purely technical phase, which, however, lasts a further 45 moves. But in such positions is it worth hurrying? 'l;g7 27 <:Ji;f1 28 lbaJ l:tc8 1 27 29 .i.e4 lle8 A rather surprising move, especially since it was apparently planned by Tai in his home analysis. 43 . . .t'Llc4 would have offered a more tenacious resist ance. If 29 . . . ltJc4 White would have won by 30 l:k3 ! (30 ltld2+ 3 1 l:txd2). . . . 30 l:la4 31 . .i.f5 32 l!b4 l:le7 ltc7 Forcing a weakening of the h-pawn. 32 33 l:t a4 34 h3 35 . .i.d3 44 The repetition of moves is explained convenient strong-point other than e5 . White, naturally, is aiming to reach the time control, to avoid taking any com mitting decisions before home analysis. .i.e4 .i.d3 .i.e4 <it>el 45 46 47 48 49 50 ttlc4 ttle5 ttlc4 The incautious 39 .i.d3 would have repeated three times, and so to a draw. llad4 l:lld2 lld1 f4 lLle5 llc3 :ct+ l:tcJ f5 l:t4d3 .i.g4 :b3 llcc2 lia2 <it>g6 ihd2 lhc4 ihd2 <it>f2 lle3 "1b3 The two extra pawns guarantee White a win. He only needs to be careful not to make a mistake, and for this it is useful to gain time for thought. led after 39 . . . ltle5 to the position being 39 40 41 42 43 ttlc4 gave up the pawn, Tai had in mind 44 . . . tlm+ 45 exf3 l:te3+ 46 � l:lxfJ+ 4 7 <it>g 1 l:t.xg3+ 48 lt>hl , but not without reason he avoided it, since Black has no compensation for the piece. Or perhaps he hoped that I would be afraid of this continuation? by the fact that Black's knight has no 36 37 38 39 .i xf5 It can only be assumed that, when he h6 %lc5 �c4 �e5 51 52 ttlc4 lte3 lie8 52 llb3 would have led to a repetition, allowing Black to claim a draw. 52 53 54 55 56 ttld2 c.t>f6 l:te5 Jlf5+ :Ie5 .i.h5 �g6 c.t>t"6 llc3 . The next time control has been reached, and White, after calm reflec tion, can begin advancing his pawns, without which he cannot get by. 57 h4 58 .to 59 ·JL d5 60 .i.f3 1 28 , llc4 .:cc2 lla4 . hiaa2 · 61 62 :et h5 l:!a4 J:l.c3 on the clock, and in addition, the advance g3-g4-g5 can no longer be avoided. These are more than sufficient reasons for Black to curtail his resistance. Game 289 Tal-Botvinnik Ulorld Championship Return Match Afoscow 1961, 1 0th game Caro-Kann Defence The black rooks have finally left the second rank, in order to hinder the advance of the white pawns. But at the same time this relieves the white pieces of any concern about the e2 pawn. 63 64 65 66 67 · .i.g2 l:r.dl .i.d5 <it>el .i.g2 :c2 %ta4 l1.d4 1bd4 <ii>f2 .i.e4 l:.f5+ lld5 @f3 ll:\bJ lbxd4 tDe6 :b2 rl;g7 'i&>f6 Black resigns Tai satisfied himself that White was not intending to inake any error. Besides, from the 73rd move the two players had each gained an extra hour e4 d4 e5 c6 d5 For many years this continuation was considered to be harmless for Black. White's next move reveals that it has some positional basis. 3 4 lla3 The pin inevitably leads to the exchange of one pair of rooks, which is useful for the conversion of the material advantage. 67 68 69 70 71 72 73 1 2 3 ... h4 .i.f5 h6 4 . . . h5, which looks more active, is also often played, as, for example, in the 14th game of the Return Match. However, the text move is quite logical. Black wants to play . . . e7-e6, after which he will be out of danger. There fore it is not surprising that White immediately takes vigorous action. 5 g4 Here Boleslavsky recommended the quieter continuation 5 �d3 .itxd3 6 'i'xd3 e6 7 h5. 5 ... .i.d7 After 5 . .Jth7 6 e6 fxe6 White would have had a certain compensation for the pawn, which was confirmed, in par ticular, in the game Gufeld-Spiridonov ( 1 96 1) : ·7 .td3 .ixd3 8 'i'xd3 'i'd6 9 f4. . . 6 1 29 h5 12 This too is useful, in order to prevent . . . h6-h5. However, in the 1 8th game (No. 293) Tal refrained from this move, restricting himself to the prophylactic 6 c3 , while in the same year Bronstein played 6 c4 against Portisch. Now Black must not delay his attack on the centre. 12 6 7 8 9 c3 i..h3 9 10 ... libJ 1'b6 Usually Tai avoids the exchange of queens, but here he changed his habit, and wrongly so, since in the given situation tl1e exchange favours Black. 10 11 iixb6 11 ... • • • cxd4 After 1 1 cxd4 �b4+ 12 tLlc3 liJxd4 1 3 'i'd 1 .i.c5 White would have lost a pawn. axb6 • . • l£la5 But now, of course. Black avoids the win of a pawn ( 1 2 . . . tbb4 1 3 �d2 l:xa2 1 4 I:txa2 ti:)xa2), since then the initiative would have passed to the opponent. Instead of this he rapidly advances his b-pawn, imitating the plan that Capa blanca demonstrated in his encounter with Janowski ( 1 9 16). This game was given by the third World Champion in his book Afv Chess Career. c5 l£lc6 e6 ..ieJ Inconsistent, to say the least. 9 f4 . came into consideration. to place the knight at f3, from where it would have securely defended the central d4 pawn. cxd4 lJ 14 lDcJ .ifl b5 b4 14 . . .t:Dc4 could also have been played, but Black has no objection to the white knight occupying an insecure position at b5 and even penetrating to d6 . All this will merely assist the mobilisation of Black· s forces. 15 l£lb5 �d8 A rather rare occurrence, when by the 1 6th move none of the kingside pieces of both White, and Black, have left their initial squares. . It is true that the white bishop left its post at fl for a · short ti.me, but then thought it best to return. 1 30 bxc3 b3. White is therefore forced to give up a pawn, so that his bishop can participate in the blocking of Black's queenside pawns. 16 ltlfJ lbc4 Again rejecting, and not without reason, the win of the a2 pawn (16 . . . lt:\b3 17 !!bl lixa2), since it would be hard to Black to bring his bishop at f8 into play. 17 .i.xc4 18 lt:\d6 23 24 25 26 27 dxc4 .i.xd6 There was no point in sacrificing the exchange (18 ... SLc6 19 �xf7+ �e8 20 lDxh8 i.x:f3 etc.), since in any case Black's position is good enough. 19 exd6 20 lt:\e5 exd5 lbf6 �d7 @e6 <bxd6 Here we can take stock: Black is a pawn up and has the better position. Tai' s subsequent clever attempts to complicate the play are unable to change the course of events. .i.c6 White appears to be creating compli cations, but in the end it all reduces to further simplification, and this is merely playing into Black's hands. The white knight proves to be out of play in Black's territory, and in general the World Champion's pieces become disunited. d5 .i.d4 �d2 {fjf7 lt:\e5 · 20 il..xh l 21 lbxf7+ �e8 22 li)xh8 .i.e4 There is the unpleasant threat of 23 . . . c3 24 a4 c3, which is not parried by 23 it.d2 on account of 23 . . . c3 24 13 1 28 f3 29 g5 30 h6 31 {fjf7+ 32 ltJxh6 .i.h7 hxg5 gxh6 @e6 ICigS lbg4 34 liJe3 35 l£ig4 .i.f5 .i.g6 ilf5 36 ltJe3 37 l£ig4 .i.g6 <it>d6 3 3 l:te 1 + was threatened, when Black loses a piece. 33 Now all Black's thoughts are aimed at safely escaping from time trouble. 38 the5 .*.f5 . quick draw. In so doing,. he quite reasonably assumed that. .White would not avoid the Exchaiig� Variation of the Slav Defence, but he failed·: to appreciate that it is not so easy for Black to gain a draw even in this variation, if he does not have a prepared plan of action. 4 cxd5 5 lf)tJ 6 .i.f4 cxd5 li)c6 .i.f5 Two experts on the Slav Defence, Emanuel Lasker and Vasily Smyslov, used to prefer 6 . e6, although in this case too Black does not have an easy game. .. Black unhesitatingly gives up his g5 pawn, since the passed f3 pawn is not dangerous. The outcome will be decided by the breakthrough of his pawns on the queenside, which White is unable to avoid. 7 e3 e6 39 lDt;+ 'it?d7 40 thxg5 the7 thc6 . 41 f4 42 ti)fJ <ii>d6 This last move was sealed by Black. White resigned, without resuming the game. 8 .i.b5 Game 290 Botvinnik-Tal World Championship Return Match Moscow 1961, 1 1th game Slav Defence · 1 d4 2 c4 3 li)cJ li)f6 c6 d5 A psychological mistake. After losing the two preceding games, Tal decided to take a breather and make a The system associated with this move was one that I prepared · back in 1 946, i.e. 15 years before the present game, but I had never managed to employ it in practice. Only late in 1947 could I have played this against Trifunovic in the Chigorin Memorial Tournament, but I refrained: the tournament situation was such that I was quite ·�ppy with a draw. There I played 8 'ib3 .tb4 9 �b5 0--0, and the chances quickly became equal. 132 It should be added, only, that 8 .i.b5 is by no means a new move. A game in which it o.ccurred was given in Griinfeld's Taschenbuch der Eroff nungen im Schach. There 8 . . . t'.Dd7 was recommended, and if 9 ifa4 l:tc8, after which Black's position is· not worse, since the acceptance of the sacrifice ( I O i.xc6 �xc6 1 1 'i'xa7 ii'.c8) i s not without its dangers for White. All this was established long ago, but I not unreasonably thought that my opponents would avoid the pawn sacrifice� since White was apparently ready to go in for this variation. This assumption proved justified . . . 8 .i.b4 'iia5 9 tiJe5 10 ..txc6+ bxc6 ..txc3 11 0-0 'fixc3 12 bxcJ Black continues to stick to his direct drawing tactics. However. it should be mentioned that Pomar' s attempt to improve Black' s play by 1 2 . . . :cs did not prove satisfactory in view of 13 c4 with the threat of c4-c5 (Game 340). Now the exchange of queens is in evitable, since if 1 3 . . . ..i.c2 there follows 14 J;lb l followed by 1 5 ::tb2 or 1 5 :b3 1 2 • Because o f this, th e white rook moves to the c-file with gain of tempo. 13 14 llfxcl This leads to a further deterioration of .Black ' s position. His main defensive trump . is the centralised position of his king. After removing it to the wing, apart from anything else he will have to reckon with the threat of ttJxc6. lLle7+ was and lLlxf5. Therefore 14 . . preferable. . tLlci.7 15 f3 16 lLlxc6 17 a4 h6 17 18 .i.d6 lDd7 lDb6 :res Exploiting his spatial advantage, White selects the a7 pawn as a target, since it will not be easy for the black bishop to come to its aid. 19 .i.c5 13 'lfcl! 'irxct 0-0 After Black has prepared to occupy c4 with his knight, the threat of ttJe7+ and ltJxf5 has lost its strength, since 133 White would have problems over the defence of his e3 pawn. However, for the momenh 9 .. ltlc4 is unfavourable on account of 20 e4 �g6, and now either 2 1 exd5 exd5 22 l'De7+ and 23 ft.)xd5, or simply 2 1 iJ.xa7. Black also has a difficult game after 19 . . . t'.Llc8 20 lla3 ! This means that he has to agree to the loss of a pawn, but at least he is able to give it up in such a way as to provoke the exchange of knights. Then the only minor pieces left on the board will be opposite-colour bishops. . 19 20 ltlxa7 21 .i.xb6 i.dJ rlxa7 l:ta6 Now White's plan becomes apparent: to play e4-e5 and exchange on f6, after which the advance of the f-. and g-pawns will enable him to achieve his goal. Although it is natural that Black should want to blockade the passed a-pawn, the rook is not well placed on this square. 22 a5 23 lla3 e4 <&t>e3 :.ac3 g4 ct>f2 28 29 -h4 30 h5 llab8 lic6 32 hxg6+ 33 :Jc2 <&t>xg6 fxe5 After this the white bishop also gains the opportunity to participate in the play on the kingside. 34 dxe5 %th8 35 lth2 l:tcc8 Of course, not 3 5 . . . h5 36 :chi . 36 @d2 iJ.b3 .J::t aa8 :eb8 r!c8 After strengthening the placing of his pieces and pawns on the queenside and in the centre, White intends to open up the game on the kingside, in the hope of either penetrating with his rooks into the opponent's rear, or of creating another passed pawn. llbc8 g6 Passive waiting is not to Tal 's liking, but the move played merely assists White's intentions. .i.c4 f6 This decision also cannot be approved. The development of White's initiative would have been hindered by 23 . . . f5 . 24 25 26 27 28 30 . . . 31 e5 But now the a-pawn will be able to advance directly to a7 . 37 38 39 40 a6 a7 ltal .i.e3 .i.c4 l:.h7 :as It only remains to 'piay l:tb l -b8 or l:tahl, but Black's next' move - his last in time trouble . - · 1e_qds to a different outcome. 134 40 l:r.b7 41 l:!xh6+ 'lig7 ; 42 l: ahl ktb2+ Black resigns • • l l . . .cxd5 1 2 .ixd5 lDc6, which was known from the game Boleslavsky Najdorf ( 1 953). Alas, such a prosaic decision does not suit Tai, and he plays for complications, i.e. for the worsening· of his own position. • {. � . · 11 12 i.b3 Game 291 Botvinnik-Tal b5 ilb7 World Championship Return Match Moscow 1961, 13th game King's Indian Defence 1 2 3 4 5 6 d4 c4 lDcJ e4 f3 .ie3 lDf6 g6 .i.g7 d6 0-0 e5 Whereas after the I 0th game my opponent was definitely aiming for a draw, after his victory in the 1 2th game he was in an aggressive mood, as indicated by his choice of opening. Taking this into account, White takes a psychologically correct decision: he first of all exchanges the queens. 13 0-0-0 If 1 3 l:.c 1, according to analysis by Boleslavsky, Black gains the better game with 1 3 . . . aS ! (avoiding traps such as 1 3 . . . cxd5 14 :c7 l:td7 1 5 l:xb7. or 1 3 . . . �d7 14 liJe2 cxd5 1 5 .ixd5 i.�d5 16 l:k8+ i.f8 17 exd5). Therefore White sets a more substantial trap - a positional one. 7 dxe5 Of course, 7 d5 or 7 l2Jge2 leads to more interesting play. 7 8 'ifxd8 9 lDd5 dxe5 l:txd8 13 White, naturally, avoids 9 ti:Jb5 lba6 1 0 l2Jxa7 lDb4 or 9 ftd l :xd l + 10 @xd l i.e6. which would have been in Black's favour. 9 • • • lDxd5 c6 If 9 . . . tDe8 the strongest is 10 0-0-0. 10 cxd5 11 ..tc4 • • • c5 To my surprise, Tai, makes the position a closed one, which is not in keeping with his style. Of course, he should not have left White with a protected passed pawn and restricted the possibilities of his queen's bishop. After 13 . . . as ! 14 dxc6 lLlxc6 1 5 .:t.xd8+ l:xd8 1 6 l2Je2 a toughly equal position would have been reached. 14 .i.c2 Inviting Black to choose the variation 135 White's plan was not difficult to find. It was · successfully implemented in a similar �osi_tion in a training game Boleslavsky-Botvinnik (Voronovo 1 952), and consisted in the preparation of an attack on the black pawn chain by b2-b3 and a2-a4. It is surprising that, although White does not conceal his intentions, my opponent essentially does not even attempt to counter them. 14 15 ttle2 16 ltlc3 tl)d7 .i.f8 a6 Perhaps the losing move. The bishop at b7 is shut out of play for a long time, and, more important, White is able to carry out his idea without hindrance. Black should have played 16 . . .b4 17 �a4 c4 followed by ... ..ta6, leading to unclear complications. 17 b3 18 il.d3 19 .i.e2 l:tac8 ltlb6 :l d6 Now the threat is parried: in the event of 20 . . .b4 2 1 ttJbl c4 22 bxc4 ttJxc4+ 23 .i.xc4 :xc4 24 .tk l white breaks through on the c-file. · 20 21 llcl 22 a4 f5 ltf6 bxa4 Since after 22 . . . b4 23 ttJb l Black is left with numerous weaknesses, he takes another decision, also, however, a painful one: he opens the b-file, which will soon be occupied by White. 23 bxa4 White, of course, does not exchange knights (23 tt:Jxa4 ttJxa4+ 24 bxa4), since the black knight is badly placed. 23 . . . a5 23 . . . c4 24 'tt>c2 �b4 25 l:.b l .i.xc3 would not have changed anything, since both 26 ..i.xb6 and 26 'it>xc3 �xa4+ 27 �c2 are good for White. 24 <iftc2 c4 .i.b4 25 ltbl 26 ltla2 Forcing Black to unblock the open file. 26 ..ic5 27 .i.xc5 :.xc5 This move creates the concrete threat of 20 . . . b4 2 1 lDb l c4 (the knight at b6 is defended), but the rook at d6 is no adornment to the position. 20 @b2 136 I saw that by 28 f4 I could gain a material advantage, since it is hard for Black to defend his minor pieces (28 . . . fxe4 29 fxe5 30 �d.2 .:Ixd.5+ 3 1 @e3 etc.). But at this tense moment, in the fifth hour of play' r did not want to allow my opponent even the slightest tactical counter-chances. �f2 28 29 l:tb2 • • Slightly more tenacious was 29 . . . fxe4 30 fxe4 (30 ttlxe4 JL.f5) 30 ... ..td7 3 1 :hb 1 .ixa4+ 3 2 tiJxa4 ttlxa4 3 3 l:tb8+ l:f8 (in the game on the 32nd move this could not be played because of 3 3 f4). Even so, after 3 4 d6 l:kc8 3 5 l:.xc8 l:xc8 36 l:tb7 there is no doubt that White is winning. 30 l:t.hbl ltti l:l.xb7 'it>f6 l:tc8 37 38 39 40 41 d7 rld8 .lxc4 ltlc5 :n+ @g5 .tb5 fxe4 fxe4 Black resigns Game 292 Botvinnik-Tal World Championship Return Match Moscow 1961, 1 5th game King's Indian Defence ..ixa4+ If 30 . . . ttlc8 there would now have followed 3 1 f4 - a possibility which would have been excluded by the preparatory exchange 29 . . . fxe4 30 fae4. 1 2 3 4 5 6 ltlxa4 31 ltlxa4 32 l:tb8+ .:1b7+ d6 �xb7+ 11xh7 Or 3 6 . . . l:.c6 37 d7 lld6 3 8 hc4 ttlb6 39 l:If7+ �g5 40 i.b5. .i.c8 .i.d7 • 33 34 35 36 d4 c4 lhc3 e4 f3 lhf6 g6 .i.g7 d6 0-0 c6 ..te3 Avoiding the exchange of queens which occurred in the 13th game after 6 . .. e5 (No.29 1). The situation in the return match obliged Tal to think only in terms of winning. 32 • • • ' �g7 It is already known that if 32 . . .l:f8 there would.have followed 3 3 f4. 7 i.d3 e5 8 lhge2 exd4 Usually after 6 . . . c6 Black transposes into the variation involving . . . a7-a6 and . . . b7-b5. This was how Smyslov played against me in the 8th game of our 1958 return match (No.259), as did Larsen in · Leiden ( 1970) - see No. 3 80. This exchange can hardly be ap proved, since White can advantageously 1 37 Only after his rook has left the comer square does White make · this move, strengthening his c4 pawn, since now the weakening of the al-h8 diagonal is of no significance. recapture with his bishop. 8 ...lLlbd7 was better. 9 .i.xd4 14 15 ii.bl 9 • • • c5 Black takes on too great obligations. Such a plan would have chances of success, if he could compensate for the defects of his position in the centre with piece pressure, but he has no such prospects, in view of the fact that he is behind in development. 9 ... tDbd7 I 0 0-0 lDe5 1 1 f4 tDxd3 1 2 'ifxd3 also cannot be recommended for Black. Subsequently this variation occurred in the game Hiibner-Gligoric ( 1973 ), and the Yugoslav grandmaster demon strated the correct method of play for Black: 9 . . . d5 10 cxd5 cxdS 1 1 e5 ttJfd7 12 f4 tDc6. 10 11 ilf2 0-0 tbc6 White ignores the threat to exchange his bishop ( ...l£lc6-e5xd3), which would merely be playing into his hands. 11 12 1id2 13 l:.ad1 14 bJ �ab8 llfd8 16 f4! At this point, when the two sides have completed their development, White, 'as is his right', is the first to begin active play. There is immediately the prospect of an attack on f6 (f4-f5, ..ih4 etc.). In order to parry it, Black decides on an exchanging operation, as a result of which White gains the advantage of the two bishops, and also a very active position in the centre and on the kingside. It need hardly be said that the dis appearance of the queens signifies the wrecking of Tai' s plans, made in his preparations for the game! 16 17 18 19 20 a6 i.e6 'lfa5 138 hJ it)xel :xd2 ll'igJ .i.g4 i.xel 1ixd2 :eS Naturally� Black would have been quite happf to exchange his d- and c pawns for the . white e- and f-pawns (20 e5 dxe5 2 1 ..txc5). To avoid this, White has to make a rather awkward move. i..f8 The variation 20 . . hS 2 1 �xd6 h4 22 li:Jhl ltJxe4 23 .i.xe4 I:txe4 24 ..i.xc5 (or 24 i.xh4) would have led to the loss of a pawn� therefore Black is forced to spend a tempo defending his d6 pawn. 20 . 21 Itel 22 t[}fl lie6 22 23 :de2 libe8 J.g7 At the first opportunity the knight leaves the g3 square, since at the given moment it is not needed for the defence of the e4 pawn. White's coming offensive have been hindered by 23 . . . h.5 . should 26 lDh2 t[}f8 All Black's efforts are concentrated on occupying the d4 square. Meanwhile, in such situations this manoeuvre does not play any particular role, and it is unable to prevent the development of White' s initiative on the kingside (cf., for example, Game 87). 27 J.h4 2s .:n lDe6 ll d7 A serious error, after which Black's position becomes difficult to defend. One can understand Tal not wanting to restrict his most active piece, the bishop at g7, but 28 . . .f6 was nevertheless necessary. Then White would have had a choice between 29 f5 liJed4 30 llef2 and 29 i..g3 f5, in both cases with counterplay for Black. Apparently my opponent evaluated the position somewhat routinely, and did not sense just how dangerous it was. Later Tai said that he lost the return match, not imaging that the former champion would act so decisively. 29 g5 24 g4 White .not only restricts the space available to the opponent's pieces, but he also prepares the activation of his own king. 24 lDd7 25 ��2'_.�,... lt6e7 After this seemingly · · paradoxical move� shutting in the bishop at h4, 139 White forces the exchange of his opponent's bishop at f6� after which it is doubtful whether Black can save the game. 29 . . . h5 30 gxh6 31 lDg4 i.xb6 .i.g7 32 33 34 35 ..i.xf6 lf)g7 lDh5 l:r.ed8 Preventing ttJg4 for one instant. After 29 . . . b5, which, incidentally, Black was also free to play earlier. the develop ment of events would not have changed. After 3 1 . . . lbxf4+ 3 2 ltxf4 i.xf4 3 3 t'ill'6+ White would have won a piece. lDf6+ -*.xf6 nd2 ll.cJ 40 f6 41 l:td5 42 bxc4 43 <it>f3 b5 bxc4 ltb7 :b4 44 .i.xb4 45 .i.xb5 46 exd5 lDxb4 lDxd5 gxh5 This exchange sacrifice also does not save Black: White immediately takes play into a won rook ending, rejecting a material advantage. 47 .:bt The last move in the game that required accuracy! Black has no useful moves, and he can only passively await the outcome. 36 37 38 39 i.c2 i.dl Jl.g4 f5 � <l;e7 :c1 @es White had only one more move to make before the time control, and, instead of 40 l::tdf2 with the threat of 4 1 fxg6, he was tempted into playing for 'suffocation'. 140 47 48 :b6 49 <it>f4 <i>g8 <t>h7 50 <t>g5 51 <it'xh5 52 h4 llg8+ l:g3 %te3 � How can Black hope to save an endgame, in which he has had to spend so many tempi on decentralising his king? 52 . . . :c3 was no bettei; on account of 53 Itxd6 l:hc4 54 :ct? 'iti>g8 55 l:td8+ @h7 56 etc. lle5+ �f8 53 :xd6 54 55 56 57 58 �g4 ct>f'4 �el %ba6 <t>d3 �g6 i:tf5+ l:h5 llxh4 �5 7 Checks to the white kiiig do not help, since he hides from them at a4. 11h2 59 :Ic6 60 llxc5 %lxa2 61 l:tc7 �xf6 ci>e5 62 lld7 63 lle7+ Black resigns. Naturally, he could not be satisfied with either 63 . . . �6 64 d6, or 63 . . . 'itd6 64 l:.xf7. Game 293 World Championship Return Match Moscow 1961, 18th game Caro-Kann Defence e4 d4 e5 h4 g4 cl • • e6 8 lLle2 8 9 lLlaJ 8 ltJa3 would perhaps have been safer, to prevent the black bishop from reaching the fl -a6 diagonal. • .. .i.b5 But now this is simply unfavourable, since it allows Black to force useful exchanges. 9 .i.e3 should have been played. 9 10 11fxe2 11 cxd4 12 bxa3 Tal-Botvinnik 1 2 l 4 5 6 • So, in the centre we have a pawn formation, typical of a variation which was employed in the 1 9th century by Steinitz, and in the 20th century by Nimzowitsch. The analogy can be con tinued with the fact that now Black does not experience any opening difficulties. .i.xe2 cxd4 .i.xa3 c6 d5 .i.f5 h6 il.d7 In the 10th game Tai played 6 h5, preventing . . . h6-h5 . But now he had evidently decided that after 6 . . . h5 he could well accept the pawn sacrifice. For the same reason Black prefers to avoid this continuation. 6 7 .i.g2 • • • c5 White aims to provoke . . . e7-e6 as soon as possible, to rid him of his con cerns about the g4 pawn. However, at g2 the bishop has' no particular future. 14 1 A semi-open position has been reached, in which the black knights will be no weaker than White's bishops, which are condemned to defending his numerous weaknesses (a3 , c4, d4, g4). 12 • • • lLlc6 13 .i.e3 14 'iWl lia5+ 14 lDge7 For the moment White avoids the exchange of queens, hoping to develop an attack. Over the course of several moves Black does not take the a3 pawn (and White does not defend it)� both players rightly consider tliat it is more important for them to complete their development. 15 l:t.bl 16 .i.bJ l:tb8 16 17 Itdl 11a4 A loss of time. This square would be better used for the rapid inclusion in the play of the rook at h 1 . knight to c4, or to exchange · queens and play a cheerless ending. In the latter case the doubling of the enemy pawns cannot provide any consolation, since in the end the exchange of rooks on b6 will be inevitable, and Black's pawns will resume their normal form. 19 20 21 22 'if xa6 h5 l:tbl <t>gJ bxa6 <it>d7 ltb6 lDa5 23 :xb6 24 f4 axb6 li)c4 Let the move 22 . . .l�hb8 remain in reserve. Of course. the d4 pawn had to be defended, but not in such a passive way. Therefore 1 7 d2 tiJa5 should have been played, although this continuation did not look very tempting for White. 'if 17 18 �g2 'ilxaJ Wa6! 25 ..i.cl White tries to retain the two bishops, which are his only consolation in this position. 25 26 27 28 29 This sets White a difficult choice: either to allow Black to transfer his lldl aJ f5 ltxct tDc6 li)b4 lDa2 tDxcl So, White has also been deprived of his last trump - the tWo bishops. .. . b 29 5 <J;;e7 30 .:tat 142 37 l:cl Just in case, Black moves his king off the diagonal Qn which the enemy bishop is operating. 31 <iti1'4 32 g5 33 <&ti>xg5 34 .*-xf5 37 a4 was stronger. Even so, after 37 . . . b4 3 8 i.:f3 liJd2 ! (but not 3 8 . . . tiJb6 39 a5, or 3 8 . . J:k7 39 :c l ! ) 39 · i.xd5 %k3 40 .i.e4 (or 40 ktd l ltd3) 40 . . .b3 ! (weaker is 40 . . J�c4 4 1 �e3 t2Jxe4 42 <iirxe4 b3 43 ltb 1 l:tb4 44 <itid5, when White has drawing chances) White would have been unable to combat the passed. b-pawn. :.cs hxg5+ exf5 :c6 37 . 38 .i.f5 • . f6 In time trouble White makes an error, leading to the loss of a second pawn, but, generally speaking, in this position the conversion of Black ' s material advantage is not very difficult. 38 • .. fxe5+ Weaker is 38 . . . t2Jxe5 39 llgl (39 l:te l flc4). What is there for White to do in this position? The attempt to hold on with 35 l:c l is refuted by 3 5 . . . ttJxe5 36 :e l f6+ 37 <iirf4 :c4, w°hen he loses another pawn. Therefore, having discovered the only weakness in Black' s position - the g7 pawn. Tal clears the file in order to attack this pawn. M4 The start of some mutual time trouble errors, which, however, do not affect the evaluation of the position. Here, for example, the quickest way to win was with 3 5 . . . lDb6 followed by the invasion of the rook on the c-file. In trying to force the opponent to block the g-:file and simultaneously gain time, Black allows some complications. 35 36 .i.g4 39 dxe5 40 l::td l Or 40 :e l �6 (4 1 lhe5 J:tc4+). 40 41 lth6 llc6 Otherwise 3 7 i.f3 ltJb6 3 8 ltc 1 , and White seizes the initiative. lLlxe5 �d6 D.c5 White resigns .i.e4 Game 294 Botvinnik-Tal World Championship Return Match Moscow 1 961, 21st game King' s Indian Defence 1 2 3 d4 c4 lLlcJ ltlf6 g6 ilg7 Before this game Tai was still the World Champion, since I had only 1 2 points and my opponeµt still had theor etical chances of drawing the match by 143 winning the remrurung four games. Therefore he chose a complicated open ing - a draw was of no interest to him at all. 4 e4 5 f3 d6 l£lbd7 In avoiding simplificatio� Tai chooses a not very successful variation. It should also be mentioned that we had spent three days on the previous game, in which for a long time he had been hoping to win, and after I had saved it he was, naturally, rather depressed. 6 .i.eJ 7 l£lge2 8 e5 0-0 9 10 . • • f5 a6 . 0-0-0 Played to control the b5 square for counterplay with . . . b7-b5, and also to free the queen from having to defend the c7 pawn in· view of a possible lbb5. Even so, 10 ... l£lc5 is probably stronger (Boleslavsky-Keres, 1 952). 11 'ifi>bl l£ldf6 Black intensifies the pressure on the central e4 pawn, and in view of the threatened . . . b7-b5 White is practically forced to exchange on f5 . 12 exf5 13 l£lg3 gxf5 13 'l'e8 · Now B iack does not have a great choice. If 1 3 . . . f4 14 liJxh5 fxe3 1 5 ltJxf6+ 'i'xf6, then White emphasises his positional advantage by 16 'i'c2, avoiding the double-edged play after 16 'ii'xe3 e4 ! 1 7 fxe4. The manoeuvre 16 'i'c2 (instead of 16 'i'xe3) was one that I later carried out in a game with Medina (No. 324). The exchange of knights on g3 leads to the opening of the h-:file, and that says everything. • • • d5 White accepts the challenge and goes in for a complicated game. However, there was no particular risk in this, since Tai did not like closed positions. The Encyclopaedia considers 8 'i'd2 to be stronger. 8 9 .. lid2 . l£lh5 White avoids the win of a pawn (9 g4 tt)f4 1 Q �xf4 . exf4 1 1 ..txf4), which after 1 1 . . . f5 12 exf5 gxf5 13 g5 would have handed the initiative - to Black. 14 1 44 i.clJ! �xg3 21 :dbl 22 fhgs+ 23 :h6 In the spirit of the position was the pawn sacrifice l4 . . . e4 ( 1 5 · ctJxh5 fol lowed by fxe4)� although it could hardly give sufficient compensation, as Biack' s pieces are not mobilised. )4 . . . f4 would still have not given anything, while after 1 4 . . . Wfg6 1 5 liJxh5 'i'xh5 1 6 h3 the unpleasant g2-g4 cannot be avoided. Even so, it is hard to agree with Black's decision. In the game White's attack becomes virtually irresistible. 15 hxg3 This is the whole point. White controls the g6 square, which is of decisive importance. · 23 c5 ilg6 b5 18 19 klh4 'it>xg7 bxc4 • . . fxg4 Even so, Black should not have given up the b l -h7 diagonal without a fight. After, for example, 23 . . . e4 24 gxf5 .i.xf5 25 fxe4 .th? he could still have held the position. However, even in this case White had the decisive contin uation 24 fxe4 tt:lxg4 25 l:txd6. A standard advance in the King's Indian Defence, after which the . . . b7-b5 counterblow could have gained in strength. However, White ' s offensive develops more quickly. 16 i..h6 17 g4 18 ..txg7 1lg5 hxg5 Of course, it was tempting to ex change the bishop defending the enemy king, but the h6 square should not have been given up without a fight. The pre paratory 1 8 %th4 was more consistent. . . • 24 fxg4! Now White sacrifices a pawn (by retreating his bishop), but on the other hand his king will be completely safe. However, the alternatives 19 . . . C'Dg8 20 gxf5 .ixf5 21 Itg4 and 19 . . . e4 20 gxf5 .ixf5 2 1 fxe4 were equally hopeless. 20 i.c2 the simplest. The pressure on the ffile will be irresistible. 24 25 lig6+ 26 :n 27 llg7+ h6! Cleverly played. Black covers the h6 square and forces the exchange of queens, which, alas, can no longer halt White' s attack. It becomes clear why White should not have been in a hurry to exchange the dark-square bishops (cf. the note to White' s l �th move). .i.xg4 <!/e7 <i>e8 <l;f7 Other replies were equally hopeless: 27 . . ':f7 28 l:xf7+ r:l;xi7 29 tt:le4, or 27 . . . �d8 28 lhe4 (but not 28 l:xf6? llxf6 29 l:g8+ and 30 ltxa8 on account of 30 . . . l:.tfl+) 28 ... lhd7 29 J:ixf8+ thxf8 30 iDxd6. 145 . 28 lDe4 li:ld7 The white knight is immwie because of 29 .i.a4+. 8 e4 c6 29 lihd6+ �d8 30 lbf8+ iCixf8 31 . ·lCixc4 .i.d7 32 :n 'l;c7 33 d6+ Black resigned Here the game, and with it the title of World Champion. So, by transposition we have reached a well-lrnown variation, which is gen erally considered satisfactory for Black. It may be remembered that it occurred in two games from world championship matches (Botvinnik-Smyslov, 14th game, 1954, and .Botvinnik-Tal, 6th game, 1 960), and both ended in a win for Black. Regarding 8 . . J:te8, see Games 1 13 and 2 10. 9 Game 295 Botvinnik-Pachman European Team Championship Oberhausen 1961 King's Indian Defence 1 2 3 4 5 6 i£if3 g3 .i.g2 c4 thc3 d4 i£if6 d6 e5 g6 ll.g7 Only this move transforms the game from an English Opening into a King' s Indian Defence. 6 7 0-0 iCibd7 0-0 h3 In Game 1 50 I played 9 b3 . 9 10 :et 1fb6 10 11 ihxd4 exd4 thg4 . • • White avoids d4-d5 and maintains the tension in the centre - a method that was employed in this position by Semion Furman. • • . 10 . . . l:Ie8 was also possible. After l 1 . ..l:te8 12 :e2 the lunge 12 . . . �g4 can be countered well by 1 3 l'.:td2. 12 like2 , . After 1 2 hxg4 ifxd4 Black has an excellent game. 146 12 . . . 13· b3 psychological ones, at that time it was the strongest! In a If 13 c5 White has the convenient later ... game Portisch-Gligoric retreat 14 tt:Jc2, defenrung the rook at ( 1 964) the continuation 16 £Cic3 'i'd8 17 al. l:te2 We7 a5 lte8 14 i.eJ 15 :bt 1 8 f4 £Cied7 19 · .:tf2 also proved advantageous to White. 16 • • ll'c7 • Subsequently Black in fact fails to find a good plan. The only question is whether White will find one . . . 17 18 19 20 Steinitz 'ifc2 lDcJ l:tbdl ltlde2 long ago lDed7 lDf8 lDfe6 recommended refraining from exchanges, when the opponent has a cramped position. 20 21 1fd2 • . • .i.f8 The plan involving f2-f4 appears to At that time this position was little known among the broad mass of regular be the most logical. Reckoning that in the given position there is no reason to players, although it had already occur hurry , for the moment White adopts red in master games. Black is counting waiting tactics. on exploiting the strength of his piece pressure on the centre, while White is hoping in time to restrict the activity of the black pieces. 16 llft! This paradoxical move was em ployed by Furman not long before the present game, and it probably came as a surprise to my opponent. White removes his rook to a passive position. in order to deprive Black of tactical chances involving an attack by one of his knights on the rook at e l . For example: 16 'i'c2 lDed3 17 .l:ted l £Cib4 etc. It cannot be asserted, of course, that 16 %tfl are is the only move, but if all factors taken irito account, including 147 21 22 i.b6 23 .ttfel �d7 �g7 Of course, this is a weake1ung of the position, but Black is preparing . . . lDe5f7. However. White himself retreats his bishop �thout any compulsion. il.eJ 24 25 .i.f4 26 lhd4 27 'fic2 lhc5 l%d8 .*.d 7 27 28 .i.eJ 29 a3 lDge6 :eS 30 31 exf5 32 g4 The white pawns are very strong. Black's scattered pieces and exposed king make his position critical. 1'e5 It is clear that sooner or later Black will have to return his extra knight, and he intends to do this in an advantageous way or to force simplification. How ever, after the next move, which came as a surprise to him, it transpires that he should have returned the piece immediately with 32 . . ltJxf5, since all the same the knight at g7 is out of play. 32 Finally a concrete plan is contem plated - White intends to drive away the knight from c5 by a2-a3 and b3-b4. gxf5 ltlg7 . • • . 33 .i.d4! Decisive. It soon becomes clear that the exchange of one pair of rooks does not ease Black's position, and a counter sacrifice will no longer be possible. lbel+ .i.c8 33 34 llxel 35 :dt .f!e8 36 b4 37 axb4 axb4 lDa6 Black must strengthen his control of f7, in view of the tlrreat of f5-f6-f7+. f5 29 Black tries to free himself, but he overlooks a tactical blow. However, after 29 . . . l2Jxd4 30 �xd4 it would also have been bad to play 3 O . . . f5 because of 3 1 b4 axb4 32 axb4 ltlxe4 3 3 ii.xe4 ! fxe4 34 ltlxe4. 30 lhxf5? But now in the event of 30 b4 axb4 3 1 axb4 lLlxe4 or 30 exf5 ltlxd4 and 3 1 .. . .i.xf5 Black would have success fully defended. At the same time, the positional sacrifice of a piece made by White does not leave Black any choice. · 148 38 ltle4! l:lxe4 39 .i.xe4 ltJxb4 to bring it out to f5 or even g4, whereas 7 . . . i..e7 is in any case necessary. Othernrise the terrible threat of 39 0\f6+ cannot be parried. 40 'lfd2 41 cxd5 Or 40 . . . tiJa6 4 1 'i'g5. 42 .i.f3 8 9 d5 cxd5 Here Black was due to make a sealed move. However, in view of his numer ous weaknesses and the bad placing of his pieces. together with the strong white pawns on the kingside. further resistance is pointless. Therefore resigned . Black ti:lcJ .ieJ il.e7 · In the first half of this event, playing White, Unzicker defeated me in excellent style in a French Defence. In order to take first place on the team leaders' board, I was obliged to gain my revenge. Hence the decision to avoid well-known continuations and from the very start to prevent the opponent from using his opening knowledge. 9 dxc5 ii.xc5 I 0 i.g5 or I 0 l£ia4 is usually played. Game 296 Botvinnik-Unzicker European Team Championship Oberhausen 1 961 Queen's Gambit 1 2 3 4 5 6 c4 ll)fJ g3 .ig2 0-0 cxd5 d4 li)f6 e6 d5 c5 li)c6 exd5 9 7 An opening transformation typical of modem tournament play. After the first move the game was an English Opening, then a Reti� then it seemed to I ransfonn into a Catalan� but in the end a Tarrasch Defence was reached, and that means a Queen's Gambit. 7 • • • i..e6 Black should not have been in a hurry to develop this bishop. since in some cases it might be more advisable ... c4 My calculation proves correct. Now a complicated battle ensues. and White advantageously establishes his knight in the centre of the board. 10 /l)e5 0-0 After 10 . . . 'ifc7 1 1 .i.f4 ilb6 White could have begun an attack 12 e4. 11 li)xc4 - A few decades earlier I had observed one of Levenfish 's games, in which he carried out such a combination. The position is opened up, and the white 149 bishops prove more active than Black 's. The alternative was 1 1 b3 . 11 12 d5 13 lllxd5 Black Wlderstandably aims for sim plification, but this move !eads to the loss of a pawn. After 16 . . . .i.e6 ! i..c 3 Wb6 he would have had a reasonable chance of occupying the open file and e>..i>loiting the weak d3 square. · dxc4 l£ixd5 .i.f6 Since 1 4 l£ixf6+ 'i'xf6 followed by . . . l:tfd8 and . . . lDd4 is quite safe for Black, White cannot prevent 14 . . . .ild4. Ten years later another plan for Black, 1 3 . . J:k8 14 :c 1 b5, was tried by Spassky in a game with Taimanov, but with 1 4 'ifd2 followed by Itfd l White could have gained a clear advantage. 1 4 ttlf4, a s suggested by Polugayevsky, i s also good. 17 .i.xe4 ttlxd4 17 . . . 'iixd4.. then White has If instead a pleasant choice: 18 'i'xd4 �xd4 1 9 �xc4, or 18 .i.xc6 'i'xdl 1 9 llixd l bxc6 20 l::t.xc4. in both cases with an imposing advantage. 14 :ct i.d4 15 .txd4 ..lxd5 lDe6 18 .:.xc4 Black retreats his knight and assumes that after 1 9 .ixb7 l:tb8 he will be able to regain the b2 pawn. However, my opponent overlooked a tactical subtlety. 14 'I'd2 would also have been answered by 1 4 . . . .i.d4, and if 1 5 lDf4 'i'b6 16 .t.xd4 lD.xd4 ! 14 . . . 17 'l!!' ;.�;:.y.: A ;:r1;· J;: : (:: .. ;;�,w.� � I ;;;,' /?,'� ��* Of course. not 1 4 . . . .txb2 1 5 :xc4, and White' s initiative increases. �}rtJ iJ��:-W%-. JJ'f/, � �:}PJ' ffx�%ffl!"' : :),� �JrtJ :M.f(4, ,-, .-/.-.. .. f Wr ·�-'.;/N/ e :;(.::'/,.' t 19 W'c2! Defending the b2 pawn with gain of tempo, after which one of the black pawns - b7 or h7 - is lost. 16 e4! 17 19 20 .i.xb7 21 il.g2 Exchanges - 16 i..c5 .llxg2 'it>xg2 :es l1xc4 1fxd 1 1 9 l:.xd 1 Ihe2 could in no way have satisfied White. 18 16 . . . i.xe4 g6 llb8 When the opponent' has a queen and knight, a fianchettoed bishop covers best of all the approaches to the king's position. 1 50 21 . 22·. · . bl 23 11e4 24 l:tel 1!ff6 tt:\d4 i!fd8 tt:\f5 25 l:tc6 26 l:tc2 'ifb2 29 .i.d5 The enemy rook must not be allowed to go to e6. 29 30 .ic4 31 '9e4 In this situation it is unfavourable for Black to win the queen for two rooks (24 . l:le8?). . . 1fb5+ 'ifd7 A prophylactic move. Since the bishop is no longer defending its king, instead the long diagonal is covered by the queen. The place for this rook is at e2, about which, however, I soon forgot! l:ld6 31 32 J:ce2 lhd4 �f5 33 l:e3 34 ll3e2 �d4 35 :b2 Not agreeing to a draw . . . 'lfhJ+ 35 36 1ig2 Va3 26 . . . 27 'i'e5 Intending to continue 28 ct.d5. 1i"b4 27 A natural move but 36 'it?gl was preferable. In the game Black could have developed an initiative by 36 . . . ii'h5, forcing, in view of the threat of 37 tDfl , the reply 37 .i.e2. , ... 28 'iff5 <irfl The obsessive idea of transferring his bishop to c4 leads White into difficult ies. This was the reason for his last move, since the immediate 28 i.d5 would have been answered by 28 . l:txd5; He should have connected his rooks with 28 %lce2, and after 28 lDd4 29 :e4 'i'd2 30 'i'f6 ! he would have gained a virtually irres istible attack, to say nothing of his material advantage . 36 . . . . 28 . • • . • ltb6 37 f4! In this way White eliminates the weakness of his f3 square and co ordinates the actions of his pieces. 151 'itb.5 37 . 38 nf2 39 g4 After arranging his kingside pawns in the most advantageous way, White takes play into the endgame it'a5 46 47 <it>xel 48 gxf5 Restricting the mobility of the black knight, which at a propitious moment might have gone via f5 to e3 . 39 40 f5 • • ttle6 • ife4 ifeJ lhel+ f5 48 g5 would have led to more com plicated play, but, having little time for thought, I chose a simpler continuation. TI1is leads to an abrupt sharpening of the play, which is fully justified, since the white pieces are now well placed. 40 41 42 . . 48 gxf5 lldl :sd4 The sealed move. I also considered the combination 42 fxe6 1!xe4 43 exf7+ @f8 44 l:xd l :xg4, but 45 :e l is parried by 45 . . . :gl +, and I was unable to find a forced variation, leading to a win. 42 'i'xel+ • • • A clever move, which I overlooked in my analysis during the brief dinner break. However, it did not prove so difficult to find a way to win. I was expecting 42 . . . l:txel + 43 Wxel 'i'xe l+ 44 �xel gxf5 45 gxf5 ltJg5, when I was intending to win as follows: 46 f6 h6 4 'iti>e2 l:.e4+ 48 �d2 and then .i.d3. 7 43 'ii' xel ltJf4 Nothing immediately can be done 44 :xf4? ltxf4+, or 44 ifxdl ? Ihdl mate, but Black has no active moves, and White is after all a pawn up! 44 fxg6 An essential exchange. Black was threatening by 44 . . gxf5 45 gAf5 to give White a weak pawn at f5 . . 44 45 h4 46 Ac2 hxg6 �g7 White has a minimal material advan tage, the exchange of the f- and h-pawns is inevitable, and with play on one wing, as the experts teach, a knight is stronger than a bishop. Also, it should not be difficult to exchange the knight for the, two remaining white pawns, and this will lead to an ending with rook and bishop against rook, which by no means always reduces to Philidor s winning position. However, let us see how events develop. · 49 ii.fl! Vacating the c-file for the rook and relieving it of the need to control the g2 square, to where the knight could have gone. 49 152 @h6 50 llc5! 50 51 52 :a5 53 <bf2 • • • <ifi>h5 'it>xh4 �d7 Alas, Black has nothing else. lhf5+ thg4 57 l:la6 57 58 .id5 59 .i.e4 'it>h5 the5 With the deadly threat of 58 :g6. Black has achieved his aim, but only apparently. The tragedy is that his king is not only cut off from the queenside, but may also be in danger. Threatening 60 :as. Black has several possible moves, but they all merely · worsen his position. For example: 59 . . . @h4 60 11116 mate, or 59 . . . tt\f7 60 i.f3 + 'iti>h4 6 1 :as, or 59 . . . tt\g4 60 i.f3 .:tf7+ 6 1 <it>g3. As for the rook, it has to guard the knight and the pawn. It follows that mate or loss of material is unavoidable, and so '!J.e7 Trying, in tum, to cut off the white king from the queenside, for where, however, it is not aiming. <it>f3 After 54 !tf7 55 i.c4 the black rook would have been unable to defend simultaneously the knight and the pawn. tlJg6 54 56 @£4 The rook ending after 56 . . .�xc4 57 bxc4 would have been hopeless for Black, while after 56 . . .ttJg6+ 57 <i>fs his knight would have been shut out of the game. Preventing . . . a7-a5-a4 and tempting Black to exchange · the kingside_ pawns as quickly as possible. . . . Black resigned. 55 il.c4 Grune 297 Botvinnik-Wade Hastings 1961162 Ruy Lopez e4 thf3 3 .i.b5 4 i.a4 5 .ixc6 1 2 Here it can now be guessed that White is intending to encircle the enemy king. For this it is important to take control of the f7 square, after which the white king can no longer be driven off the f-file. 55 . .. e5 thc6 a6 thf6 A continuation that was employed many times by Flohr, and then also by Kholmov. In recent years it has gained a nwnber of other supporters. tbe5+ 5 dxc6 6 d3 6 0--0 or 6 lDc3 is equally good. 6 . thd7 • • • . 153 • The most sensible way of defending the e5 pawn. . 7 lhbd2 f6 In the Exchange Variation of the Ruy Lopez, this is usually played before . . . liJf6. Here, however, for the moment the e5 pawn is not threatened and . . . f7-f6 weakens Black' s position. 7 . .. J.e7 or 7 . . . g6 was preferable. 8 0--0 .*.c5 It will not be possible to maintain the bishop on this square, and hence White gains a further tempo. 9 c3 10 d4 11 tLlbJ exd4 �b6 Black probably thought that White would be forced to take on d4 with his paw� but his main concern is to hinder the exchange of the opponent's doubled pawn. kingside, which will aid the develop ment of his initiative . l£lg6 Of course, not 12 . c5 (in the hope of exchanging queens) because of 1 3 lDxc5 .i.xc5 14 'i'h5+. 12 . • • .. 13 .i.eJ 0-0 13 . . . c5 could now have been played, but after 14 li:lf5 the c5 and g7 pawns would have both been attacked. 14 11fc2 15 tLlf5 W'e7 The exchange on f5 is unavoidable� and White seemingly devalues some what his pawn majority. But on the other hand, for his other knight he acquires an excellent strongpoint at e6. 15 16 exf5 • • • .i.xf5 lDe5 Perhaps Black should have gone in for 16 . . . i.xe3 1 7 fxg6 i.a7 1 8 gxh7+ �h8, since in time the h7 pawn would be lost, he would retain his active bishop, and the white knight would no longer be able to reach e6. 17 .i.xb6 18 l£td4 19 lDe6 12 tLlfxd4! After 12 cxd4 .i.g4 Black would have achieved a comfortable development. Now. however, for the moment his light-square bishop has no future, whereas White has a clear pawn majority in the centre and on the 1 54 cxb6 llti 19 20 :ret llfc8 b5 20 . . tbd7 followed by . into consideration. 21 22 23 24 lleJ l:tael f4 %lh3 . The position is such that, despite the . . �f8 came Wear material equality, White' s forces a to be more numerous. Now he is threat ening to take the c5 pawn, arid for the c.ti>b8 .S:g8 tDg4 moment Black indirectly defends it. 31 32 11'g2 11'c6 1Fd7 After the exchange of queens Black It is obvious that the black knight would have similarly been llllable to will have to retreat to h6, to weaken the avoid soon losing material. He evidently opponent' s pressure on the rook' s file. thought that after the exchange of rooks But this signifies that it will be shut out he would gain some hopes of perpetual of the game. 24 25 26 27 28 1fe2 g4 l:td3 �fl check. liae8 lDh6 1id7 'ifc8 33 34 35 36 tDxc5 1if7 llxe8 :.xe8 lhe8+ 'ifxe8 'ife4 'i'd8 Now it is obvious that Black must . . . g7-g6 avoid the exchange of queens - he is threats to his king may arise along the already one pawn down, and the loss of g-file. It is curious that at fl the king a second is inevitable. White reckons that after 37 1id4 Ve7 38 tDe4 tDg8 39 g5 Black resigns feels more safe than on the customary hl square, since Black has in reserve . . . c6-c5 followed by . . . 'i'c6. 28 29 h3 30 fxg6 31 l:leJ c5 g6 hxg6 Grune 298 J.Littlewood-Botvinnik Hastings 1 961162 Sicilian Defence 1 3 4 5 2 e4 ttlt3 d4 ttlxd4 tllcJ c5 d6 cxd4 tllf6 g6 In the Dragon Variation Black carries out a plan that was introduced by Reshevsky, in which the position of the queen's knight is determined later. 155 6 .i.eJ 7 f3 8 .lc4 i.g7 a6 b5 begin play on the file that is then opened on the queenside. Therefore Black castles on the kingside. i.b7 �bd7 In order to create counterplay, Black must push back both white knights, but it must be done only in this order, since after 1 5 . . . eS 16 llJc2 and the inevitable 1 7 ltJe3 White has a good game. 12 13 cxb3 14 i.h6 15 1lxb6 Sticking to the afore-mentioned plan. Black (when White has already dev eloped his bishop at c4) develops his bishop at b7, and then plays his knight from b8 via d7 to c5. .in order to exchange White's king's bishop on b3 . 9 .lb3 10 il'd2 lDxbJ 0-0 i;xb6 b4! 16 e5 In keeping with his style of play� Littlewood goes in for great complica tions and . . . loses. More cautious was 16 tlJce2 e5 1 7 ttJc2 a5 with a double edged game. 16 . . . lhd7 If 16 . . . dxe5 1 7 tDf5, and White wins the queen. 17 h4 Now the immediate 1 1 .th6 is the most unpleasant for Black� and after 1 1 . . . �xh6 1 2 1i'xh6 t2Jc5 1 3 0-0-0 llJxb3+ 14 cxb3 'i'b6 1 5 'iti>bl 0-0-0 1 6 b 4 I encountered certain difficulties i n a game with Krutikhin (No. 3 15). The attempt by Reshevsky in a game with Bisguier ( 1 957) to avoid the exchange of the dark-square bishops by 10 . . . h5 did not prove successful. White simply replied by castling kingside. 11 0-0-0 12 <ii>b t �cs Preparing after the exchange on b3 to recapture with the c-pawn (Boles lavsky' s well-known manoeuvre) and to After 1 7 exd6 e5 or 1 7 . . . bxc3 White would have lost substantial material without any compensation. But now, since the black knight is pushed back, the opening of the h-file is unavoidable, and it would seem that White is bound to conclude the game with mate. How ever, Black finds a way of returning his knight to f6 and of parrying the threats on the kingside. 17 . . . 18 h5 bxc3 dxe5 18 c2 + 19 <&tixc2 dxe5 would have led to a transposition of moves, but White could have changed the course of events: 1 9 ttJxc2 g5 20 'i'xg5+ <ifi>h8 2 1 exd6, and for the piece · he has three pawns plus a positional advantage. . . . 19 hxg6 1 56 26 llcl 1ixa2+ The simplest - an extra bishop is sufficient. 27 1ixa2 lhxa2 28 %t.xd8 :xd8 White resigns Game 299 Robatsch-Botvinnik Hastings 1961162 French Defence Other continuations too would not have maintained the attack, e.g. 19 tiJc2 (with the threats of 20 hxg6 and 20 lhd7) 1 9. . . g5 20 'i'xg5+ �h8 2 1 h6 ilg8 22 ii'f5 i'..c 8, or 19 bxc3 exd4 20 :txd4 'i'a5 21 :Ixd7 'i'f5+ 22 'it>al .i.c6. 19 • .. lDf6 A move which was hard to foresee when White began his cavalier attack with 16 e5. Boleslavky's idea is a good one, of course, but in the given situation the absence of the pawn from c2 allows Black to achieve the impossible (if 20 QJf5 or 20 tt:Je6 there follows 20 . . . c2+! 2 1 �xc2 'i'c8+). 20 bxc3 21 gxh7+ exd4 21 22 �xd4 23 9'e3 �f8. Cit>h8 1i'a5 lDd5 The attack would also have petered out after 2 1 lhd4 'i'a5 22 :tf4 fxg6 23 klxf6 htxf6 24 ii'xh7+ 1 2 3 4 5 6 e6 d5 i.b4 c5 .i.xcJ+ 1fc7 Since lengthy tournament experience has shown that after 6. . . t:De7 7 1"g4 it is not easy for Black to equalise, I began playing 6 . . . ii c7, in order to answer 7 ifg4 with 7 . . . f5. True, even in this case, as shown, for example, by Game 1 8 1 , after 8 'i'g3 Black still ha s some difficulties. But when on his ne:x1 move White plays 7 ttJf3 , this leads to the usual variations, in which Black feels comparatively secure. lhxc3+ 1lad8 7 lhfJ .*.d7 8 a4 9 .i.aJ l£ic6 Regarding 7 . .. l:i'Je7, see Game 235. Black prevents 8 i.d3 , on which there could have followed 8 . . . c4 9 i..e2 .ta4, and after castling long, his king will find a secure shelter on the queenside. Therefore White immediately advances his rook's pawn. Black has not only two extra pieces. but also the initiative. 24 1fd2 25 �al e4 d4 lhc3 e5 a3 bxc3 The Encyclopaedia considers that 9 .i.d.3 is stronger. 1 57 13 14 ifh4 f5 ci;f7 ; The logic of this move is that Black s king' s rook is more useful at h8 than at f8. The threat of . . . g7-g5 becomes highly unpleasant. 15 g4 White hopes to find salvation in complications, and not without reason. 15 9 • • • tLlc4 Driving back the dark-square bishop, since the light-square bishop at d3 is more valuable. 16 .tel 17 %!.et b6 Here 9 . . .cxd4 10 cxd4 11f'a5+ can hardly be recommended, since after 1 1 °Wd2 'ii'xa4 12 ii'g5 Black ends up in a difficult position, and after 1 l . . .ifxd2+ 1 2 <t>xd2 White's chances in the end game are clearly preferable. l:tag8 g5 10 11fd2 In a similar situation in a 1943 game, Tolush played against me 10 i.e2, which, of course, is more advisable. 10 11 i.d3 tLlge7 h6 12 0-0 13 11rf4 tLlaS If Black wants to castle kingside� this move is obligatory, since the g5 square must be controlled. This indirectly defends the a4 pawn ( 1 3 . . . .txa4? 14 dxc5), but allows an un pleasant reply, after which the situation changes in favour of Black - he acquires good chances on the kingside. White should have simply played 1 3 i.b2. However, Robatsch evidently assumed that White's plan should involve an attack on the black king. Alas, now Black seizes the initiative. 18 1ih3 But this is a mistake. White should definitely have opened up the position by 1 8 'i'h5+ <it>f8 1 9 gxf5 . Then after both 1 9 . . . �xf5 20 i.xf5 exf5 2 1 e6 .i.e8 22 'ii'h 3, and 19 . . . exf5 20 e6 .te8 2 1 'i'h3 (2 1 . . . g4 22 .ixh6+) great com plications arise. Now Black is able to shut the kingside and then to occupy himself with the weak a4 paWll. 15 8 18 19 1ig2 f4 tLla5 29 ttllr4 To carry out his plan, Black is forced to move his knight from its strong post at c4. 20 b4 21 .i.e2 22 h5 c4 1id8 30 /i)xe6 White voluntarily eliminates all possibilities of activity on the kingside, but this essentially does not change any thing, since the opening of the h-file would merely have been playing into Black ' s hands. 9e8 22 23 <it>hl 24 ..tdt 25 li)gt lt:Je7 By playing 27 . . . liJg8, in anticipation of the knight sacrifice Black defended his h6 pawn (to free his rook at h8 from this task). Now the knight returns, to defend the g6 square. Although in this way White regains a pawn, the simplification is hardly in his favour: Black retains a solid material advantage. 30 31 llxa4 'i'xe6 1if7 32 33 34 35 1if4 ltxf4 :hf8 /i)ec6 • ..txa4 1i'd7 <it>e8 • • Black nips in the bud any possible advance of White 's kingside pawns, and then he forces the exchange of queens. It has now become clear that White wants to make use of his last chance, a piece sacrifice on f4, and so Black removes his king in advance from the danger zone. f3 'i'xf4 'it>g2 <it>g3 26 li)bJ 27 'i'h2 28 .i.xf4 Possibly 28 t;Jxf4 gxf4 29 Xi.xf4 was stronger, when it would have been easier for White to advance his f-pawn. 28 0 0 0 \V gxf4 Black begins what appears 'to be a logical manoeuvre. He defends his a7 pawn, in order to restore the mobility to his knight at a5. However, soon it trans pires that this plan has to be abandoned. Even so, the situation on _the board is such that the four wasted tempi do not change anything. 1 59 36 .i.e2 37 l:lbl i£1b7 A clever counter-chance. Now the 'natural' 37 . . . <t/e? 38 l:.b5 �e6 would be met by the spectacular 39 ltxd5 ! (and if 3 9 . . . �xd5 40 1'.xc4 with mate in the centre of the board). Therefore Black's knights have to retrace their steps. 37 38 l:tb5 39 ltaxa5 lllba5 llle7 44 l:tb6 was not possible in view of 44 . . . l:.4f7! 45 llxc6 l:.a7 46 l:b6 a3 . 47 l:ibl a2 48 :al <l;e7 followed�. - by . . . :b8-bl . 4;c7 44 l£la7 45 :at White resigns. After 46 l:.xa4 ttlb5 • • • the c3 pawn cannot be defended, and his entire position collapses. White has no other possibility. He cannot wait while Black consolidates his forces and exploits his material advantage. 39 40 l1b8+ 41 l1b7+ bxa5 �d7 'it>d8 42 l:lxa7 43 l:lb7 l£lc6 Game 300 Botvinnik-Bisguier Hastings 1 961162 Reti Opening 4 1 .. .<it>e8 could have been met by 42 :bs+ c:l;f7 43 l::tb7 (but not 43 !1xf8+ <it>xf8 44 <ifi>xf4 because of 44 . . . a4 ). Of course, the rook cannot leave the seventh rank, but now Black activates his a-pawn. 43 44 :bt • • • a4! 1 gJ llltJ .i.g2 4 0-0 5 c4 6 bJ d5 ltlf6 e6 .i..e7 0-0 c6 7 .*.bl b5 2 3 6 . . . c5 is more active. 6 . . . d4, see Game 230 . Regarding Such a manoeuvre makes sense when there is a white pawn at d4, but in the given situation it merely restricts 160 Black's forces. 7 . . . l'Llbd7 was more natural, and after, for example, 8 'ii'c2 b6 9 tlJc3 i.b7 10 e4 dxe4 1 1 ti:Jxe4 c5 the chances are equal gain a material advantage. But here it is already hard to offer him any good advice. (Botvinnik- 0 'Kelly, 1 962). 8 dJ 1i'c7 i.c6 16 . . . a5 was preferable. Then it would A logical reply. White defends his c4 8 9 ltlbd2 1 0 1i'c2 11 e4 have been more difficult for White to carry out the plan involving a2-a3 and pawn and prepares e2-e4. i£lbd7 .i.b 7 llc8 the blockade of the queenside, and after 1 7 'i\b5 Black's queen' s bishop would have been activated. 17 :lfel 18 a3 out this advance, he has nevertheless achieved a slight advantage. This may not be the strongest contin 'ifb6 �e8 c5 11 12 e5 13 d4 uation, since even without it White can always prevent the activation of the black pawns, Despite Black' s artificial play in the by controlling the b4 square. But in this position, how many 14 bxc4 c5 players would have been able to reject things would not have been so bad for the possibility of blockading the queen him. But he 13 . . . bxc4 g6 Here too 1 7 . . . as was probably better. Since White has managed to carry opening, after 15 c5 16 1id3 overlooks a curious manoeuvre, involving a pawn sacrifice. side ! In the given situation, the fact that Black has two connected passed pawns is of no significance. 18 19 1ha3 20 i..c3 21 i£lb3 22 ll.a5 bxaJ lta8 i.d8 lCJg7 After the exchange of bishops the white pieces will be free to occupy the dark squares. 14 b4! · · 22 23 .ixd8 24 lLla5 25 l:.e2 26 lEiel cxb4 White's king ' s laright is aiming for Black allows his opponent a pro tected passed pawn at c5, ifb7 l:fxd8 1'c7 a6 the blockading square b4·. 26 27 'i'cJ which seriously cramps his pieces, merely to 161 �f5 27 • . • 38 39 40 41 b4 This attempt to gain counterplay merely leads to Black returning his exira pawn, and remaining even without that slight consolation which he had for his positional problems. 28 'ifxb4 29 1Wc3 30 :ea2 ltdb8 1!b5 30 31 ltlxc6 32 ii.fl f6 Ylxc6 fxe5 As a rule, the superiority of a rook over a knight is especially appreciable in the endgame. 41 11fxd8 42 l:txd8 ltlb5 43 tLle5 'iti>g7 44 tLlc6 Black resigns. He loses Now, in view of the inevitable .i.fl , Black is bound to lose material. i.xb5 dxe5 1i'd3 ltxa6 1i'xa6 1ixb5 d4 'ihc5 %ha6 'I' xe5 another pawn. After the retreat of his rook, Black would have lost not only the a6 pawn, but also immediately the e6 pawn, and this would have been even more serious than the loss of the exchange. 33 34 35 36 37 1if6 ltl:f8 tLld6 tLldJ 1ic8+ lta8 'iid8 Game 301 Skold-Botvinnik Stockholm 1962 Sicilian Defence 1 e4 2 ti)fJ 3 d4 It would appear that Black can still resist, but with his next move White pushes back the enemy pieces, and the rest becomes clear. c5 g6 Jl.g7 This move is perhaps more accurate than 3 . cxd4, when not only 4 ltlxd4, but also 4 1Wxd4 is possible. · 162 .. 4 tLlc3 5 lLlxd4 cx.ci4 d6 6 i..eJ 7 i..c4 In the given situation this natural developing move is an error, which Black immediately exploits. Rauzer's move 7 f3 or else 7 i.e2 is correct. 7 8 • • • .i.b5+ lt)xe3 lt b8 tlfb6 12 13 'ifxe3 14 .i.b3 . l£lf6 l£lg4 ct>f8 After Black's knight manoeuvre he is forced to give up the right to castle, but this is not dangerous. 15 'i'f3 White avoids going into a slightly inferior ending, hoping to create some threats thanks to the weakness of Black's fl. It is hard to say which decision is the more rational. 15 16 17 18 19 9 ifd2 If the bishop moves from e3 there can follow 9 . . . 'ib6, while the active 9 i.g5, preventing 9 . . . 'ifb6 because of 10 i.xe7+, can be met, e.g., by 9 . . . h6 10 i.h4 g5 1 1 .i.g3 with a double-edged game (Ljubojevic-Sosonko, 1 978). 9 10 .i.c4 g4 h4 l:tgl 0-0-0 .i.f6 h6 1fc5 g5 h5! With a temporary pawn sacrifice Black parries the opponent's threats and regains the initiative. a6 liJc6 20 gxh5 21 'iidJ 22 f4 g4 a5 22 23 1ixc3 .i.xc3 1 0 . . . tDxe3 was more accmate. Now White could have avoided the exchange of his bishop. This leads to a favourable endgame for Black, but White is already deprived of any active possibilities. White should have played 12 .i.d4, and if 12 . . . .i.h6 13 'ii',d3 e5 14 .ic5, or 12 . . . tt:Je5 1 3 i.e2. After 23 bxc3 a4 the position of the white king would have looked very dangerous. 11 liJxc6 12 hJ bxc6 163 · 23 .24 bxc3 11xc3 f5 In this way Black securely defends his passed pawn. 25 e5 26 c4! d5 26 .i.e6! cxd5 27 cxd5 28 i.xd5 � 29 J.xe6+ <i&>x;e6 For the moment Black is two pawns down. and yet there is no longer any doubt that he will win. 30 :d4 31 l:tc4 l:txh5 l:tb4 32 l:bb4 axb4 The exchange of one pair of rooks nips in the bud the opponent's attempts to create counterplay. <it>d5 34 35 lldt+ 36 l:td7 37 'it?e2 llxh4 <it>c5 Iih3+ The enemy king cannot, of course, be allowed to go to e4. A clever move, which does credit to my opponent. It is not easy for Black to find a reply. The position is not a simple one, but Black finds the correct plan. He is prepared to give up material, to force a rook ending in which he will inevitably be able to create two connected passed pawns. 33 'it>d2 34 <i!i?e3 l:tf3 The concluding stage of the plan, begun by Black back on the 26th move. l:bf4 38 :xe7 39 �e3 40 <t>d3 lte4+ <it>d5 41 a3 42 �e3 43 <it>e2 !td4+ f4+ bxa3 There is no defence against the advance of Black's connected pawns. Apart from anything else, Black is now a pawn up. 44 l:!a7 �xe5 45 .:xa3 fJ+ White resigns A disregard for material loss for the sake of positional gains in a rook ending was demonstrated by ·capablanca in his famous game with · Tartakower (New York 1 924). 1 64 to switch earlier than normal to the consideration of tactical variations. · Grune 302 Botvinnik-Lundin 8 Stockholm 1962 English . Opening 1 2 3 4 5 c4 lDcJ g3 cxd5 li.g2 9 7 l[}f3 a3 .ll.e6 . . .. now the white queen will be able to _/ move to d2. 9 • • • f5 This game too shows that it is advisable for Black to prevent b2-b4 by 9 . a5, as Portisch played against me (Game 363). . . 10 b4 11 .i.d2 .i.f6 Ignoring the threat of 1 l . ..e4, when there could have followed 1 2 dxe4 fxe4 1 3 ltlxe4 .i.xal 14 'i'xal with more than sufficient compensation for the ex change. But if White had routinely played 1 1 i.b2, B lack would have advantageously replied 1 1 . . . e4 12 dxe4 fxe4 1 3 ltld2 e3 14 fxe3 0-0. 0-0 lDc6 0-0 A successful attempt to defer castling and to immediately make the pawn advances a2-a3, d2-d3 and b2-b4 was made in Game 27 1 . There White' s king remained in the centre until the 1 5th move. 7 8 d3 9 b4 was premature on account of 9 . . ltld4 with the threat of I O . .tb3 . Brit e5 lDf6 d5 lDxd5 lDb6 Black was not yet forced to make this move, and he could have played 5 . . . i.e6 (cf. , for example, Game 363), but subsequently he would all the same have had to either retreat the knight, or exchange it for the knight on c3 . 6 . • . ll.e7 11 12 llcl <i>b8 13 b5 lDa5 • • . If 12 . . .ltJd4 White could have played 13 ltJxd4 exd4 14 lDa4 ltJxa4 1 5 'i'xa4 with the initiative. The knight is insecurely placed here, but after 1 3 . . . ltld4 White also has a clear advantage. 14 :.b1 Defending the b3 square. Now it appears that Black's position is un promising, but Lundin, a resourceful tactician, finds an interesting possibility. 14 , White intends to advance his b-pawn, saving time on d2-d3. This forces Black . . • e4 A clever attempt to· gain counterplay. Black exploits the fact that the d3 pawn 1 65 has to defend the c4 square ·against the invasion of the enemy knight. 15 . lDel lhac4 With this temporary piece sacrifice Black rids himself of his badly placed knight, but ends up with a weak pawn in the centre. 15 . . . exd3 would have been dangerous for him� on account of the opening of the long diagonal and the increased activity of White's king's bishop. 16 dxc4 17 ltlxe4 ltlxc4 17 18 i.b4 19 il.xe4! fxe4 �e8 .i.h3 White is forced to return the piece, but with favourable consequences, the chief of · which is the weakness of Black's central pawn. 20 ifc2! hoping, then 20 . . ifxdl 2 1 .:txdl .i.xg2 If 20 .i.g2, for which Black was . 22 ttJxg2 i:.xe2 and the outcome is unclear. The move played is a subtlety, which Black did not anticipate. White sacrifices the exchange, but his pieces will dominate the board, which, in combination with his kingside pawn majority, decides the outcome. 20 21 <&ti?xfl 22 .i.xd6 23 .i.xh7 .i.Xrl lDd6 1ixd6 Of course, one black pawn will be unable to hold four enemy pawns on this sector of the front, but even more significant is the weakening of the king's position, which is furthered by the presence of opposite-colour bishops. 23 24 lDf3 25 .i.g6 26 <&ti?g2 l:ad8 11'e6 l:d5 :f8 Black's pieces have covered the breaches in his king's fortress, but the advance ofjust one white pawn destroys the defence. 27 e4 28 lib4 %td6 28 29 .i.f5 30 llc4 31 e5 i.e7 1lh6 .i.d8 Threatening to win immediately by 29 e5 .txe5 30 :h4+. 1 66 Now there is no defence against tt:lli4 . 31 32 �b4 33 1Wc3 34 11.f4 Covering the lld5 l::td2 @g8 f2 pawn and setting up an ambush (35 .i.h7+ 'i'xh7 36 'i'c4+). 34 35 lld5 Black resigns 11f3 Initially 9 ilh6 looked tempting, but after 9 . . .i.xh6 10 liJxf6+ White does not achieve anything significant, as is also the case after 9 tiJxf6+ .i.xf6 10 il.h6 ltJg8. Game 303 Botvinnik-Soderborg . Stockholm 1962 English Opening 1 2 3 4 c4 ltlcJ e4 g3 ltlf6 g6 e5 9 This arrangement of White's pawns in the English Opening was first employed by Nimzowitsch. I copied this same idea from him and used it for Black in the Closed Variation of the Sicilian Defence (for example, Game 224 ). I also later played this way with White, although in Game 306 Black developed his king's knight at e7. 4 5 .lg2 6 ltlge2 7 dJ 8 lhd5 .i.g5 . 9 ltle7 . • ltlfg8 10 'l'd2 11 Ji.el 12 ltldc3 13 f4 b6 c6 ltlf6 ltld7 14 bJ c5 15 Ilbl 16 0-0 17 lhb5 18 b4 llb8 .i.g4 lhc8 If 13 . . . ltJg4 there followed 14 ..igl . would have The stabilisation of the centre is to the advantage of White, who has the possibility of pawn breakthroughs on both wings. ltlc6 d6il.g7 il.e6 Black should not have been in a hurry to make this move; now the advance of the white knight to d5 acquires greater strength. • Now White further increases his lead in development, but Black has nothing better. So, White switches to positive actio� and after the first blow there soon follows a second . 18 167 • • . b6 19 f5 .ilxe2 not so easy for White to ,take advantage of this. lDc6 A forced exchange, as otherwise the position of the bishop at g4 may prove insecure. But without this bishop the light squares in Black's position are weakened. 20 'ifxe2 28 il.g4 29 lDd5 lDd4 g5 30 .i.xd4 White happily exchanges his inactive bishop; opposite-colour bishops nor mally favour the active side. 21 f6 White sacrifices a pawn� to gain the opportunity for play on the h3-c8 diagonal and the half-open f-file. Whether this compensation is sufficient is for the moment an open question. 21 22 23 24 25 li.h3 lDc3 a3 'iid l cxd4 lDxd5 32 cxd5 33 �c2 a5 b5 34 35 36 37 38 1fxa5 lla8 :xa5 :xaJ And this exchange is even more advantageous to White: it opens the c file. of which he seizes control, thanks to the dominating position of his bishop on the h3-c8 diagonal. lDxf6 a6 lDe7 0--0 Hoping to weaken the opponent' s pressure by exchanging the queens. Threatening fia4 in some cases, and simultaneously preparing l:b2-f2 . 25 26 l:.b2 27 l!bf2 30 31 1ia4 tDe8 '1:ic7 f6 27 . . . CiJe6 was not possible because of 28 l:.xf7 l'J.xf7 29 i.xe6� so Black decides to strengthen his position still further. Although his position is passive and his light squares are weakened, it is 1f xa5 bxa5 lk7 .flfcl �b7 The final subtlety. Black was threat ening to exchange on,e . pair of rooks by 3 8 . . . l:c3, but now this is met with 3 9 .&!a l followed by :aa7. 168 Generally speaking, in the King's Indian Defence Black has to solve more difficult problems, and from this stand poin� preference should be given to 4 c4. However, Unzicker, a staunch sup porter of open games, did not want to go into a closed opening. 4 �f6 0-0 5 .i.e2 6 i.f4 6 0-0 or 6 h3 is more often played. 6 �c6 7 d5 • 38 39 :cc7 Beginning some checks. 40 � 41 <ifalel 42 <i>d2 h5 :at+ so-called 'spite' l%a2+ l+ lUa8 l:t a Even such a massive sacrifice is unable to save Black from being mated. 43 Jbg7+ ©f8 44 l:tbf7+ <it>e8 45 .i.d7+ @d8 46 :g8+ Black resigns. Mate is just • • My opponent must probably have overlooked my reply. There was no point in freely allowing the advance of Black's central pawn. However, another development of events - 7 0-0 .i.g4 8 d5 lDb8 9 h3 i..xf3 1 O .ixf3 c6 1 1 'i'd2 also did not bring White any advantage (Lehmann-0 'Kelly, 1 966). 7 - • • • e5 four moves away: 46 . . . @c7 47 .i.c6+ �b6 48 l:b7+ <t>a5 (48 . .. <&t>cs 49 l:xb5 mate) 49 l:txb5+ and 50 :xa8 mate. Game 304 U O�vmpiad, Varna 1962 Pirc-Ufimtsev Defence nzicker-Botvinnik 1 e4 2 d4 · 3 Q)fJ 4 lDc3 g6 �g7 d6 8 dxe6 8 i..g5 is more logical, since in the game Black rids himself of any opening difficulties. 8 9 0-0 169 .t.xe6 lle8 10 !tel 11 h3 h6 g5 minimal advantage, ficient for a win. 19 .i.xd4 20 .i.xc6 White faces a choice: he has to re treat his bishop either to h2, after which it will be cut off from the queenside and the central square d4, or to e3, but then the attack on the c7 pawn is removed, and Black can advantageously open the centre. 12 J..e3 13 exd5 14 lLlxd5 15 c3 16 �exdl 'lfxdl llad8 a6 My opponent quite naturally relies on the drawing character of endings with opposite-colour bishops, especially since retreating the bishop would have allowed Black to develop his initiative. 20 i.xc6 21 ltel f5 22 f3 .i.b5 23 b4 b6 24 lhe8+ d5 lbxd5 11f xd5 Since Black's kingside is slightly weakened, it is useful for him to exchange queens:. in order to rule out the possibility of an enemy attack. although insuf It cannot be said, of course, that the manoeuvre begun with this move is incorrect. Even so, it is excessively direct, which creates the preconditions for subsequent omissions. 24 �f2 suggests itself. 24 25 a4 26 a5 :xe8 .i.c4 l:.e6 17 .i.b5 White is fully justified in aiming for further simplification. The point is that Black's bishops are slightly more active, and their pressure on the queenside pawns is appreciable. 17 . . . 18 lbd4 J.d5 .ixd4 Black goes in for an ending with opposite-colour bishops, where he has a The outcome is that the black pieces are slightly more activ�, although White has not yet any reaso.n to be concerned. But the following exchange on b6 must be criticised, since it gives Black the 1 70 theoretical possibility of creating an outside passed pawn. 27 axb6 28 � cxb6 � Nothing is achieved by 28 . . . :e2+ 29 �g 1, when the b6 pawn is under attack. 29 !tel 30 <it>xet l:Xet a5 A slight surprise. Since after 3 1 .ixb6 a4 3 2 �d2 i.fl Black could also have created a passed pawn on the kingside, White is forced to reject the Greek gift. 31 bxa5 32 g3 33 �d2 bxa5 a4 A significant inaccuracy. It would have been better to restrain the pawn not with the king, but with the bishop (3 3 i.c5). The magical drawing power of opposite-colour bishops (and with equal material ! ) apparently caused my oppon ent to relax. 33 34 <iifilc2 a3 h5 f-pawns would also have guaranteed Black a win, but probably the game could still have been saved by the manoeuvre i.b6-d8. 35 36 . . • .i.e5 f4 3 6 i..f2 was no better on account of 36 . . . gxh4 37 gxh4 <iifile6 followed by . . . cM5 and ... i.d5. 36 • • • <&t>e6! Gaining an important tempo for the defence of the f4 pawn. gxh4 37 J..c7 38 .i.xf4 Or 38 gxh4 �5 etc. h3 38 . . . 39 g4 h4! The only way to win. The h-pawns, although doubled. are strong. White will have to give up his bishop for one of them, and the other will queen. 40 i.h2 White resigns. lte2 If 4 1 <i2tb3 there follows 4 1 . . . .ixf3 42 «it>xa3 i.xg4 43 @b4 'it>d5 44 c4+ 'it>e4 etc. In this game my experienced oppon ent played too passively, openly aiming for a draw. Game 305 Filip-Botvinnik O�vmpiad, Varna 1962 Queen's Indian Defence 1 d4 2 c4 3 lDt3 4 e3 5 il..d3 35 h4 The losing move. After 3 5 f4 h4 36 gxh4 (36 ii.fl g4) 36 . . . gxf4 the two �f6 e6 b6 .i.b7 d5 It would appear . that the vanauon chosen by White caimot bring him any 171 particular benefit, since, compared with similar lines of the Slav Defence, Black has the possibility of advancing his pawn to c5 not in two moves. but in one. 6 0-0 7 bl 8 ..tb2 9 tDe5 'flc7 11 . . . 12 ti:}cJ tDe4 lbdf6 13 lbb5 14 lDxd6 'fle7 lbxd6 15 dxc5 16 .:act 17 ltfdl bxc5 l£ife4 a5 After 1 1 lDxd7 lDxd7 1 2 dxc5 !txh2+ 13 �hl ..ie5 White would not have achieved anything, but the move played weakens the central e4 square and allows Black to exploit this to gain counterplay. ..i.d6 0-0 iDbd7 Although this lmight invasion must be deemed premature, the. chances are also equal after 9 lDc3 c5 1 0 cxd5 exd5 1 1 :c l 'ile7 1 2 Vi'e2 :ads (Petrosian Polugayevsky. 1970), 9 10 11 f4 It stands to reason that Black does not object to the exchange of the c3 knight for his d6 bishop, since this makes it easier for him to control e4. A year later I carried out a similar idea in a game with Taimanov (No.3 17). c5 In order to occupy the e4 square with the other knight, and to drive back White's centralised knight by . . . f7-f6. 10 11'e2 After 10 lL\xd7 lL\xd7 1 1 dxc5 l2Jxc5 the standard sacrifice of two bishops would not even have given perpetual check: 12 i.xh7+ �xh7 1 3 'i'h5+ �g8 14 ii..xg7 @xg7 1 5 'iig4+ �h6 16 'i'h3+ �g6 1 7 'i'g4+ 'i'g5. A slightly different plan 10 lDd2 cxd4 1 1 exd4 lDe4 1 2 ttJdf3 lDxe5 1 3 dxe5 i.c5 14 iic 2 lDg5 15 l2Jxg5 iixg5 16 cxd5 h6 was tried in Vaganian Polugayevsky (1976), and again Black gained a satisfactory position. 18 l£if3! White recognises the imminent danger and takes the correct decision to exchange his fine-looking but 1 72 · ineffective knight for one of the black knights. With this he weakens the opponent' s pressure in the centre, which became possible due to 1 1 f4. In this respect it is interesting . to draw an analogy with games 134 arid 1 49. 18 19 itld2 20 9xd2 lUd8 ltlxd2 dxc4 Now 2 1 'i'c3 f6 22 .i.xc4 suggests itself, when White equalises without any problems, and he has the prospect of subsequently exploiting the half-open c-file. The decision taken by Filip is faulty, since Black is again able to occupy the e4 square. 21 bxc4 22 ifc3 23 .ifl 24 25 26 27 llfb3 lhd8+ lldl 'iixdl 'lfc7 l:xd8 l:xdl 'iic6 Both white bishops are restricted, and Black gradually intensifies the pressure. After . 20 l:.xd2 dxc4 2 1 bxc4 (2 1 Axc4 l2Jxc4 22 bxc4 Ae4) 2 1 . . . .te4 Black would have retained a minimal advantage. 20 . . . which his advantage in the endgame assumes real proportions. 28 .i.c3 29 h3 30 <3i>h2 a4 h6 lLld6 With the WleqUivocal threat of 3 1 . . .ti:)fS 32 1fd2 h5, when White will no longer be able to play g2-g4 because of . . . �h4-f3 . Therefore White hurrie s to play g2-g4, but in so doing he weakens the position of his king. 31 g4 32 . 'iid3 .i.f3 <iilf7 ile4 t'6 33 g5 33 . . . hxg5 lLle4 34 fxg5 Threatening 35 . . . 'W'c7+ 36 <itigl 'i'g3+. 35 'ittgl White resigns. After 35 . . . tlJxg5 An oversight in a difficult position. 23 ltlb7! Defending the a5 pawn and the important d8 square. Black invites his opponent to exchange the rooks. after 1 73 Black is a pawn up with the better position. 13 Game 306 Botvinnik-Robatscb 'ifd7 O�ympiad, Varna 1962 English Opening 1 2 3 4 5 6 c4 g3 i.g2 lElcJ d3 e4 g6 Ji.g7 e5 l£le7 0-0 I had already played this in Game 303 , in ·the notes to which I described the origin of this plan. It is based on the conviction that in the given situation the weakening of the d4 square is of no great significance. 6 7 l£lge2 • • • d6 f5 Now 14 @h2 f4 15 gxf4 h6 would have . led to complications, which White preferred to avoid, and so he relieves the situation somewhat. Later I employed this active idea of Robatsch in a game against Benko (No.362). It would appear to be Black's best plan. 8 0-0 In the afore-mentioned game I dev eloped my queen's knight at c6, which is perhaps more promising. 9 i.eJ b3 11 ild2 10 l£lf6 l£lh5 l£lc6 Black's knight manoeuvres are not very efficient. This set-up .could have been achieved in three moves. whereas he has spent five. In addition, White can now play his knight to d5, where it will hinder . . . f5-f4. However, for the mom ent Black retains a defensible position. 12 l£ld5 13 . i.g5 i.e6 A dubious manoeuvre. since at g5 the bishop is badly placed. 14 15 16 17 18 e.xf5 f4 @b2 .:ael .tfJ :ae8 <if;b8 lLld8 18 19 20 21 22 cxd5 l£lc3 .i.b4 1i'f2 i.xd5 li)f6 lt)f7 li)g8 l£le7 gxf5 Because of Black's cramped position, it is not easy for him to parry even such a simple threat. If he retreats (18 . . . il)f6) he loses the exchange ( 19 i.xf6 .i.xf6 20 .i.h5). Since 1 8 . . . .i.f7 seems too passive, Black exchanges the centralised knight at d5, which, however, weakens somewhat the light squares in his position. Black sacrifices a pawn in the hope of gaining some counterplay. His idea would be justified irt the event of 23 'i'xa7 lDg6. but 23 �xe7 and 24 'i'xa7 174 30 was possible. But White does not want to part with his queen's bishop, and he prefers to exchange his king's bishop. . • • lDe5 It becomes clear that Black has ·achieved his goal: his pieces are active, and the enemy king is exposed. ltlg6 31 i.g3 32 'l'fJ 1fg5 33 i.f4 34 1ixf4 35 :gt illxf4 1ih5 35 36 :1g5 37 11f3 38 l:tg2 39 �e2 l:tf6 'tlb7 .i.b6 39 40 41 42 43 :!g8 :xg2+ ffg7+ 'ifxd4 Jl..g7 Avoiding 32 . . . �xd4 3 3 'i'h5+ followed by l:.xf5 13. Now White prevents . . . f5-f4. The alternative, 35 �e 1, was not easy to assess in time trouble. 23 24 25 26 27 .i.b5 .ixg6 11'xa7 gxf4 .i.xel ltlg6 hxg6 exf4 :Ixel The exchange of rooks will all the same be necessary. 27 :!xel could also have been answered by 27 . . . g5. 27 28 'Iffl 29 d4 30 1ixf4 l::t g6 g5 gxf4 'l'e7 l:tg2 <ifi>xg2 <it>fl 1ixf5 30 ..id2 came into consideration. This position must be considered drawn, for the main reason that both 175 king lack any pawn protection. Even so, White makes some active attempts, in which he is prepared to sacrifice material, since he has the draw 'in hand' . 44 �g2 45 <ilt>g3 1id2+ 1fxb2 46 ltle4 4 7 ltlg5 c6 •e5+ 48 11fxe5 49 'iifi'f3 il.xe5+ cxd5 53 ltle6 d3 54 55 56 57 d5 <it>h5 <it>h4 @xh3 All the same the pawn cannot be saved, and now at least the white king will have to move from its active position. 45 . . . i.xc3 was not possible on account of 46 'ifc8+, 47 ifxc7+ etc. 1 4 �xd3 a4 l£ic5 l£ixb7 Black has managed to exchange White's kingside pawns, and he hopes to stop the passed a-pawn with his bishop, without the help of his king. Black is forced to go in for the ex change of queens, since White' s queen and knight have created irresistible threats� in full accordance with Capa blanca' s rule, that queen and knight are stronger than queen and bishop. Logic suggests inevitable. If 58 58 . . . j_c7 59 a6 succeeds in cutting from the d5 pawn. 58 ltlc5 Now Black is a pawn up, but this is only a temporary situation. 50 ltle6 51 �f4 With a final threat: 59 a5 i.c7 60 a6 i.b6 6 1 �d4, and the white king breaks through. �h7 58 . . 59 <itrd4 • White exploits his main trump: a pawn ending will always be won for him, thanks to his outside passed pawn. 51 52 @e4 d4 @h6 that a draw is a5 there follows .i.b6, and Black off the white king il.c7 �g4 A curious mistake, giving the ending a textbook character. · After 59 . .ias the bishop would have occupied the longer a5-e l diagonal, from which it could . not 1 76 .. I have already mentioned many times that success in chess is decided not only by talent, but also other qualities, including the character of a player. And Fischer's character was always in adequate, as the reader will probably agree, after playing through our game. In later years Fischer gained some outstanding successes, but illness apparently tore him away from . chess, . which was very, very regrettable: the chess world suffered a loss tliat was very hard to bear. be driven of(.; and then Black would have approached with his king, easily gaiping a draw.· Now, however, White immediately gains control of the critical a5 square. 60 ltlb7! 61 'it>xd5 62 �c6 63 ltld6+ 64 ltlc8 Black resigns. The <i>f5 .il.b6 .i1..a7 @e.6 a-pawn queens by force. In this game grandmaster Robatsch underestimated the dangers in an ending with bishop against knight, just as Unzicker did in Game 304 in an ending with opposite-colour bishops. However, it has to be admitted that my opponent played more resourcefully here than he did in Hastings (Game 299). 1 c4 2 d4 3 ltlc3 g6 lLlf6 d5 So, the Grtinfeld Defen�e. Interest ingly, it was only in this game that I was able to use some analysis that I had prepared for the return match with Smyslov ( 1 958). 4 ltlf3 5 ffbJ Game 307 Botvinnik-Fischer O(ympiad, Varna 1962 Grtinfeld Defence This was my only meeting at the chess board with Robert Fischer (born 1943). True, seven years later the Leiden Chess Society held urgent dis cussions with us about staging a training match, but they were unsuccessful. Ten years after the Olympiad in Bulgaria, the young American won the title of . World Champion, but already long before this he enjoyed great popularity. I think that tlris was assisted not so much by his eccentric opinions and actions, as by the lively, dynamic play which he demonstrated in Iris games. .i.g7 Fischer knew, of course, that I employed the Ragozin Variation, which begins with this move. Soon it also transpires that the Smyslov Variation suits us both. 5 6 1fxc4 7 e4 dxc4 .i.g4 0-0 It is this move that characterises the Smyslov Variation - one of the most original methods of piece play against a pawn centre. 8 .i.eJ 9 �e2 ltlfd7 It would seem to be more promising for White to castle on the queenside, than on the kingside (and also than to refrain temporarily from castling), but 1 77 he chooses a more cautious plan. Later I came to the conclusion that 9 'ib3 should be considered. 9 ... and I did not know this ! Now White tries to seize the initiative. 14 d5 li)c6 But not 14 llJb5 on account of 14 . . . 'i'xc5 1 5 dxc5 �4, and not 14 e5 1fxc5 15 dxc5 lbcl7 16 f4 g5. Black too does not follow the recom mendations of theory, which considers that the fate of the queen's knight can be determined later. and that for the moment 9 . . . ltlb6 should be played. 10 :d1 11 1lc5 li)b6 11 12 h3 1ld6 • .i.xf3 13 %tfd8 1lf6 16 . . . 17 e5 lhed7 • The f-pawn is forced to advance, so that it should not be under attack. Of course, White could not be satisfied with 17 "Wixc7 'i'xb2. If 1 3 i.xf3 there would have followed 1 3 . . . 'ifxc5 14 dxc5 liJc4 ! It was this position that I had analysed in the Winter of 1958. True, in Varna, Semion Furman later told me that by playing 1 3 . .. e6 Black could equalise, but during the game Fischer 15 . . . 16 f4 • Of course, 1 5 f4 is bad for White on account of 15 . . . ltJec4, e.g. 16 .i.xc4 'i'xc5 17 .txc5 ltJxc4 1 8 e5 lDxb2 1 9 :d4 f6. After 1 2 d5 ltJe5 1 3 ltJb5 'i'xc5 14 i.xc5 Black would have eliminated the threats by .. 14 . . . c6. i2 . . . 13 gxf3 lt)eS . 1 1 'i'd3 .i.xf3 1 2 gxf3 e5 13 d5 ltld4 leads only to an equal game. • • 14 15 li)b5 In my analysis I had examined two continuations: 17 . . . 'i'h4 18 ila3 g5 19 l:d4, and 17 . . . ilf5 18 ilb4 a5 19 'i'd4 c5 20 dxc6 bxc6 2 1 .i.g4 c5 22 'i'xd7, and I came to the c<;>nclusion that White retains the advantag�. Fischer, however, found a third. possibility. 178 17 . . • 11xf4 A move, typical of Fischer's energetic style. It is true that here too White still has a good game, but the surprlse shocked me. I was extremely vexed by my poor analysis, and so sub sequently I made several weak moves. i.xf4 The point is that if 1 8 'ii'xb6 Black could have continued 18 . . . 'i'e4 1 9 f3 'i'h4+ 20 .tf2 'i'b4+. after 24 .i.c4 (with the intention of continuing llel-e7) White would have activated his pieces still further. Fischer thought that he could have forced a draw by playing 24 . . . ltle6 25 i.h2 tfJ.d4 (with the threat of 26 . . .ttJf6) 26 :bl i.c3 27 %kl .ib2 etc. (here and sub sequently, Fischer's opinions are cited from the Russian edition of his book My 60 Memorable Games, Moscow 1 972). However, it is not clear why White should repeat moves, when instead of 25 i.h2 he can play 25 i.g3 (25 . .. ctJd4 26 :d1). 18 18 19 tfJ.xc7 20 d6 21 exd6 22 0-0 tfJ.xc5 lt.ac8 exd6 i.xb2 tfJ.bd7 23 ltd5 23 Now White controls the e5 square. • . • b6 24 25 lLlxe6 lLle6 25 26 :d3 27 lle3 fxe6 lbc5 e5 28 29 30 31 i.xe5 :xd6 :d7 thxd7 • Black is a pawn up, but the white pieces are very active. Black tries to restrict their mobility by preparing . . . �e5, but he commits an inaccuracy. Correct was 22 ... ttJcd7 23 i.f3 i.e5 24 i.xe5 ltlxe5 25 .ltxb7 llb8, when White has only a minimal advantage. • • White completely loses his head. He assumed that the alternative 25 ..ih2 ltJd4 26 i.g2 lDf6 was even worse for him, but Geller pointed out 26 l:txd4 i.xd4 27 l:.el, with which Wl.&_te retains a good game. Here too Fischer did not agree: he continued the variation by one move - 27 . . . i.c5. But we too can continue the variation - 28 ltld5. Now, however, after the exchange of the c7 knight, things · reduce to a prosaic endgame. Before this move I was still hoping for 27 . . . i.d4 28 :a3 e5 29 .ig5 l:txd6 30 i..e7 l:.d7 3 1 .i.g4, whereas now Black eliminates White· s passed pawn without loss of time and obtains a won ending. 24 .i.f3 A poor move, since there is nothing for the bishop to do here. Meanwhile, 179 Jl.xe5 l1xe5 lle7 llxd7 3 8 . . . lle l , after which the white bishop is unable to occupy the favourable c2 square (39 i.c2 llcl). 39 il.c2 40 cilf6 'iti>g5 41 <.t>gl lhe4+ 42 .i.xe4 lbe4 � An endgame specialist of the class of Capablanca or Smyslov would have immediately transferred his king to d6, defending the knight, after which the advance of the queenside pawns would have decided the outcome. This is when the defects in the character of my opponent began to tell. Reckoning that the game was easily won for him, he was angry with me for playing on, and in his fervor, already after the time control, he talces a rash decision. By 4 1 . . .%:tb4 42 a3 :d4 43 f3 a5 Black could have created a zugzwang position: White' s king has to guard the h4 square, his rook - c4, and his bishop - dl. A s for the resulting rook ending, in the last few moves Black has worsened the position of his king, and this is of great significance. 32 .i.g4 This move is wrong, since the rook ending does not leave White with any chances of saving the game. 3 2 l:ie 1 <ii>f8 33 i.d5 was stronger, when he is still prepared for a prolonged resistance. 32 33 ltet 34 �g2 l:lc7 � White misses a chance to move his bishop from its bad position by 3 4 i.e6+. 34 35 lteJ 36 ltf3+ thc5 %le7 36 37 llcJ 38 .i.dl �g7 l:e4 l::td4 • • • 36 �f3 suggested itself. so as after the exchange of rooks to occupy the central d4 square with the king, and place the bishop at c2 and the pawn at f4. Alas, then the bishop would be lost 36 . . . h5 37 l:.xe7+ (37 ..ic8 %:tc7) 37 . . . @xe7 38 ..ic8 �d8. After this carelessness by my young opponent I acquired some hopes of saving the game. I was expecting 1 80 Although 42 moves had already been made, play . continued. I was not in a hurry to adjourn the game, since I thought that, in such a position, further play at the board gave more chances of drawing than after home analysis. And Fischer's entire behaviour expressed his indignation at White continuing to resist in this 'hopeless' position. .He clearly wanted to demonstrate both to the players in the Olympiad, and to the spectators, that such a position did not require any analysis. But let us see how the game continued. 43 l:taJ From what follows, if we look at the next diagram, it is evident that the immediate 43 lk7 was correct, not fearing 43 . . . l:ta4. But, as the reader will see, to find such a difficult decision at the board was impossible. However, Fischer too promptly pays the cost of playing on without home analysis, mis sing the winning continuation 43 . . .a5, and if 44 :b3 :b4. The whole point is that the pawn ending is hopeless for White: 45 !txb4 axb4 46 f4+ �5 47 @f3 �e6 48 �e4 <it>d6 49 �d4 b5. But if he avoids the exchange, then the rook ending too is won for Black. 43 44 • • • l:tfJ lle7 J:lc7 After 44 . . . <it>h6 and 45 . . . �g7 White would have had to seek other ways of saving the game, and perhaps he would have been altogether unable to find any. 45 a4 The time for play had expired, and Fischer had to seal his move. After dinner I began �y night-time analysis, one of the mosfprolonged· in my career. 45 . . . llc4 did not concern me, on account of the immediate 46 a5 bxa5 (or 46 . . . b5 47 J:tf7) 47 %If7 a6 48 h4+ 'ifi>h6 49 :d7 with a draw. 45 . . .<it>h6 46 %td3 :cs 47 h4 %:ta5 48 l:ld4 is also not dangerous for ·white. But what was White to do after 45 . . . :cs ? For a long time the future appeared gloomy, until Geller found for White a unique idea of counterplay. When deep into the night he left me, it only remained for me to work out the · details of this find. 45 • • • l:r.c5 This means it was not in vain that we spent so much time on this move. 46 l!ti �a5 It can be seen that, if the immediate 43 :tc7 :a4 had been played, basi�ally the same position would have ari�en. But what is the secret of the analysis? 47 :xh7!! ' I overlooked this defence', Fischer later wrote. There is nothing surprising about this - after the game it transpired that during the night my opponent had slept soundly. In addition, the decision to allow · the opponent two connected 181 passed pawns looks extremely paradox ical. It is base(:! . on the fact that, when these pawns begin advancing, the b6 square is weakened. Then, using checks, the black king will be forced away from the g6 pawn and it will be won, when White too will have two passed pawns on the kingside. 47 48 • • 51 52 A thematic variation, giving concrete form to the previous note, would have looked like this: 48 . <itf'6 49 l:tb7 l:.a5 50 �g4 b5 5 1 f4 a6 5 2 Itb6+ <:/;fl 5 3 !tb7+ with a draw. .. 49 50 51 .n.n+ l:g7 � �es flat 52 53 54 55 56 What can Black do now? We have al ready seen that nothing worthwhile comes of 5 1 . . . �6 52 �b7. Had Fischer made a thorough analysis of the ad journed position, he would most prob ably have reached this position and chosen 5 1 . . . <i>d4, the most unpleasant move for White. However, he was confident of winning, and had analysed the ending light-heartedly. • • b5 h5! The surprising thing about this move is that White does not pick up the undefended g6 pawn, but temporarily sacrifices another pawn himself. Fischer admitted that he ioverlooked this reply when he playe4 5 1 . . . b5 - a rare instance for the future World Champion. Incidentally, after this move I could not restrain myself. and, going up to our team captain Lev Abramov, I said to him one word: ·praw' . Great was the general astonistu:nent when later we learned that at this point Fischer had protested to the; deputy arbiter, that 'Botvinnik was being prompted during the game' ! ltxa4 <it>f5 • h4+ • Towards morning I had found a worthy reply to this move. <li>g2 ltg5+ l:txb5 f4 llaJ+ gxh5 'ifi>d6 h4 <ifi>c6 The reader should not be surprised at how long the play dragged on in this theoretically drawn ending. There are players who find it hard to adjust to a tum of events that is unexpected for them. 57 58 59 60 61 62 63 :bs 'ifi>h2 f5 ltb5 f6 :b6+ lta6 hJ+ a5 cl;c7 '3;d6 cl;e6 rM7 · Here Black should have offered a draw. After all, being a pa� down, QY the unwritten· laws of sporting ethics I 182 could not be the initiator of peace negotiations. 63 · 64 65 66 67 68 �g6 a4 ltc6 l:ta6 l:r.c6 l:la6 @gt .&tg6 + 'iit> a5 60 ltg5+ 'it>a4 61 l1g4+ <i>a3. Here 62 l:h4 b2 63 b7 blif 64 h8'1' is inevitable. My evaluation of the position was as follows: 'Black cannot win, since his rook is restricted by his king. ' Fischer continued the analysis: � l:t d3 a3 64 'ii b3+ 65 <it>e2 W'dl+ 66 �e3 ltbl . • • Only here, with his face a white as a sheet, did Fischer shake my hand, and with tears in his eyes he left the hall. However, our battle did not end at this point. The main topic was the question of whether or not 51 �d4 would have won. ••• After 52 l:lxg6 Black begins advan cing his pawns: 52 b5 53 h5 b4 54 b6 • • . b3. 54 . . . J::th l 55 <itrg2 :lh5 56 l:ta6 b3 57 l:txa7 l:.xh6 58 l:.b7 <i>c4 59 <bf3 leads to a theoretical draw. Now White will pursue the enemy king. Not wishing to allow the rook onto the b-file, it will be forced to occupy an unfavourable position: 55 l:tg4+ <it>c5 56 J:lg5+ �c6 (56 . . . <it>b4 57 l%.g7 b2 5 8 h7 llhl 59 :Ixa7) 57 :g6+ �b7 58 llg7+ �.a6 59 183 and, after making the further moves 67 iff8+ <i>a2, he concluded that 'White's king will be without she�r from the coming avalanche of checks' . Here, however, there are straight away two mistakes. First of all, as I established, after 68 1lc5 White can successfully defend. This was later demonstrated in detail by the Moscow master Anatoly Kremenetsky in the magazine Shakhmaty v SSSR ( 1 977, No. 2). And the 1 3-year-old Garry Kasparov (later to be World Champion) found an elegant way to draw in the diagram position: 67 l:tc4! �bl+ 68-:cJ 11e1+ 69 <i>d3 1ffl+ 70 �d2 (70 <t>e3? 1ib3+! ) 70 W'xf2+ 71 �d3. .•. This, it would seem, put an end to the lengthy arguments involving our game. They forced much racking of brains, plan is to develop his · bishop at e2, castle kingside, and then ·begin the standard pawn attack on the queenside. Incidentally, this is what happened in the game Reshevsky-Geller ( 1953), where after l 1 . . ...i.f5 White carried out this plan unhindered. The course of events ex-plains, how ever� why theory considers it more sensible to develop the knight at f3 . both during the play, and especially in analysis. After exanfuring this game and the associated eve"nts, the reader will prob ably agree that Fischer' s human charac ter did not match his great chess talent. Game 308 Petrosian-Botvinnik World Championship Afatch Moscow 1 963, 1st game Nimzo-Indian Defence 1 2 3 4 d4 c4 lLicJ 'ii'c2 4 5 6 7 cxd5 ..tg5 .i.xf6 lLif6 e6 . i.b4 This continuation was popular in the 1 930s, but then it went out of fashion. It guarantees White against any surprises. but it is hard to expect any real achieve ments from it. If White wants to avoid compli cations. this move is necessary, since after 7 i.h4 c5 Black can try to seize the initiative. This is well illustrated by Game 1 14, Keres-Botvinnik ( 1 94 1). Now, however, a further exchange of minor pieces is unavoidable. 7 8 9 10 aJ 1fxc3 eJ 11 �2 1fxf6 i.xcJ+ c6 0-0 White decides that the best square for his knight will be f4 . His subsequent lte8 11 d5 exd5 h6 This simple move prevents the knight from going to f4, and now the plan, which White had apparently outlined before the game, proves impossible. Was this the reason for Petrosian' s subsequent uncertain play? 12 ltlgJ The natural solution - the knight has to make way for the bishop! It is doubt ful whether 1 2 ltJcl would have been any better. 12 • • • g6! Only this energetic move, creating the threat of . . . h6-h5-h4, can cause difficulties for White. .Anything else would have left him free to complete his development. 184 As a result, it has to be acknow ledged that the entire opening idea, involving 1 1 ll'le2, can hard:ly be approved. 13 f3 This move could be understood, only if White were intending to castle queen side, since then the f2 pawn would have to be safeguarded. But since White subsequently avoids (quite reasonably! ) this possibility, f2-f3 proves to be a loss of time, to say nothing of the fact that it weakens his position. 13 14 ... .te2 h5 lDd7 Black is not in a hurry to play . . . h5h4, as long as the lrnight has a retreat square at fl . But if 1 5 0-0 there would have inunediately followed 15 . . . h4 16 tiJh l 'i'g5 (but not 16 . . . 1i'e7 because of 17 e4 dxe4 1 8 fxe4 'ifxe4 19 1Lc4) and then, for example, 17 e4 . dxe4 18 f.xe4 Itxe4 1 9 �c4 (or 19 'i'f3 CDf6) 19 . . . ltlf6 20 'i':f3 �f5 with an obvious advantage. Therefore White altogether gives up the right to castle, defending the e3 pawn with his king and leaving fl free for his lrnight. 15 16 17 @f2 tDfl tDd2 17 18 19 l:hel h3 h4 lt:)f8 It is obvious that White must bring his king's rook into play. It is no less obvious that Black should double his rooks on the central file. lle7 �f5 This move has only one plus: White is freed from the need to calculate every time the variations with . . . h4-h3. But its positional minuses are also obvious: the g3 square is weakened, and · the possi bility of advancing g2-g4 is no longer on tl1e agenda. 19 20 21 tLln 1'd2 kiae8 lDe6 Sticking, and not without reason, to waiting tactics. If, for example, 2 1 i.d3, then 2 1 . . . cS ! 22 .ixf5 cxd4 23 exd4 'i'xf5 24 .U.e5 'i'f4 25 .:tae l :c7, and White has serious difficult�s. 21 I spent a tDg7 long time considering 2 1 . . .iDgS followed by a piece sacrifice 185 on h3. After 22 <itigl �xh3 23 gxh3 ctJxh3+ the white king has three crucially different moves: (a) 24 <itih2, when things most probably end in mate: 24 . . . �xe3 25 t:bxe3 'i'f4+ 26 'it>hl tbt2+ 27 @gl 'ilg3+ 28 @fl (or 28 ltJg2 h3 29 i..fl h2 mate) 28 ... tt:lli3 29 ..tdl ifgl+ 30 <itie2 liJf4 mate: (b) 24 @hl 'illg5 25 @h2 'i'gl+ 26 'it>xh3 :!xe3, when it was not easy to suspect that despite his two extra pieces, White is helpless - he does not have a single satisfactory move; (c) 24 \tig2 lhe3 25 @xh3 (25 l2Jxe3 'iig5+) 25 ... :xe2. True, White could have tried to parry the attack by 22 ifdl (instead of 22 @gl), but then all the same there would have followed 22 . . . il.xh3. and if 23 gxh3 t:De4+! There can follow 24 @g2 'i'g5+ 25 'it>h2 t:Df2 26 'i'd2 (26 f4 'i'f6, but not 26 ... 'i'f5 27 iLg4) 26 ... l:lxe3 27 .ltdl 'i'f4+ 28 @g2 lbxdl 29 !%.axdl 'i'xf3+ 30 <t>gl l:e2 3 1 lhe2 .D.xe2 32 'ifg5 'i'f2+ 3 3 <t>h 1 lhb2. and White would have to resign (variation by Averbakh). This would have been a logical conclusion to the game, and would have vividly demonstrated the danger of such ill-founded waiting moves as 13 f3 and 19 h3. Now, however, Black has merely some positional advantage. and the battle continues. · 22 lladl 23 :ct 24 :c3 25 �gt 26 .i.dt tZ\h5 ild6 lbg3 tZ\h5 :e6 27 'i'e7 1if2 Black has completed all the neces sary preparations for beginning the advance of his kingside pawns - without this White's fortress cannot be taken. 28 29 30 .i.b3 Jl.dt g4 g5 .*.g6 It is very curious that, at precisely the moment when White does not face any immediate threats, Petrosian, who throughout the game has stuck to cautious tactics, tmexpectedly and in correctly launches into complications. Who knows. perhaps he was hoping to exploit his opponent's imminent time trouble? 30 31 • • • li)xg3 hxgJ lDf4 Apparently White overlooked or underestimated this move, after which a lengthy resistance is impossible. 32 'ii'h 2 33 1id2 c5 32 ...llJd3 would also have won. Of course, not 33 . dxc5 d4, or 33 �xc5 l:txe3 . · 33 186 • • • c4 34 i.a4 34 35 ... .i.c2 Merely provoking the advance of the b-pawn, which is advantageous to Black. b5 Or 35 i.xb5 :bs 36 i.a4 liJd3 . 35 36 37 38 <t>t'l �g2 exf4 6 exd4 d5 7 8 cxd5 'ifb3 lhxd5 This position could have been reached from a well-known variation of the Caro-Kann Defence. if instead of the strongest move . . . i.°g4 Black had played . . . g7-g6. lBxb3+ 'fff6 lBf4+ Practically forcing the exchange of knights. since 8 . . . liJb6 9 d5 or 8 . . . e6 9 ii.gs is advantageous to White. Otherwise 3 8 . . . i.xc2 and 39 . . . ltJd3. lhel 38 39 fxg5 lie6 40 f4 lie2+ White resigns, as he loses a piece. 8 ... 9 .i.c4 tLlxcJ Grune 309 Botvinnik-Petrosian World Championship Match Moscow 1963, 4th game English Opening 1 2 c5 c4 t;)cJ lLJc6 This game shows that if Black wants to fianchetto his king's bishop, he should do this immediately: 2 . . . g6 3 tDf3 i..g7 4 d4 cxd4 5 lDxd4, and only now play 5 . . . ltJc6. 3 4 lLJf3 el A cunning move. pr�ared in the quiet of my study. Black cannot play either 9 . . . lDe4 10 .ixfi+ @d7 1 1 'iie6+. or 9 . . . i..g7 10 i.x:f7+ � 1 1 bxc3 lDas 1 2 'i'd5. 9 g6 . An important subtlety! The threat of d2-d4-d5 forces Black to take cow1ter measures. 4 • • . lBf6 If 4 . . . i.. g7 there would have followed 5 d4, and White will. :always be able to play d4-d5. This occurred in Game 349. 5 d4 cxd4 • • • e6 9 . . .lDd5 10 i.xd5 e6 1 1 i.xc6+ bxc6 was also possible, but Petrosian prob ably chose the right course. 10 bxcJ i.g7 If 10 . . . l'.Da5, then 1 1 i.b5+ i.d7 1 2 1Wa 4 liJc6 1 3 d 5 ! ! exd5 1 4 0-0 with a strong attack for White. This trap, also found in home analysis, is successfully avoided by Petrosian. 1 87 As a result of the opening · battle, White has provoked . . . e7-e6, weak ening Black's position, and has also gained a tempo. 11 .i.aJ This exchange safeguards White against the threat of t'.1Ja5, since he will have a queen check at b4. After 1 5 0-0 tDa5 1 6 'i'a4 .ixa3 1 7 .i.xd7+ 'i'xd7 1 8 'i'xa3 lbc4 Black has nothing to fear. It is interesting to note that with out 1 5 . . . tDa5 Black cannot equalise: 1 2 .i.b5 Ad7 1 3 0-0 .i.xa3 14 'ifxa3 "fle7 1 5 'i'b2 0-0 16 .:tfe l (Andersson Huguet, 1 973 ). . . . �f8 15 . . • �xf8 16 0-0 <:J;;g7 The surprise in the opening has not unsettled my opponent. This quite well-known manoeuvre, which forms the basis of Black's defen sive plan. was one that I underestimated during my preparations for the game, and, frankly speaking, I was surprised that Black was able to avoid the itmnediate danger. I needed time to gather my thoughts, and with this aim White repeats moves, but even so he does . not take the optimal decision. 12 il.cl· . Tai played more strongly against Pohla ( 1 973) : 12 0-0 i..xa3 1 3 'i'xa3 'i'e7 14 'i'c l , and Black was unable to eliminate the defects of his position. Sl.g7 12 . . • 13 .ii b5 i.. d 7 14 15 i.aJ · .ixf8 .i.f8 After 13 . . . 0-0 14 .i.xc6 bxc6 15 .ia3 :es 16 0-0 White has an obvious advantage. 17 il.e2 The turning point. Despite his resourceful defence in the opening, Black would still have had to overcome considerable difficulties in the event of the prosaic 1 7 .i.xc6 i.xc6 1 8 lDe5 ifd5 19 f3 ! (1 9 . . . f6 20 lDxc6). Now the ex change of queens on b3 leads to a better ending for White, but if Black avoids the exchange, the position of the white queen at b2 will be highly unpleasant. Unjustifiably avoiding tl:iis variation; White allows his opponent to gain com plete equality. 1 88 17 18 19 20 21 c4 'i'e3 li)e5 Jladl b6 'iff6 :he8 :ad8 ile7 6 7 0-0 a4 7 8 'ife2 a6 Petrosian knew that in this variation of the Queen's Gambit . Accepted I preferred this move of Rubinstein. My opponent was happy to go in for this line, assuming that nothing new could be devised here. Even so, it should be said that 7 'i'e2 is more often played. • • • �c6 iJ.e7 In the opinion of theory, 8 . cxd4 is stronger, but after Petrosian had gained reasonable play in the 6th game of the match, he decided to repeat 8 .i.e7. .. ... 9 dxc5 Here the avoidance of simplification by 22 f4 f6 2 3 tiJf3 ti'd6 would have led to double-edged play. Therefore White takes a wise decision - to force a draw. lDxe5 22 c5 23 24 11xe5+ 'iff6 cxb6 axb6 Draw agreed Black is obliged to exchange on e5, in view of the threat of lbc4-d6. 9 Game 3 10 Botvinnik-Petrosian World Championship Match A1oscow 1 963, 8th game Queen's Gambit Accepted 1 2 3 4 5 d4 c4 lt)f3 el .i.xc4 d5 dxc4 tiJf6 e6 c5 .itxc5 Annotating my game with Keres from the 194 1 Match-Tournament, I pointed out than that if 9 dxc5 Black can advantageously reply 9 . . . lLle4. Just in case, Petrosian avoids this move, fear ing some surprises, especially on the principle of ' let well alone', since in the afore-mentioned 6th game 9 . . . i..xc5 fully justified itself. For my part, I did not fear 9 . . . lbe4, for the reasons outlined in the notes to 189 all, Black is threatening to win the e5 pawn only with his knight, and his centralised queen may become a target. Game 1 6 1 , and although there was a mistake in the analysis, the overall evaluation remained accurate. The point is that Polugayevsky would appear to be right, in recommending that if 9 . . . CDe4 White should immediately play 1 0 ltJd4 ! tL\xc5 ( 1 0 . . . ltJxd4 1 1 exd4 'i'xd4 1 2 :ct1) 1 1 lDxc6 bxc6 1 2 4Jc3 . 10 li:)g4 e4 After 10 . . . 'i'c7 1 1 e5 ltJg4 1 2 .if4 f6 1 3 liJbd2 ! (but not 1 3 exf6 'i'xf4 1 4 i.xe6 i.xe6 1 5 1 5 'i'xe6+ @fK although according to theory White can never theless count on a draw) a position from the game Botvinnik-Vilner (Leningrad Championship, 1 930/3 1 ) is reached. A possible continuation is 1 3 . . . ltJgxe5 (Vilner preferred first to play 1 3 . . . 0-0) 14 tL\e4 i..e7 (or 1 4 . . . ila7) 1 5 lDxe5 fxe5 ( 1 5 . . . ttJ.xe5 16 �ac l ) 1 6 ilg3 0-0 1 7 'i'g4 with the initiative for White. 11 e5 Up till here we had repeated the previous even-numbered game, where after 1 1 i.f4 'i'f6 1 2 .tg3 lbge5 1 3 ltJxe5 tL\xe5 14 ltJd2 Black achieved good play. I should mention that Tal recommended playing 14 �a2 0-0 15 ltJc3 . with the intention by <iifrh l and f2-f4 of preparing an attack. However, for the 8th game I had prepared a pawn sacrifice, and I wanted to test it. 11 • • • li:)d4 Since. White cannot reply 12 'i'e4 on account of 12 . . . lDxf3+ 1 3 gxf3 lDxf2. he is forced to exchange knights. 12 t£ixd4 13 l£ia3! 13 14 15 l::t xaJ bJ The simplest. While defending his bishop. White creates numerous threats: 14 lDc2 and 14 h3. It can also be considered that for the moment his central pawn is immune: 13 . . . ltJxe5 14 i.e3 (14 l:d l 'i'g4) 1 4 ... 'i'd6 1 5 l:tfdl 'ilc7 16 i.xc5 1 6 ifxc5 1 7 i.b5+ axb5 18 :ac 1 ltJc4 1 9 ltJxb5 etc. Here my opponent realised that he had walked into a prepared variation. To his credit, he chose a continuation which, although dangerous, was the most logical, depriving White of his king's bishop. 1f xd4 This position had to be analysed in · detail before the game. of course, and I came to the conclusion that previously it had been incorrectly evaluated. After .i.xaJ t£ixe5 In search of something better than this prosaic move, I thought for 45 min utes. since I realised that the decision taken in my preparatory :;malysis was not ideal. After the game it seemed to me that it would have been better to play 1 5 :Cdl 1 90 White would have retained some advan tage thanks to the poor position of the enemy king. Now, however, the advantage passes to Black. 'i'g4 ( 1 S . . . 'ifxc4 16 'ii'xeS 0--0 17 .i.h6 'ifg4 1 8 :Iad3 'i'g6 1 9 .i.xg7 f6 20 ..txf6 'ilxf6 2 1 .:1.g3+, or 18 . . .f6 19 'Wic7 'i'g6 20 i.xg7) 16 'i'c2 (but not 16 f3 'i'fS) 16 . . . tDxc4 ( 16 . . . ifxc4 17 :c3, 16 . . . i.d7 17 ..te2, or 16 . .. 0-0 17 ..te2 'i'fS 18 'i'c7) 1 7 l:tg3 'lbs 18 :gs 'i'h4 19 l:txg7 with the threat of 20 i.gS. Failing to find these possibilities at the board, I made the previously planned move (the drawback to which is that the rook is temporarily shut out of the game), after, alas, wasting much time in thought. There were also other recommend ations. After l S �dl 'i'g4 Aronin suggested playing 16 'i'd2 0--0 17 .i.e2, Keres mentioned lS ..ia2, and Matanovic - 15 !tc3 0-0 16 %Xdl. But would this have given White a clear advantage? 'ilc5 15 . 16 . 16 17 18 19 bxc4 .i.a3 l:td2 if)xc4 .i.d7 iff5 Hardly any better was 19 'i'd2 f6 (20 'ilb4 .0--0-0), but Polugayevsky con sidered 19 'ilb2 to be stronger. 19 • . . .i.c6 Black avoids queenside castling on account of 20 i.e7 J:de8 2 1 i.d6, when his king feels uncomfortable, and parries the_ threat of 20 :cts. 20 :et h5! White's uncertain play has been excellently e;...-ploited by Petrosian. With the intention of finally safeguarding his king by . . . f7-f6, he prevents a possible g2-g4. 21 'i'e3 • l:la2 Sensing that the fruits of my home analysis had been wasted� as in my game with Fischer I begin playing weakly. By continuing 16 ..tb2 ltJxc4 (or 16 . . . f6 1 7 'i'hs+ g6 18 'i'h6) 17 bxc4 i.d7 18 !tg3 0-0-0 19 l:xg7 191 21 • • • f6 Since 2 I.. .i:th6 is risky in view of 22 I;ld5 (or perhaps even 22 'ib6 kig6 23 'i'c7 l:xg2+ 24 'it>fl l:tgl+ 1 7 25 �xgl 'i'g5+ 26 <it>fl. when Black can only simpler. when it is hard for Black to strengthen his position. hope for a draw), Petrosian prefers to take play into an ending, where he will have a ininimal advantage. Here I offered a draw, but there was no reply, and the game continued. 22 23 24 25 1ixe6+ l:txe6+ l:Ie7+ a5 The immediate possible. 26 27 .i.d6 .:txe8 'If xe6 <3Jf7 lt>g6 �ad8 25 .. J:the8 33 34 35 36 ... fJ was l:.he8 c.t>f2 Jl.b4 i.a3 i.d6 'ittf4 g4 .i.c4 .i.b5 l:ic3 40 lte3 :c4+ And now the highly important fourth rank is in Black's possession. lixe8 �el+ l:tal ltbl :b3 l:tc3 36 37 38 After the g-pawn has advanced, another pawn comes under threat. 39 :le4 .i.c6 111.is is the whole point: White has not managed to play �f2. 29 30 31 32 Ji.e6 All the same the c-pawn would have had to advance, but now Black can threaten the g2 pawn. White incorrectly activates the black rook. After 2 7 J:.e3 It.xe3 28 fxe3 :es 2 9 <t>f2 the draw would have been obvious. 27 28 l:t.c2+ .i.d7 'iitg3 h4 c5 41 42 43 @gJ il.c7 fxg4 44 ll.f4 �a4 bxg4 .i.d7 Now White can no longer defend all his weak pawns: one of them is lost. If 44 h5+, then 44 . . <:Jilf7. . Of course, not 44 . :xas, when there would have followed 45 l::te7 J:.a3+ 46 i.e3 .i.c6 47 h5+ <itih7 48 h6. .. 33 :d4 White incorrectly concedes the · second rank. 3 3 c5 :a3 34 i.c7 was 1 92 45 l:lb3!! Game 3 1 1 The obvious 45 h5 l:xa5 46 i.d6 i.e6 47 g5 £xg5 48- :f3+ �e8 49 .f:.e3 @d7 50 :Le5 l!af+ 5 1 @fl g4 52 :gs would probably have lost to 52 . . Jlf3+ and 53 . . l:lf7. The move played, which was found in adjournment analysis, lures the black bishop to c8, since from c6 it will not be attacking the g4 pawn. And this is sufficient for a draw. Botvinnik-Petrosian World Championship Match Moscow 1963, 1 0th gaine Queen' s Gambit Accepted . 45 46 . • . g5 . ... fxg5 Now the draw is obvious: after the exchange of rooks, Black will not have the two widely separated passed pawns at a6 and f6, necessary for a win in the ending with opposite-colour bishops. 47 48 49 50 51 52 hxg5 .i.eJ :b6+ c6 lhc6 lk5+ d5 dxc4 ltlf6 e6 8 1i'e2 cxd4 After his lack of success in the opening of the 8th game (No. 3 1 0.), Petrosian avoids 8 . . . .ie7 and reverts to the continuation that occurred in the 2nd game. lba5 @g6 <it>f5 bxc6 l:laJ 9 10 11 In the event of 52 llxc8 11.xe3+ Black would have retained an advantage. But if now 52 . . . �e4, then 53 :xc8 l:xe3+ 54 <Jtg4, and a draw is inevitable in view of the wealmess of the g7 pawn. �g6 52 53 54 55 d4 c4 tDf3 el .i.xc4 c5 There has probably been no impor tant event, other than the 1 963 World Championship Match, in which the given variation occurred so frequently, although the course of the play showed that in this opening too it is not so easy for Black to equalise. a6 6 0-0 ltlc6 7 a4 .i.c8 White rids himself of his weak pawn, and threatens after 47 gxf6 gxf6 to obtain a passed h-pawn. 46 1 2 3 4 5 %1.dl exd4 lDcJ il.e7 0-0 And here White deviates from the afore-mentioned game ( 1 1 .i.g5). 11 • • . "" ltlb4 1 1 . . . tiJd5 deserves preference, as my opponent played in the 1 6th game of the match (No.3 14). Itc6+ �h5 l:.a4+ <it>f4 <it>g3 Draw agreed 12 A game, typical of my play in the match with Petrosian. Interesting ideas, but unsuccessfully implemented. .i\g5 A slight surprise, the true worth of which was not immediately appreciated by the future World Champion. Earlier I had played 1 2 tiJe5 (see, (or example, Game 44), when " with 12 . . . .i.d7, 193 suggested by Chekhover, Black could have quickly equalised. bishop has moved from c 1 and the d l square is defended by the rook). 15 16 17 .i.xd5 l:Cxd5 tl\xg5 lDxd5 .i.xg5 h6 The only defence against the threat of 1 8 'i'd3. 18 19 11d2 l:bd7 hxg5 11f6 20 21 22 23 l:l.xb7 1la5 1i'b4 :ad8 :d6 %lfd8 l:td4 24 25 26 'l'b3 ifc2 1fc7 1 9 . . . 'ilib6 would not have saved the pawn on account of 20 a5 1i'b3 (20 . ifbs 2 1 I;ld5) 2 1 :l.a3, when Black has merely lost time. . . 12 ... .i.d7 :n The simplest way of defending is to attack the white queen. This leads in surprising fashion to the loss of a pawn. 1 2 . . . t'hfdS was perhaps stronger� as Petrosian played against Reshevsky in 1 970 (the continuation was 1 3 tDxdS lDxd5 14 jLxe7 t'iJxe7 1 5 "i'e4 tDds 1 6 tDes t'fil6 17 Wf4 with equal chances). I need hardly tell the reader that this entire variation and the following con tinuation were prepared by me before the game. Nevertheless, I made my next move not without some hesitation, since I foresaw that the conversion of the extra pawn would entail great difficulties. 13 d5 exd5 14 tDxd5 lDbxd5 In this way White removes his queen from further attacks, but Black's pieces are active and well coordinated. This and the following exchanges are forced. If instead 14 . . . ti:)fxdS 15 jLxe7 t'iJxe7 16 lDe5 ltJbdS 17 .itxd5 lDxd5 1 8 :xd5, then Black does not have Chekhover's saving manoeuvre 18 . . . .ilg4 19 11'c4 1ixd5 20 'i'xd5 l:tad8 (in contrast to the afore-mentioned Game 44� the white Black cannot implement his threat of 26 .. J�xfl. ( 2 7 .Uxf2 :tcil+) because of 27 'iixd8+. Therefore he offers to go into a 194 double rook ending, where he has grounds for hoping �or a draw. 26 27 28 29 11'xf4 h4 :b4 30 31 32 33 gxf3 b3 'it>g2 :.b8+ 1!ff4 gxf4 lt.c8 tJ So, throughout the game White has maintained the advantage that he gained in the opening, but a draw would appear to be inevitable. 39 Black temporarily sacrifices a second pawn, to break up White' s pawn chain on the kingside and to hinder his defence of f2 . licc2 lib2 .U.d3 39 40 41 42 43 All the same the two extra pawns cannot be held, and White makes an attempt to activate his second rook. 'it>h7 f6 ltb7 :dxb3 11.el llxb3 After 36 lia7 l!b4 37 l:ee7 l:lxa4 38 33 34 35 36 lle6 l:lxa6 �g2 l:ta8 g5! �g6 :r4 :a4 a5 a6 Draw agreed Game 3 12 Petrosian-Botvinnik l:.xg7+ 'it>h6 Black would also have retained every chance of securing a draw. 36 37 38 'it>g3 By 39 �l White could have tried to activate his king, without prematurely advancing his a-pawn. Now, however, moving the king into the centre will involve the loss of the f3 pawn. Alas, this was the penultimate move in time trouble! World Championship Match Moscow 1963, 13th game Queen' s Indian Defence �xb3 l:lb4 l:lxh4 1 2 3 4 d4 c4 lDtJ �f6 e6 b6 gJ In the following game (the 17th) where the Queen' s Indian Defence occurred, Petrosian avoided the fianchetto of his king's bishop. '-.i.b7 i.g2 /Le7 0-0 0-0 lhe4 thc3 I have already written in my not'es to 4 5 6 7 • • • Game 1 56, where I played 7 . . . d5 (as also in Game 45), that when I was not 1 95 obliged to play for a win with· Black, I was more inclined to continue 7 . . . ttJe4 . That was the case in my game with Alekhine (No.96), who chose 8 'i'c2. 8 9 �xe4 d5 • . • .i.f6 9 exd5 10 cxd5 .i.f6 1 1 ltle l i.xg2 1 2 ltlxg2 c5 1 3 lLlr4 d6 is also not bad for Black. . . . 10 lDel No better was 10 lDd4 .i.xg2 1 1 �xg2 'ife7 1 2 ltlc2 d6. 10 • . • ll.. xg2 exd5 1 1 �xg2 This exchange could . also have been delayed. After 1 1 . . .:es 1 2 lDr4 a5 1 3 e4 ctJa6 1 4 i:te 1 exd5 1 5 ctJxd5 ltlc5 16 lic2 c6 in the game Lilienthal Taimanov ( 1 954) Black maintained the balance. 12 cxd5 prospects 13 dxc6 14 1!fc2 15 l:t.dl on the dxc6 ilxe4 White played more consistently in the 3rd game of the match - .9 ttJe l , but 9 .i.f4 is also possible. 9 gammg clear queenside. c5 Forcing �te to take this pawn en passant, as otherWise Black continues . . . d7-d6, . . . ltld7 followed by . . . b5-b5, If instead 14 'i'xd8 :xd8, then Black, in contrast to White, easily mobilises his forces. c5 14 • • • Possibly 15 'i'e4 would have been more far-sighted, practically forcing Black to reply 1 5 . . . ctJa6 (or 1 5 . . . ltld7). Then his knight would have been unable to reach c6 so easily. 1'e8 15 .. . 16 i..f4 White foresees that he will have to play e2-e3 to defend the central d4 square, and so he develops his queen' s bishop i n advance. tbc6 16 17 .i.d6 Jl.e7 The tempting 17 . . . lDd4 did not achieve anything because of 1 8 1hd4 (less good is 1 8 Wc4. i..e7) 1 8 . . . it.xd4 1 9 lDf4 followed. by·.i.xf8. . 18 • e3 . · A forced move, although it weakens 1 96 immediately develops a dangerous attack, . and he is always assured of a draw. Therefore Petrosian tries to block the long diagonal. the light squares (in particular f3). Generally speaking, White has not de rived anything from his opening idea of 9 d5, and in addition Black's queenside pawn majority is a significant factor. 18 19 20 21 llxe7 lDf4 ila4 After 2 1 CDd5 :xd5 22 J:lxd5 �b4 23 'i'd2 ltlxd5 24 'i'xd5 .:.ds Black would have seized control of the d-file, but it would have been simpler for White to begin exchanging rooks immediately. 22 :xd8+ 1fxb5 �g2 'it>gl 'if xd8 30 31 <it>gl ild5 'iixe4 · Vdt+ it'f3+ 1fdl+ l:bd8 If 22 'i'e4 f5 23 'i'g2, then 23 . . .c4! (securing a favourable post for the knight at d3) with the threat of . . . g6-g5g4. After the immediate 23 . .. g5 24 tt:Jd5 'iff7 25 f4 White achieves a good game, since he drives back the enemy knight. 22 23 24 25 . 26 27 28 This check and the following one are to gain ti.me on the clock. 29 <itg2 1ff3+ li)eS 21 b5! 24 Black nevertheless gains control of the a8-hl diagonal, but the exchange of his b-pawn for the e-pawn reduces the sharpness of the play in the coming ending. :ds 1ixe7 g6 %id1 e4 '1xd8 c4 It was dangerous for White to accept the pawn sacrifice - 24 l!xd8+ 'i'xd8 25 'i'xa7, since after - 25 . . . g5 Black 11xd5 31 3 1 .. . 'ib l + 32 <iftg2 'i'xb2 33 'i'a8+ '3;g7 34 'i'xa7 would have left Black with only very problematic winning chances, however he · continued 34 . . . 'lb5 or 34 ... h6. Therefore-he trans poses into a _kni�t ending, counting on the greater activity of:bis king. 197 32 tllxdS · · tlld� 33 34 35 b4 axbJ b4 cxb3 f5 Since his king catulot go to e2 (because of . .. lDc 1 +), White seeks (and finds) a way to draw by using his queenside pawn. 35 36 • • • b5 <Ji;f7 ltle5 b6 �1 b7 ltlc7+ ltla6 'l&>e2 lhb4 43 44 45 46 �dJ 'l&>c2 <i>bJ a4 a3 The saving move. ltlb8 'l&>c7 'l&>xb7 <it>b6 Black tries i�ediately to penetrate with his king into the white pawns. The preparatory 46 . . . lDc6 would also not have won. If 36 . . . 4itte6, then 3 7 b6 ! 37 38 39 40 41 42 43 a5 <it>e6 ltlc6 When I began my adjournment analysis. I immediately discovered that 40 lDb6 ! would have quickly led to a draw: 40 . . . 'it>e7 (40 . . . <it>d6 4 1 b8'i'+ lDxb8 42 lDc4+) 4 1 @e2 <it>d8 42 llld5 etc. However, this inaccuracy on the last move before the time control does not have serious consequences. The impression is that White is losing, since after the elimination of the a-pawn his king will be too far from the kingside. In fact, it is able to return just in time. 47 48 49 50 51 <i>xa3 'l&>b3 <i>c2 <i>d2 <i>el �c5 'l&>d4 'l&>e4 @f3 White succeeds in defending every thing. As often happens in a chess game, one tempo is lacking either for a win, or for a draw. 51 52 53 54 <i>g2 h4 <i>fJ tl:\dJ ltld7 <t>t'l Draw agreed An interesting game. Game 3 13 Botvinnik-Petrosian World Championship Match Moscow 1963, 1 4th game Queen's Gambit 1 2 3 40 d4 c4 lhcJ d5 e6 , JJ..e7 This move was mentioned back in the notes to Game 59. The point of it is that Black tries to provoke tiJf3 before �d6 1 98 d4 square ( 1 0 dxc5) and allow the opening of lines. the development of tlJ.e bishop at g5. As this and other games from the match showed, this continuation is by no means better than the usual 3 . . . tt:Jf6. 4 s 6 cxd5 Ar4 e3 7 g4 10 • • • lDc6 exd5 c6 .i.f5 Essential. 6 . . . tt:Jf6 7 i.d3 leads to the Exchange Variation of the Queen's Gambit with the bishop at f4 instead of g5, which is to White' s advantage. This idea was first employed by me in that Game 59. Then, in almost the same position (instead of . . . i.e7, Black had played . . . tt:Jd7) I played it in 1952 against Bronstein (see training games) . 7 • • • 1 1 @fl When the rook has to remain on the h-file, and queenside castling -is ruled out, the king has to determine its own fate. .te6 The point is that after 7 . . . i.g6 8 h4 i.xh4 9 'i'b3 b6 10 tiJf3 1 8 i.e7 1 1 ctJe5 White has a clear advantage. 8 h3 In the preceding even-numbered game (the 12th), I played 8 i.d3 . and 8 h3 is a step forward. but probably the most energetic is 8 h4, which I chose in Leiden ( 1 970) against Spassky (No. 379). It is interesting that, 15 years after the tournament in Leiden, the plan with 8 h4 was used by Kasparov in the 2 1 st game of his second match with Karpov. 8 9 • • . i£if6 ..tdJ The following counter by Black in the centre would have been less favour able for him after 9 ttJ:f3, which I em ployed in the 18th game of the match. 9 10 • • • lDtJ c5 The position of White's king is not so secure, that he can give up the central 11 12 . . . @g2 0--0 cxd4 Now White is able to defend his out post in the centre, and in this connection Romanovsky with good reason recom mended 1 2 . . . :es, with the idea of transferring the knight from f6 to :f8. 1 2 ... l:.c8 has also been played. 13 tt1xd4 Since White' s king is now secure, he need not concern himself that Black will avoid the exchange of knights and try to exploit the absence of the white pawn from d4. And the exchange of knights is to White's advantage: he inunediately gains the possibility of gaining control of e4 by the advance of his f-pawn, when he will dominate in the centre. 13 1 99 l£lxd4 Four years later a game Geller Spassky continued 1 3 . .i.d6 14 i.. xd6 'ifxd6 1 5 tbce2 .:res 16 .&te l .i.d7 17 i..b 1 , to White' s advantage. 18 .. 14 exd4 ltld7 15 ilc2 lDf6 If Black had made use of this opportunity to fight for the initiative 14 . . . ctJe4 15 ltJxe4 dxe4 16 .i.xe4 f5 , then after 17 gxf5 ..txf5 18 'i'd3 his compensation for the material deficit would have been insufficient. Typical of Petrosian. He is afraid of weakening his position by either . . . h7h6 or . . . g7-g6, and so he is not concerned about losing two tempi, if he considers the move to be the most sensible in the given situation. 16 f3 Now White has a clear positional advantage. It would have been rash to be tempted by 16 g5 tllli5 17 �xh7+ �h8, since there is no way of contin uing the attack, and the bishop at f4 and the pawn at g5 are threatened. to say nothing of the bishop at h7. 16 17 • • • · .*.e5 :cs .i.d6 l%ael It would again have been a mistake to go chasing a pawn ( 1 8 Jtxf6 'i'xf6 1 9 i.xh7+ 'ii?h 8), since the initiative would have passed to Black. The character of the play, arising with the given pawn formation after the exchange of the dark-square bishops, was well known to me from the game Levenfish-Kotov ( 1 1th USSR Championship, 1939), which I annotated for the tournament bulletin. In the endgame White will have the advantage - his minor pieces are more active than the opponent' s. 18 19 . • • .i.xe5 l:txe5 It is only possible to take with the rook, since after 1 9 dxe5 d4 20 exf6 ilxf6 Black wins back the piece, gaining a positional advantage. 19 20 21 22 11f2 . %te2 libel g6 lDd7 �b6 The double-edged 22 h4 would not have been in the spirit of the position, since White' s attacking prospects would be unclear, whereas an endgame will always be favourable for him. 22 23 • • • lLlc4 i.xc4 Now there is exactly the same material remaining on the board, as in the afore-mentioned Levenfish-Kotov game. In the event of 23 . . . dxc4 24 d5 i.d7 Black loses a pawn, so he has to take on c4 with his rook, but then the d5 pawn will restrict the e;ictivity of his bishop. 23 24 200 l:r.xc4 l1d2 lite8 25 26 1!e3 b3 a6 At ..any moment Black could have increased the pressure on the d4 pawn, so the rook must be driven away. 26 27 28 29 ltla4 itlb2 itld3 l:tc6 b6 a5 · 35 itld3 35 36 37 gxf6 1t'g5 This, of course, is to gain time on the clock. After the knight manoeuvre has forced a slight weakening of the enemy pawns, White returns it to the centre. 29 which is not now possible on account of 34 . . . .i.f5 35 gxf6 l:xf6 36 'l'e5 J:d6 37 �e2, and in the end Black loses his d5 pawn. 34 i..f7 . f6 Petrosian underestimates the dangers arising in the endgame from the fact that a white pawn will be at g5. It was completely bad to accept the pawn sacrifice: 37 . . . 'ii'xd4 38 ti.)eS . However, Black could have defended tenaciously with 3 7 . . . rl;g7 38 ttJf4 rJm, · although after 3 9 ri;g3 he has no useful moves. 38 30 .i.e6 'I' xf6 'l'xg5+ hxg5 h4 White prepares g4-g5. to weaken the dark squares in the opponent 's position. 30 31 J..f7 l:txe8+ White is always ready to exchange one pair of rooks. Black's counter chances are reduced, but Iris weaknesses remain. Jl 32 33 34 'l'e3 g5 lbf4 .ixe8 Ji.f7 �e6 White prevents the opponent's intention of playing his bishop to f5 , 20 1 38 a4 Black grows nervous, and makes things easier for White-. The exchange of a pair of pawns does not bring any relief, whereas removing control from the c5 square is irrational. Here it would have been more advisable to stick to waiting tactics. 39 bxa4 After 3 9 ltJe5 :c3 40 bxa4 l:r.a3 4 1 llb2 :xa4 4 2 l:txb6 l:txa2+ all the queenside pawns are exchanged, which eases Black's defence. 39 40 41 42 43 44 a5 li)c5 Clti>gJ <i>f4 <i>e5 :lc4 bxa5 .t.f5 a4 a3 . . • llxd3 Axa3 <i>xd5 <i>c6 ltb2 � llg2 l:lxg5+ h5 52 53 54 d5 d6 <it>d7 llg2 l:lc2+ After the forced exchanges, which have led to White winning a pawn, Black's only hopes, naturally, rest on the advance of his passed pawn. White's plan, begun with 3 3 g5, has been logically completed. His king has penetrated into the black position. 44 48 49 50 51 :b4 Or 44 . . . :c2 45 .:txc2 .i.xc2 46 ttJa6 3'.b3 47 lDb4 .i.c4 48 �c2, and White remains a pawn up. 54 45 lhd3 45 46 47 <i>d6 ci>c6 Of course, with such an active king White can go into the rook ending, and the enemy rook must not be allowed to go to b2. If 47 . . . :as, then immediately decisive. l:lb5 <iJf7 .i.xd3 48 lDc5 • • h4 • A mistake in Black' s adjournment analysis, which leads ' to defeat. He should have tried 54 . . . g5 55 l::ta5 <itn'6 56 �d8 h4 57 d7 h3 , although after 58 :a6+ r:l;g7 59 l:te6 19 h2 60 l:.e l ci>f7 6 1 a4 White should nevertheless win. .:n 55 f4 Here there was no longer any point in playing 55 . . . g5, in view of 56 fxg5 <li>g6 57 r:l;d8 r:l;xg5 58 d7 'ittg4 59 :as h3 60 'itte7 etc. is 202 56 <i>cs 57 l:ta7+ Black resigns - promotes to a queen. :xr4 the white pawn tl1e bl-h7 diagonal. This, was conf inned, in particular, by the subsequent games Petrosian-Spassky ( 1 97 1 ), which went 1 2 'i'e4 li.Jcb4 1 3 lDe5 :a7, and Kuzmin-Suetin ( 1 976) - 1 2 h4 CDcb4 1 3 h5 h6. Game 3 14 Botvinnik-Petrosian . World Championship Match Moscow 1963, 16th game Queen's Gambit Accepted 1 2 d4 c4 d5 dxc4 After his failure in the 14th game, Petrosian reverts to accepting the Queen's Gambit instead of declining it. 3 4 5 6 7 lt\f3 eJ -*.xc4 0-0 a4 lt\f6 e6 c5 a6 Exactly the same as in the 2n4 6th. 8th (No.3 10) and 10th (No.3 1 1) games of the match. 7 8 . 1ie2 • . lLlc6 cxd4 12 13 8 . .. :il.e7 successfully passed the test in the 6th game, but the opening of the 8th game went badly for Black. 9 l:[dl ile7 exd4 lDc3 0-0 11 ... lDd5 14 15 16 17 18 19 An improvement compared with l l . . .tiJb4, which occurred in the 10th game. Now White can no longer aim for d4-d5, and his subsequent plans involve an attack on tl1e kingside, exploiting tl1e absence of the knight from f6. .i.d3 White considers that after . . . CDcb4 he will be unable to tr.ansfer his bishop to ... g6 Also after 14 . . . tDf6 1 5 ifh4 Black would soon have been forced to weaken the pawns covering his king. And here the similarity ends with the 2nd game, where 1 1 i.g5 was played. 12 lhcb4 .*.d7 13 . . . qjf6, which occurred in the game Gligoric-Korchnoi ( 1 965), also led t9 some advantage for White. 14 1We4 This is perhaps stronger than 14 ltJe5. 9 . .. 'Wic7 occurred in Games 1 16 and 161. 10 11 ... .i.bl lhe5 1ff3 1ig3 h4 ltJf3! i..f6 li.g7 .1\e8 ti)c6 White must avoid exchanges, in order to maintain the tension. 19 • .. f6 To take control of the g5 square, Black is forced to weaken his e6 and g6 pawns. 203 30 ltJe5 lbxe5 31 32 33 dxe5 f3 :et l:le8 :e2 lla2 34 Ji.gs Now White acquires a dangerous passed pawn, but if 30 . . . 'tfe6 there could follow 3 1 lllxc6 'i'xc6 32 f3 :e2 33 %itbc 1 with a d�gerous initiative. 20 lbxd5 22 hxg6 22 23 24 25 tl:ih4 11.xbl It is advantageous for Black to keep his rook on the second rank, but without the support of the other pieces he is unable to create any threats to the white king, and at a2 the rook is as though out of play. As will be seen from what follows, Black's position is critical. However, on this occasion too, as in other games from the match, I played uncertainly and squandered my advantage. exd5 If 20 . . . 'i'xd5 the reply 2 1 ..ta2 is unpleasant, but now Black is deprived of his strongpoint at d5, and White's positional advantage becomes obvious. 21 h5 /1)e7 If 2 1 . . . g5 there could have followed 22 ti:Jxg5 £xg5 23 'i'd3 . This exchange eases Black' s defence. and therefore it should not have been hwried. but White was ab-eady thinking about saving time before the control. il.xg6 li.xbl 't!f d7 l:tfi b3 25 . . . lllf5 could not be played immediately because of 26 'iig4, but now Black has prepared this move, and it has to be parried. f5 26 irfJ 27 28 flgJ lbf3 tl:ic6 29 ..fl.f4 lte4 �e7 Black exploits his strong trump, directing his rook to e4, where it will be very well placed. 34 :tbc 1 , activating the rook, suggests itself, e.g. 34 . . . .&te6 35 il.g5 (with the tlrreat of 36 .lti6) 35 . . . l:tg6, and White decides matters with the breakthrough 36 e6 l:lxe6 37 .i.f6 l:Xf6 (37 . . .ktxel+ 38 :xel) 38 :c7 f4 (or 38 . . . l:g6 39 irxg6 'i'xc7 40 l:te8 mate) 3 9 :es+ :tf8 40 :xf8+ 'ifilxf8 4 1 204 Vxf4+. In all .these variations what tells . is the absence of the rook at a2 from the main baJ:tlefield. 34 35 3 5 :bd 1 played. 35 36 e8'i' :xg2+ with perpetual check), but I still had to make my last move before tl1e time control. 40 41 d4 .*.f6 Played through inertia. After 4 1 e7 Black still has nothing better than perpetual check. Now, however, White will have some work to do . . . should first have been ktbdl iff7 ilg6 41 42 43 44 Forced: Black can no longer save his d4 pawn and he merely aims to safe guard his king. 37 38 1fxg6 llxg7 39 e6 1:cc2 ltgl hxg6 @xg7 :del e7+ 'it>g3 l:.d2 <t>f8 'it>e8 d3 The storm clouds are gathering over White's position, but in adjournment analysis I managed to find the only saving move. It is hard to explain why I avoided the natural 39 :xd4 lilc8 40 @h2 11cc2 4 1 _:tgL when White is a pawn up with winning chances. Apparently in time trouble the advance of the passed pawn seemed to me to be a safe antidote against the threat of . . . r!c8-c2. In fact Black is handed an extra tempo for this manoeuvre. 39 . fies If 47 :b l Black could now have replied 47 . . . llh2. Now I should have reconciled myself to the inevitable dratV (40 e7 lkc2 4 1 Of course, not 47 .. ;d2 48 :b l . In the event of 47 . . . l:xb3 48 11exd3 J:xd3 49 40 'it>h2 45 . . l:.e3 . 46 'it>f4 :ab2 The whole point is that the threat of 4 7 l:.h I has been created. 46 47 205 47 • • • l:t dl :xg2 l:lbd2 Ilxd3 a outcome �6 5 1 <JJxe7 5 1 draw is the most probable after both 49 . . . <itiixe7 50 ltb3 �b6+, and 49 . . Jitb2 50 :d6 llxg6. 48 49 l:txd2 �g5 50 <ifi>t'6 %lxd2 1%dl Black could even have 49 .. J:tg2+ 50 lt>f6 d2 5 1 ltd3. 10 ird2 11 i.h6 li:)bd7 lost: But not 50 <it>xg6 in view of 50 . . . l:tgl + 5 1 �6 d2 52 %k3 :tg6+ ! 50 51 52 53 54 f4 lle4 l:lct lld4 l1c6+ :cJ ci>g5 'it>f6 Draw agreed A step forward compared with Game 298, where after 1 1 0-0--0 Black retained the option of castling kingside. Now this would be dangerous. White achieves nothing with 1 1 a4 bxa4 (Ostojic-Ivanovic, 1 972). Game 3 15 Krutikhin-Botvinnik USSR Spartakiad, .Moscolt1 1963 Sicilian Defence 1 e4 2 lt)f3 3 d4 4 ltlxd4 5 ltlc3 6 .i.e3 f3 .i.c4 .i.b3 .i.g7 a6 b5 i.b7 .. 'l'xJt6 13 0-0-0 13 14 ... cxb3 .i.xh6 ltlc5 . That same year there were also other attempts to justify Black's set-up, but none were successful. For example, 1 2 . . . b4 1 3 ti)a4 'i'a5 14 'ifd2 0-0 1 5 a3 'i'e5 1 6 axb4 d5 1 7 0-0-0 (Mazzoni0 'Kelly). c5 d6 cxd4 ltlf6 g6 It will be remembered that this move followed by 7 f3 is the plan introduced into tournament play by Rauzer back in 1 936. 6 7 8 9 11 12 Black, for his part, sticks fully to the development plan suggested by Reshevsky. After 13 'i'g7 Black would have successfully defended with 1 3 . J:tf8. . lDxb3+ Boleslavsky 's well-known idea. The white king is more securely defended by its pawns than after 14 axb3, when Black still has chancei;. of opening the a file. In addition, the c:-ftle, which in the Sicilian Defence is usually used by Black, may here · also come in useful to 206 · White. And for the moment Black' s extra pawn in the centre i s of no significance·. b4! Preparing the manoeuvre of the knight to a5, where it will occupy an unpleasant position. Another, equally effective idea was demonstrated by Stein in a game with Veresov ( 1 963): 16 l:the 1 �b8 17 tiJd5 lDxd5 1 8 exdS. 16 17 18 lhbJ h4 <it>b8 1lhg8 21 22 g5 23 lhd5! l:lf8 lhd7 l:c8 lha5 1i'f4 lk7 %1.gc8 A loss of a tempo. 20 . . . �8 should have been played. White energetically conducts the offensive: after the forced exchange on d5 Black will have considerable problems due to the weakness of his c6 square. 23 24 25 26 exd5 'ife4 gxf6 .lxd5 lhe5 f5 Otl1erwise there would have followed 26 . f4, when the position of the knight at e5 would be impregnable. . . Preventing the active 18 . . . g5 fol lowed by .. J�g6 and . . . g5-g4. At the same time White himself begins a pawn offensive on the kingside. 18 19 20 g4 1i'b6 14 Or 14 . . . e5 1 5 ttJc2 'fl.e7 16 lDb4 (or 16 lDe3), and White's position is preferable. Black has no choice, other than to castle queenside. 0-0-0 15 'ifi>bl 16 21 Now, in view of the unequivocal threat of 22 g5, the rook is forced to return to the kingside to defend the f7 pawn. 26 27 11.hfl laxf6 Up to here my opponent had played excellently, but now he makes an error. After 27 f4 tLlg4 28 l:the l � 29 'i'd4 Black would have had to contend with serious difficulties, since his knight would not longer have been defending 207 the key e5 square. Now, however, I am saved by a chance tactical opportunity. c;t>as 27 . · 2s ; - net if White is agreeable to allowing the Pirc defence, he should not p)ay his bishop to c4� 4 ltic3 is better. .- 4 5 • • • 'lfe2 lDf6" Or 5 e5 dxe5 6 ttJxe5 · 0-0 and Black carries out the freeing advance . . . c7-c5. 5 • • • c6 In order after e4-e5 to have the reply . . . lD<l5. In some · cases . . . li:)xe4 is also possible ('ii xe4 d6-d5 ). 0-0 6 il.b3 7 0-0 7 i.g5 is perhaps more energetic, as Matulovic played against me in 1 970 (No.378). 7 28 • • • • • • a5 lDxf3! This is the whole point. The threat of 29 . . .ltJd2+ prevents White from first exchanging on c7. All that remains is that which occurred in the game . . . 29 lbfJ Draw agreed, in view of the variation 29 . . . J::txc l + 30 �xcl 'i'g l + 3 1 <itd2 'i'g2+ 3 2 · �e3 'i'g l+ 3 3 @d3 'i'd l + 34 �e3 'ifgl + with perpetual check. GaIUe 3 16 USSR Spartakiad, Moscow 1963 Pirc-Ufimtsev Defence 1 2 3 4 e4 d4 ttJfJ ii.c4 8 a4 8 9 ... lDbd2 A committing move. 8 a3 or 8 c3 was more cautious. The Encyclopaedia re commends first playing 8 e5 liJd5 and only then 9 a4. Gipslis-Botvinnik g6 .i.g7 d6 Black has· . greater difficulties in the King's Indian Defence (after 4 c4). But il.g4 d5 Temporarily shutting the b3 bishop out of the game. The e4-e5. advance is now no Jonger dangerous for Black: his queen's bishop has c...�e into play. 208 complications in the centre, for which he is prepared to give up a pawn. 10 e5 In principl� 10 c3 should have been preferred, although even then it is doubtfui whether White can achieve any advantage. 10 11 h3 14 15 16 t£lfd7 lDxfJ .i.g5 ilxf3 e6 13 'ifb6 prise. Trusting. in the power of his two bishops, Gipslis decides to initiate 16 17 .i.e7 17 18 19 20 i.d6 l:tb3 t£lg5 cxd4 Of course, not 17 l:b3 because of 17 . . . 1i'c5 . An energetic, but not altogether suc cessful move. White should have taken prophylactic measures against an attack on his centre by . . . c6-c5 . Six years later Matanovic tried to improve White's play with 1 3 c4 (No.3 74), but also with out success. To me 1 3 c3 seems to be the most sensible move. It transpires that, due to the need to defend the b3 bishop, 14 c3 is not pos sible, and the bishop, in tum, cannot move because the b2 pawn is left en c5 dxc4 · White's initiative can:hot last for long, since Black completes his devel opment without any interference and then sets his sights on the weak e5 pawn. . 1 1 e6 i.xe6 12 ltJg5 did not work because of 1 2 . . . i.xd4. 11 12 13 l:.aJ c4 .i.xc4 l:tc8 �c6 �b4 One gains the impression that White has nevertheless gained some compen sation for the pawn. In particular, he is threatening to play 2 1 l'Dxe6, and 21 :D is also rather unpleasant. 20 Jlxc4! However, this exchange sacrifice enables Black to parry the imminent tlrreats and to seize the key squares in the centre. 209 21 22 1'xc4 .i.xe5 ltlxe5 White resigns If 3 3 Itd6 or 3 3 %lb6, then 3 3 . �. 'i'f3 followed by . . . !fil4 is decisive. Gipslis has not yet lost hope in his attacking p9ssi.b\liti�s, and he avoids the � . �· exchange'fif queens· (22 'i'c5), assuming that it1 the . endgame he would only be able to battle for a draw. But perhaps this would have been the lesser evil. .... 22 23 24 25 ' . . ·. �·�·. . l:lf3 bidl ltle4 Gruiie .. :fl? Taimanov-Botvinnik USSR Spartakiad, Moscow 1963 .ilxe5 l:f8 i.g7 e5 Nimzo-Indian Defence 1 2 3 d4 c4 ltlcJ ltlf6 e6 i.b4 It may seem thoughtless to choose this opening against Taimanov, one of the connoisseurs of the Nimzo-Indian Defence. But, on the other hand, if something needs to be checked, the examiner should be familiar with all the details of the subject. 4 5 6 7 26 net 'ifc7 ltlf6+ ltxf6 lic4 h6 'ifd8 ild5 .i.xf6 ltld3 'ife4 There is no satisfactory against the threat of 32 . . . !Df4. 32 1ic2 c5 b6 .ib7 7 .i.d2, as played in Game 3 56, is weaker. g4 After this serious weakening of the kingside it is Black who will be attacking. True, also after 26 :c 1 'i'c6 27 'i'xc6 bxc6 28 tllf6+ .i.xf6 29 .ttxf6 l:td8 the endgame would have been hopeless for White. 26 21 28 29 30 31 e3 ltlf3 i.d3 0-0 rl;g7 defence 7 8 .. l2Ja4 . 0-0 'ife7 Here 8 . . . cxd4 is considered obliga tory, to secure the retreat of the bishop to e7. From this standpoint 8 . . . 'i'e7 may have claims to originality. Even so, it has to be admitted that in some games of Keres a similar idea had already occurred - at a5 the bishop will not be so badly placed as it appears. I should say that I prepared 8 . . . 'i'e7 for my match with Petrosian, but when in the World Championship Match I was unable to employ this weapon, I was very happy to play it against an 2 10 acknowledged expert on the Nimzo Indian Defence. Initially Taimanov was on his guard, but even so he quite quickly ..made the f9llowing moves� il.a5 . 9 a3 · · not understand his play so badly . . . ' 'What do you mean', replied my oppon ent to Petrosian, 'here there was the unpleasant interposition. 12 . . .ixf3 . Therefore the reader must agree that Mark did indeed do well ! · As regards other possibilities for White, later tournament games and researches by analysts (in particular Taimanov) showed that IO ktb 1 is more prom1smg for White, after which, according to the Encyclopaedia, he gains the advantage. It follows that 8 . . 'i'e7 is less strong than the well tested 8 . . . cxd4. Now (after 10 b3 ) a double-edged situation has arisen: both the knight at a4, and the bishop at a5, are on the edge of the board. Of course, the bishop is more actively placed, but it altogether has no moves, whereas White can always manoeuvre with his knight, and in addition there is the unpleasant threat of b3-b4 . Whether or not Black is able to stay afloat will only later become clear ! . ' . 10 bJ Well done, Mark! Taking account of Petrosian's inclination for exchanges in general, and in particular for those that gave him the advantage of the two bishops, in the diagram position I had prepared for the variation 10 dxc5 bxc5 1 1 l2Jxc5 'i'xc5 12 b4. In this case there would have followed 1 2 . . . i.xf3 ! 1 3 'i' xf'J ife5 14 bxa5 ilxal 1 5 'ii'�a8 'i'c3 16 i.e2 l2Jc6 with the initiative for Black. Naturally, I was intrigued: had my opponent seen this variation, or had he unsuspectingly avoided the danger? After the game we began analysing. I kept silent, awaiting a comment by my opponent. At this point Petrosian came . up to our board, and immediately every thing became clear. ' Why did you avoid the variation 10 dxc5 bxc5 1 1 liJxc5 ?', the World Champion asked Taimanov. · Aha', I thought, 'this means that I did 211 10 11 12 13 14 cxd5 .i.b2 1fe2 llacl d5 exd5 lDbd7 lDe4 Jlc6! The only possibility. Now 1 5 .i.b5 .i.xb5 16 'i'xb5 a6 17 'i'e2 c4 ( 1 8 b4 b5) looks dubious for White. Taimanov agrees to exchange his knight at a4 for the bishop at a5, after which White's position is slightly better, but the psy chological battle . has nevertheless been won by Black, since White has not in fact been able to exploit the unfortunate position of the bishop at a5 . 15 lDc3 .t 6 .*.xc3 17 . .i.b2 i.xc3 a5 l:tfc8 24 lDg4 c4 .: After the incorrect 1 7 . . . a4 18 dxc5 ! bxc5 1 9 b4 Black would have ended up in a critical position. 18 lDe5 Excellently played. If now 18 . . . ltJxe5 19 dxe5, then the bishop at b2 is activ ated, there follows f2-f.3 , driving ·. the black knight to g5, and White's initiative on the kingside becomes dangerous. 18 19 . • . l:tfdl .i.b7! lhdf6! was able to achieve the same arrangement of my knights in Game 305. 20 f3 There was no need for this weakening of the position. 20 21 22 23 a4 lic2 :dcl lDd6 lk7 lDfe8 On this occasion Taimanov did not guess my intention. 23 • • • Although the c4 square is attacked five times by White' s pieces and only three times by Black's, this move proves possible thanks to · the variation 25 bxc4 dxc4 26 i..xc4+ .:xc4 ! 27 llxc4 i.a6. Even so, I decided on 24 . . . c4 only after great hesitation: after 25;bxc4 dxc4 26 ltxc4 ! ltJxc4 27 :xc4 White would have had definite compensation for the exchange. However, my opponent was so dispirited by what had happened, that he thought for a long time and then soon made a totally bad move. 25 26 bxc4 Jle4 26 27 28 fxe4 l:xc4 28 29 30 ifxc4+ 1!fe2 After this hopeless. dxc4 White's position is li)xe4 1ixe4 As a result of this White loses a piece. 28 h3 would have saved the knight, but, of course; .not the game. f6 2 12 llxc4 .i.d5 Game 3 1 8 Kholmov-Botvinnik USSR Spartakiad, Moscow 1963 Pirc-Ufimtsev Defence 1 2 3 4 e4 d4 .ig5 5 .i.f4 ttlcJ d6 ttlf6 g6 h6 This . move can be made without fearing 5 iLx:f6, since after 5 . . . exf6 Black will follow up . . . f6-f5 with a reasonable position. 30 31 32 33 dxe5 hJ b5 fxe5 b4 1fxa4 .i.d4 1fh5 ltfl lr.f4 1ff3 'i!fd7 b5 l:i)c7 i.f7 i.d5 ttle5 Now an endgame would probably be won for Black even without the extra piece - after all, he has two connected passed pawns. 34 35 36 37 38 In order to 'persuade' the opponent to resign, in such a position it is sufficient to reach the time control. 39 40 41 1ie8 'l'b5 ttle6 1fxh4 'ifg6 :rs White resigns The Encyclopaedia considers that 5 .i.e3 is more promising, and that the game continuation, as well as 5 .i.h4 (as Unzicker played against me in the 1 956 Olympiad in Moscow), does not give White any advantage. 5 6 ... h3 .i.g7 After 6 'i'd2 g5 7 .i.e3 ttJg4 or 7 .*.g3 CDhS White's queen's bishop would have been exchanged, but perhaps this would not have been so advantageous to Black. Now, on the other hand, he is able to strike on the flank. 6 7 • • • c5 dxc5 7 d5 leads to more complicated play. One of the few games where I went in for a risky continuation in the opening. I was curious to see how my opponent would r�act to my cunning preparation. My reckoning . proved accurate; evidently I was an experienced and quite good chess psychologist. 213 7 8 • • • 1fa5 9d2 Panying the threat of 8 . ttJxe4. .. 8 9 10 .i.eJ .i.dJ 1fxc5 1Fa5 ttlc6 Black successfully completes the development of his pieces, but the prob lem with his position is that kingside castling is ruled out. not take any great effort by him to gain a draw in the resulting level endgam�. 11 12 ti)ge2 0-0 ttld7 ti)de5 By exchanging the opposing light square bishop� Black eases his defence. 13 f4 The attack on the h6 pawn can be lifted, since after 13 . . . tiJxd3 14 cxd3 0---0 the move 15 f5 would be highly un pleasant for Black. 13 14 15 16 ti)xd3 e6 .i.d7 cxdJ llael a3 White could have tried to complicate the play by 16 f5, e.g. 16 . . . exf5 17 exf5 .i.xf5 1 8 lDd4, or 16 . 0--0--0 17 fxg6 fxg6 1 8 �f7 . With his next move Black excludes this possibility - true, at the cost of exchanging his important bishop. . 16 17 18 19 i..d4 tLlxd4 '!'f2 . (jje7 .i.xd4+ ifb6 0-0-0 It can be considered that Black has successfully overcome his opening difficulties. In view of this, with his next move White offers . the exchange of queens, evidently assuming that it will 20 21 22 (jjf3 <&ti>xf2 l:cl 23 d4 11xf2+ @b8 f6! Black' s minimal advantage is due to the presence of his good bishop. Now he wants to seize the initiative with . . . g6g5. Even so, in such a position it would be hard to imagine that any winning chances can be found, were it not for the classic examples of this in the games of Lasker, Capablanca and Rubinstein. An incautious move, as a result of which the e4 square (and later f5) are weakened. It is true that for the moment White prevents 23 . . . g5 in view of 24 fxg5 hxg5 25 e5, but with his next move Black renews this threat. 23 24 • • • l:r.hf8 h4 One gains the impression that as yet my opponent is not at all concerned, otherwise he would have refrained from this new weakening. . 2 14 24 . • • • d5 Irrespective of White's reply, the f5 34 35 36 37 square will now be under Black' s control. Kholmov decides to exchange pawns in the centre, relying on the subsequent simplification on the open e file, but in the process the activity of the black bishop increases. �gt lDxgl tDtJ lt:\el! The only defence position. 37 38 lt:\g2 llxgl .i.g4 (f)e7 in a difficult lt:\f5 h5 Otherwise White would have provoked further exchanges by 3 9 llle3 . 39 40 25 26 27 exd5 ltfel g3 28 29 30 tLlh2 l:.c2 :ce2 .tcs tDc6 !lff8 31 �d2 g5 As a result of a lengthy manoeuvring battle, Black has gained a slight but real plus - an outside passed pawn. But he has to reckon with the possible invasion of his rear by the white rook. In addition, White· s chances of a draw are improved by the small amount of material remaining on the board. lDg6 41 :dJ Of course, not 30 . . . ltJxd4 3 1 !te8 l:tfd7 32 :xd8 l:xd8 33 :dl , when White achieves his aim - the exchange of the central pawns. A similar pawn thrust occurred in several of Capablanca' s games. White cannot exchange on g5, and so he is forced to allow the creation of new weaknesses in his position. lDtJ gxf4 lt:\xh4 f5 exd5 %!ti i.g4 Soon, by attacking the d4 pawn, Black will finnly seize the initiative. Then his bishop will be attacking the defender of this pawn - the knight at f3 . 32 33 lDeJ lt:\exd5 gxf4 .ng8 42 43 l%e3 :e6 h4 h3! 44 l:bg6 l:th8 This little combination, prepared in adjoununent analysis, involves a tempo rary piece sacrifice. which allows Black to transpose into a superior rook ending. 215 45 tLle2 Weaker is 45 l:txg4 fxg4 (but not 45 . h2 46 htgl ) 46 �g l g3. .. 45 46 47 llgl ..txe2 Jlg4 tLleJ Perhaps 47 ltlf6 was preferable. 47 48 • • • l:thl h2 hJ! It is this move, also found in adjourn ment analysis and probably not antici pated by my opponent, that gives Black new chances. In view of the threats of 49 . . . l:tf3+ and 49 . . . i.f3 the exchange of minor pieces is forced. 49 50 ltlxg4 �g2 fxg4 r!f3 The tragedy for White is that his king is a long way from the queenside, where the final battle develops. Black first of all eliminates the f4 pawn. It would have been wrong to play 50 . . . :b3 5 1 'it>xh2 l:txb2+ 52 �g3. 51 52 53 54 55 lbh2 d5 'iii>gJ l:tb7+ llh6+ lbf4 :d4 ¢Jc7 �b6 <ifi>c5 What can be said about this position? White forces the enemy king to defend its queenside pawns. Then the d- and g pawns will inevitably be exchanged, which leads to an ending with two pawns against two on the same wing. Does that mean a draw? Yes, provided only that Black does not succeed in exploiting his sole trump - the remote ness of the enemy king from the main battlefield. But how can he do this? 56 57 58 59 �b6 �b5 �a6 b6 l:th7 llb6+ llh7 l:th6+ For the moment only White has achieved anything. He has driven the opponent' s king as far away as possible from the d-pawn. Now the logical continuation, complicating the play somewhat, would have been 60 d6. Then White would possibly have been able to exchange his d-pawn for one of Black ' s queenside pawns, or for the g pawn, but with the black king on the seventh rank, further away from the white pawns. 60 llh7 But this, for the reasons indicated, is wrong. As is apparent from the course of the game, the immediate exchange of the d- and g-pawns does not complicate Black's task, but makes it easier. 60 61 62 lbd5 :d2 'itxg4 b4 If Black immediately threatens the pawn, then after 62 . l::ta2? 63 b5+ <ltxb5 64 :lxa7 a draw is inevitabl¢. Therefore the trajectory of the attack on the pawn is more complicated. . 2 16 . Grigoriev' , by cutting off the �ing along the rank - 69 . . . l:.g3 70 l:t.hl �a3. 62 l:lb2! The first subtlety: 63 b5+ is not pos sible, and the threat of 63 . . . l:tb3 forces the white rook to abandon its active position. There is no doubt that White overlooked this quiet move, when he played 60 l:th7. 63 :r.bJ In the most simple of rook endings a number of amazing secrets may be concealed! l:ta2! Game 3 19 And immediately comes the second subtlety: now White' s rook is tied to tl1e defence of the a3 pawn, and for the transference of his king to queenside tlle third rank is blocked. Black's king, by contrast is able to support tlle attack on the white pawns. 64 65 66 @14 <it>e4 llb7 <ifi>b5 <it>a4 Botvinnik-Van Scheltinga Amsterdam 1 963 English Opening 1 2 3 4 c4 ttJcJ g3 ttJfJ tiJf6 c5 e6 d5 5 cxd5 ttJxd5 6 7 i.g2 0-0 ttJc6 ttJxcJ 8 bxc3 i.e7 4 b6 5 .i.g2 .i.b7 is probably prefer able - in this case a variation similar to the Queen 's Indian Defence is reached. . . . White goes into a tlleoretically lost ending, since there is nothing else available to him. :xa3 66 67 l:txa7+ �xb4 b5 68 l:tb7 69 <ifi>d4 lla8 White resigns, since if 70 <Ji?d3 the simplest is 70 . . . 'itrb3 . But, incidentally, it was also possible to win ·a la And now Black avoids the Tarrasch Defence, which would have arisen after 5 . . . exd5 6 d4 . This exchange is insufficiently justified. 7 . . . i..e7 was more natural. 2 17 9 d4 10 e4 possibility that was available to my opponent on the next mov�. , 17 lifet · · · 0-0 White's forces are activated, and he now creates a passed pawn on the d-file. 10 11 12 d5 .i.f4 12 13 ... exd5 14 9'c2 b6 ll'la5 exd5 ..i.f6 Black prevents GtJe5-c6, since now if 14 ttJe5 there follows 14 . . . g5 . 14 'i'd3 is also good, as in a game Hort-Zweig ( 1 967) . . • . g6 An excellent move. Black's queen's bishop will take up a good position at f5, and at the same time the h7 pawn is covered. Of course, 1 4 . . . 'i!fxd5 was not possible because of 1 5 tDe 1 . 15 16 :adl 'ifcl ilf5 l!e8 1fd7 , A useful move: it will not be easy to blockade White's passed pawn. 14 17 The critical point. Black should have played 17 . . . .i.e4 ! greatly hindering White's actions: his knight at f3 would have been pinned in view of the possible exchange of the g2 bishop. Perhaps the plan of creating a passed d-pawn was not so strong after all. Now, however, when Black has lost an important tempo, White's attack becomes irresistible. 18 tl)eS 18 19 20 d6! ..td5 In the afore-mentioned game Smejkal won by 18 .l\e5 i..g7 19 .i.xg7 @xg7 20 ttJe5 'S'd6 21 'i'h6+ �g8 22 'ii'f4 [L)b7 2 3 g4 etc. 'ila4 1%ad8 .ie6 Or 20 . %lf8 2 1 Jlh6 ilg7 22 ti:Jxf7 .:Ixf7 23 ..txf7+ @xf7 24 �e7+. l:tf8 21 d7 22 :le4 Gaining a tempo t-0. switch the rook to · the kingside. . . · It was only here that for a short time the game Smejkal-Parma ( 1 973) took a different course. Black played 16 . . . 'i'd7 first, and so he did not have the 218 22 23 24 ..ih6 -*.xg7 4 5 i.g2 ll.e7 6 7 8 b3 i..b2 cxd5 b6 .ib7 lLlxd5 9 d4 c5 · 0-0 0-0 This opening, bordering on the Reti, the Catalan and the Tarrasch Defence, later became very popular. �;;:.j �� :,if? ;� '.� � � A t-:�� ... �� & �f� w.:/,.0 ::.,,.� , /f� c4 iig7 �xg7 o/�:f �';T "'! ,,,,,,, 8 . . exd5 would have led to a more complicated situation, since piece ex changes would have been less probable. . %:� . t t.� . . � � :,:.',�?1: �·· � �%.� 25 lih4 25 26 l:ldd4 Simagin played more strongly against Smyslov ( 1 966) : 9 . . . tt:'ld7. '"• t :i . . .. A subtlety that was probably under estimated by my opponent. He was apparently e:\.'Pecting the obvious 25 i..xe6, after which the f-file would be opened and Black would acquire some hopes of saving the game. <iit>g8 Removing the rook from the attack by the queen. 'if xa2 26 27 1lh6 Black resigned in view of inevitable mate. 10 dxc5! White exploits an opportunity to force his opponent to move an already developed piece. In addition. now Black will no longer be able to offer the exchange of bishops by . i.f6. .. Game 3 20 Botvinnik-Donner Amsterdam 1963 Reti 1 2 3 · c4 lLlfJ g3 Opening lLlf6 e6 d5 10 . .*.xc5 The position after White's 10th move occurred back in 1 93 2 in a game Euwe Van der Bosch. Black made the weaker reply 1 0 . . . i.f6, and after 1 1 i..xf6 'i'xf6 12 tt:'lbd2 White had an undisputed advantage. 11 219 lDbd2 lLld7 12 a3 · queens: 18 . . 'ifxf3+ 19 t'filxf3 �f8 20 ttJc6 :c7 21 �c2 l:.ac8 22 l::tac 1 with strong pressure. By now Black has realised the dangers of his _position, and with the threat of 1 9 . . . ttJe5 20 'i'e4 f5 (2 1 'i'xe5 llJf4+) he tries to initiate complications. lD5f6 . Black' s misadventures begin with this move. He was obliged to play 12 . . . a5, forestalling the positional threat of b3-b4-b5 with the seizure of the c6 square. In this case he need not have feared 1 3 e4, which would merely have increased the activity of the black pieces. 13 14 b4 lDd4! JJ..e7 The exchange of the light-square bishops, for which Black has been openly aiming, turns out to be to White's advantage, since it weakens the c6 square, making it easier to him to seize control of it. 14 15 'iitt xg2 il.xg2 ifc7 19 e4 19 20 b5! lD5f6 a6 21 lDc6 .lf8 22 23 24 a4 axb5 :.xal axb5 lbal l:la8 25 ktdl! In this way Black's counterplay is nipped in the bud, while the weak c6 square in his position remains. In the event of 20 . . . tiJeS 21 'i'e2, Black would have had to reckon with inevitably having to retreat his knight after f2-f4. Black apparently avoided the more active 2 1 . . . i.c5, not wishing that his bishop should subsequently come under attack (tiJb3). But this is perhaps what he should have played, since b3 is not the best square for the white knight. /?��;:·.?; 1.= f11� il/&�tJ{ : �f.:tkt 16 11fb3 An important manoeuvre. White will alwavs be able to answer . . 'i'b7+ with 'i'D: when the endgame is in his favour. . 16 17 18 ltfct 1ff3 llfc8 1fb7+ tDd5 Let us considei. a possible contin uation in the event' of the exchange of 220 . This move decides the game. On the a-file the lone rook is no danger, whereas on the d-file the white rook is in close contact with its other pieces and will play a leading role. 25 26 27 �c4 e5 · lbe8 · �c5 White can finally lift his control of d5, since the square is inaccessible to the enemy knight (27 . . . CiJc7? 28 l:.d7 ! tt:Jxd7 29 CiJe7+). 27 • • :cs • Black moves his rook from the same diagonal as the enemy queen, but . . . concedes the a-file for White to invade. 28 Ital llc7 29 :a7 ifxa7 Alas. the reply 28 . . . !:ta8 is not possible (29 l:xa8 'i'xa8 30 t'Lle7+). The queen has to be given up immediately, since after 29 . . . 'i'c8 30 ttJxb6 hardly anything can be gained for it. However, here too the compensation is clearly inadequate. Jlxa7 30 li:)xa7 31 li:)xb6 Black resigns. The b5 pawn cannot 3 4 5 6 7 tLlc3 g3 .i.g2 8 e3 0-0 b3 e6 b6 ..i.b7 ll.e7 d5 After 7 . . . 0-0 8 d4 cxd4 9 'i'xd4 according to the Encyclopaedia White has an appreciable advantage. With slight divergences, this opening variation has already occurred in Game 320 and will be repeated in Game 3 59. White's basic plan consists in fore stalling . . . d5-d4. With the white pawn at e3 this advance is unfavourable for Black: after the exchange on d4 the e file is opened, and in addition White gains a queenside pawn majority. In tum. White exchanges on d5 at a point when Black will be forced to recapture with his pawn, whereupon White plays d7-d4. Then the defence of the d5 pawn and the insufficiently active bishop at b7 will become something of a problem for Black. 8 0-0 be regained (if 3 1 . . .CiJc7 or 3 1 . . .l:tb7 there follows 32 'iic6), and it will soon have the last word. Game 321 Botvinnik-Petrosian USSR Team Championship Moscow 1 964 · English Opening 1 2 c4 lt)f3 The Encyclopaedia 8 . .. lDe4. Therefore it c5 �f6 221 recommends would have possibly been more useful for White, instead of castling, to play d2-d3 . 9 i..b 2 ttlc6 10 cxd5 exd5 11 d4 lies 12 13 l:tcl i..h J llc8 15 16 17 Now the position coincides exactly with Game 359, where White played I O d3 , which would seem t o b e more accurate than the continuation in the present game. This allows White to carry out his opening plan. A consultation game Bot vinnik and Polugayevsky against Keres and Prins (Holland 1 966) went 1 0 . . . ltlxd5 1 1 ltlxd5 'i'xd5 1 2 d4 :ad8 1 3 t'De5 'i'd6 14 dxc5 ii'xc5, and Black emerged from the opening with better prospects of equalising. l 1 . . .'i'd7 was hardly advisable. After 1 2 l:k l White gained a significant ad vantage both in Geller-Van Scheltinga ( 1 969� which went 1 2 . . J:Ud8 1 3 dxc5 bxc5 14 �4 liJe4 15 tiJd2, and in Csom-Miles ( 1 977) : 1 2 . . . :tac8 1 3 tiJe5. • • • .J:.el cxd4 it.f8 17 ifd3 l%e1 lLle5 g6 W'd8 A classic example of how to play such positions is provided by the well known game Reshevsky-Flohr ( 1 93 8). . 14 .i.b4 if xe8 A committing decision. 1 7 . . . .i.xc3 followed by . . . lDe4 suggests itself. when it would not be easy for White to demonstrate that the two bishops are a significant factor in this position. Now White gains the opportunity to establish his knight on the central e5 square. 18 19 20 Driving the rook off the c-file. since 1 3 . . . �c7 loses the exchange: 14 tiJb5 (14 . . i!.c8 1 5 tiJxc7 i.xh3 1 6 liJxe8). 13 l.1.b8 exd4 l:txe8+ aJ Black chooses a convenient moment for this exchange, ensuring the safety of his d5 and c5 pawns. White has to re capture on d4 with his pawn, since 1 5 liJxd4 lDxd4 1 6 if xd4 ..tc5 1 7 'iih4 d4 18 :led l dxe3 1 9 l:xd8 exf2+ leads to complications favourable to Black. However, now White's prospects are still better in view of the better placing of his pieces. 20 21 f3 21 22 23 'fldl . • • .i.g7 Thus, White has already achieved something: he has centralised his knight, defended the important e4 square, and occupied the open e-file. In addition Black has difficulties in bringing his rook into play. 222 l'Lla2 lpa5 a6 Threatening by 24 fi)b4 to shut the knight at a5 out of the game. 23 24 ·.i.� • • • lZ\c6 White agam · wants to play 25 fi)b4 . which would enable him to secure for a long time the position of the lrnight at e5 . Both now and later the exchange on e5 would be clearly to his advantage. 24 25 26 27 'fld2 .i.b2 lllc l 1lc7 a5 1!i'd6 .i.c8 knight from e5, but as a result the e6 square is weakened and the position of the bishop at g7 is worsened. .i.fl lllcd3 34 35 lllf4 f6. Jld7 • • ltlf5 . Now White wants to play 36 .tb4 and win the d5 pawn, and if 3 5 . . . tbc7 to · win the queen ! 35 . • 1ff8 . 35 . Jlf8 would have lost to 36 liJxd5 'i'xdS 37 %:txe8 (37 . . . �xe8 3 8 CDxf6+). . . .*.e6 tfJe7 lDg4 .i.c3 With the threat of 35 .tb4. Black finally drives away the annoy ing bishop. White, of course, avoids the exchange, rightly assuming that at e6 the black bishop will be not much more actively placed than at b7. 28 29 32 33 34 36 37 38 39 40 tbxd5 .i.b4 l'De7 lbxf5 d5 �h8 1if7 itled6 itlxf5 White achieves complete domination, and in addition he is a pawn up. 40 lle8 41 lle4 h5 42 l'Df2 · ·�·:;):} 30 b4! This leads to tl1e seizure of c6 idea that was also used in Grune 320. - 30 31 · 32 axb4 b5 an tiff:� . axb4 lDe8 In view of the thfeat of tbb4-c6, Black must immediately drive tl1e 223 42 �� g .i_� l:td8 This apparently natural move leads to Black's inevitable defeat. In my analysis of the adjourned position I considered Black ' s best chance to be 42 . . . i..h6 with the possible continuation 43 'ifd3 i.e3 , when 44 g4 cannot be played on account of 44 . . . hxg4 45 fxg4 i.xb5 ! 46 :xe8+ 1i'xe8 47 'i'xb5 i..xf2 + 48 �xf2 'i'e3+ 49 �g2 �h4+ with a draw, while after 44 �g2 i.c5 45 ..ixc5 bxc5 46 'i'c4, although White retains the advan tage, there is still much play to come. In his adjournment analysis the World Champion apparently did not take account of some nuances in the position. 43 .i.c4 Now White achieves the advantageous advance of his d-pawn. 43 44 45 d6 l:el �c8 ifd7 ..ib7 The d6 pawn is immune - in the event of 45 . . . �xd6 46 'i'xd6 'i'xd6 47 i.xd6 nxd6 48 lle8+ Black loses a piece. 46 47 lDe4 lbe4 .ixe4 lDxd6 47 . . . i.f8 48 l:te6, and the passed pawn is inunune (48 . . . ttJxd6 49 i..xd6 i.xd6 50 !lxd6 'i'xd6 5 1 'i'h6 mate). 48 .i.xd6 49 l'.Id4 .i.f8 Black noticed, of course, that after 48 . . . 'i'xd6 49 · :es+ he · would lose a rook. Also possible was 49 l:te6 �h7 50 i.xf8 ! 'i'xd2 5 1 :e7+ 'it>h8 52 i.g7+ 'it>h7 53 Si.xf6+ �h6 54 i..g7+ 'it>h7 (or 54 . . . 'iti>g5 55 %le5 mate) 55 .i.c3+ etc. 49 'it>h7 50 .*.xf8 But not 50 i.e5 because of 50 . . . .i.c5 . • 50 51 52 • • ll'xd4 ilxd4+ lhd4 i.fl It only remains for White to transfer his bishop to the gl -a7 diagonal. 52 53 54 55 i.. a3 cbf2 .i.b4 g5 lidl c/i;g7 h4 This ·diversion' is not able to change anyt11ing. 56 57 58 59 gxh4 Jil.el �g3 .i.f2 59 60 61 .tc4 f4 l:td4 �xh4 lid4 The main obj ective has been ful filled: now the rook will have to defend the b6 pawn. An oversighh:-but in a position where there was nothing at all to do! e.g. 224 l:tdl l:td6 Clearing the long diagonal for the manoeuvre i.e2-f3-c6. 61 �g6 fxg5 62 fxg5 · Black resigns 10 f3 A dubious plan, Game 3 22 against which Smyslov finds a convincing reply. 10 d3 was simpler and better. BotVirinik-Smyslov USSR Team Championship Moscow 1964 English OpeniQg 1 2 3 c4 ltlc3 g3 lDf6 e5 .i.b4 4 5 .i.g2 a3 0-0 Probably the simplest way of equal ising in this opening. A waste of an important tempo. 5 ttJf3 (Game 3 3 3 ) or 5 e4 was preferable. 5 6 .. bxcJ . .i.xc3 e4! An interesting idea. Smyslov carries over this well-known idea from the Sicilian Defence (for White) to the English Opening (for Black). Now White has to develop his king' s knight on an unsuitable square. 6 . l:.e8 7 e4 c6 followed by . . . d7-d5 is also possible (as recommended by Polugayevsky). .. 7 8 9 ltlbJ 0-0 ltlf4 l:te8 d6 b6 225 e3! 10 11 ... d3 11 12 13 1ie1 g4 13 14 15 h4 'ifg3 h6 ltlf8 ltlg6 16 17 18 lhhJ h5 .i.bl ltlh7 ltlb4 f5 By leaving the opponent with his pawn on e3, White risks getting into danger. After 1 1 dxe3 .ia6 Black would have advantageously regained his pawn, but even so this should evidently have been preferred. ..i.b7 ltlbd7 White needs to gain counterplay at any price. For this he transfers his queen to the kingside, and with its support he begins a pawn offensive there. The black knights have not only set up a secure defensive screen, but they are also ready to launch a counterattack. By skilfully manoeuvring with his knights,. .:•Black has pushed back the opponent �s pieces, and White's hopes of an attack have been sharply curtailed. Now Smyslov lands an energetic pawn blow - a significant link is his plan for developing his initiative. 19 -*.b2 1i'f6 The first point in the game when Black can be criticised. He could have retained the advantage with the active 1 9 . . . tbg5. Then in the event of 20 tL\f4 11b8 2 1 <it>h2 'i'f6 the white pieces are extremely badly placed (only not 20 . . . 'i'f6 immediately because of 2 1 'i'xh4 lDxf3+ 22 .i.xf3 'ifxh4 23 .i.xb7, and White has three minor pieces for the queen). White promptly makes use of the respite afforded to him to seize the initiative. 20 f4! .i.xhl In the new situation Black should have considered sacrificing the ex change (20 . . . fxg4 2 1 .i.xb7 gxh3 22 ..txa8 l:txa8), since it would not have been easy for White to show that his material advantage is sufficient for a win. 21 g5 This important intermediate move was apparently overlooked by my opponent when he played 1 9 . . . 'i'f6. 21 22 23 25 26 27 28 29 • • g6! hxg6 1i'b6 l:xf5 l:[afl :r6 thf8 1ig7 tllxg6 l:tf8 ifxh6 Black has managed to parry the immediate threats to his king, but now he faces a battle in an endgame a pawn down. 30 31 32 11fe5 i.c6 ltf4 • .. hxg5 fxg5 11xh4 White avoids the exchange of queens, since his chances of an attack have become very real. 23 24 24 White must not be allowed to bring his queen's rook into play - then Black's position would become hope less. Therefore, with a subtle pawn sac rifice, Smyslov mobilises all his forces for the defence. If Black is allowed time to play 25 . tDIB, it will no longer be possible to exchange pawns on g6. Therefore this must be done immediately. gxh6 lZ.xf6 .t.ct lbf6 �h7 The outcome is not so obvious after 32 l:tf7+ <it>xh6 3 3. l:.xc7 tbes (34 d4 ltlxc4 20), since the position of the white king is not secur�. Therefore White gives preference to a' ,continuation by 226 which he finally eliminates the annoy ing pawn at �� and brings his inactive bishop into . play. . 32 33 lt\g5+ 33 34 35 .lxe3 llti ci>xh6 <i>h5 �e8 l:t.b7+ 'it>g4 • • • :g8 In this way White protects his king against any misadventures. Hoping, apparently, to win a pawn (at e2). 36 37 <i>f2 An important subtlety. In the event of 37 . . . l::txe3 3 8 <it>xe3 <it?xg5 3 9 :xc7 and 40 l:r.xa7 White has a rook and three pawns against two minor pieces (effect ively an exchange advantage) and he should win without particular difficulty. 37 38 39 CiJ.e6 lDd4 :g7+1 <i>f4 41 l:tf7+ (cutting off the enemy king from the main battlefield) 4 1 . . . <it>g4 42 ttJxc6 lbdl+ 43 <it>e l �c3 44 'it>d2 (44 e4 lbxe4 ! ) 44 . . . ltJbl+ 45 <ittd l (less convincing is 45 �c2 lbxa3+ 46 <ii>b 2 ltlxc4+ 47 dxc4 :xe2+) 45 . . . lbc3+ 46 �c2 ltlxe2 47 %lxc7 etc. 40 41 42 43 cxd4 d5 -*.d4 e4 44 .i.f6 llc8 .ta4 a6 c5 A comical oversight (in adjournment analysis ! ) - true, in an already hopeless position. Black resigns: the threat of mate cannot be parried, only deferred by one move. CiJ.e7 Cjjf5 CiJ.xd4 Game 323 Botvinnik-Stein Black exchanges lmights, relying on the drawing tendencies of opposite colour bishops. The play would have developed djfferently, but also with advantage to White, after 3 9 . . . lbxe3 40 227 USSR Team Championship Moscow 1964 King' s Indian Defence 1 2 c4 �c3 g6 .lg7 3 4 5 .,6 d4 e4 ..i.eJ . f3 tt)f6 0--0 d6 b6 A year"before this game, in the USSR Spartakiad, Stein chose against me the line involving an inunediate attack on the queenside: 6 . . . lDc6, 7 . . . a6, 8 . �b8 and 9 . b5 . There Black gained a prom ising position from the opening, and the outcome was a draw. The 'change of tune' can only be explained by a fear of some preparation on my part. 7 i.dJ .*.b7 7 . . . a6 followed by . . . c7-c5 was also possible, but not immediately 7 . . . c5 because of 8 e5, when Black loses material due to the threat of i.e4. The drawback to 7 . . . i.b7 is that at an appropriate moment White can play d4d5, and the queen' s bishop will have to seek other diagonals to participate in the play. True, in the present game White did not exploit this possibility. . . . . 8 9 lDge2 0-0 After 12 . . . 1fxdl 13 l:tfxd l l'Lld4 14 i.a4 White' s position is preferable, but, of course, there is still all to play for. However, as it later transpires, Black had planned a queen sacrifice. Therefore it was more logical to refrain from' the move in the game and to play 1 2 ... llX14 immediately, in order after 1 3 l2Jd5 lDxd5 14 i.xd8 :l'Lle3 etc. to carry out the combination : in a more favourable situation than adtually occurred, since the h6 square would have remained accessible to his king's bishop. To be fair, it should be said that such inaccuracies were not typical of Stein's play. Evidently, if Black had played 1 2 . . . liJd4 immediately, White would have had to reply 1 3 i.d3 (declining the win of the queen), thereby losing an important tempo. 13 14 i.h4 ltld5 l£1d4 c5 9 d5 or 9 .i.c2 is more accurate. 9 10 • • • i.c2 ltlc6 And this is a mistake, throwing away White's opening advantage. It was essential to play 10 d5, not fearing I O . . liJe5 and . . . lDxd3. Now Black gives up the d5 square, but seizes con trol of the no less important d4 square. In the event of exchanges on the central d4 and d5. squares, Black's bishops will be better placed than White' s. . 10 11 12 dxe5 .i.g5 e5 dxe5 h6 228 14 . • . tt)xd5 A clever queen �crifice, typical of Stein's inventive style, which, however, does not present any danger to White. 15 i.xd8 · ltleJ 16 1fd3 l:axd8 17 18 'ifxe3 1fc3 li:)xc2 Black has nothilig else. If 16 . . . ll'ldxc2 1 7 .th4 lt)xa l 1 8 l:xa l, then the knight at e3 is trapped, and after 1 7 . . . tiJxfl 1 8 'i'xc2 ll'le3 19 'i'd3 the same knight perishes on the same square. A very cautious move, .but by no means the best. It was possible� not fear ing the invasion of the black rook at d2, to play the queen to b3, leaving the c3 square free for the manoeuvre ll'lc3-d5. 18 19 ttlxal 21 22 23 • • • h5 'ifa4 li)cJ :d2 .i.c8 l:tdl! The decisive move. After the ex change. of rooks, Black's hopes of an attack along the second rank will dis appear; therefore he is forced to concede the d-:file. l:txal 23 24 lixb2 �d5 .ia6 19 White' s inaccurate 1 8th move gave Black a respite, which he should have exploited for active play. 1 9 . . . f5 20 'i'c2 fxe4 2 1 fxe4 :n suggested itself, or else 1 9 . . . l:d7 immediately, in both cases followed by the doubling of rooks. The move in the game again leads to a difficult position for Black. 20 20 This attempt to bring the king's bishop into play is too late. Perhaps slightly better chances would have been given by 20 . . . b5 2 1 tLlc3 bxc4 22 ifa4 (22 . . . l:.d6 23 :!:ldl). 'ifc2! This reply, with the unpleasant threat of 2 1 'i'a4 (or lDc3 ), was apparently underestimated by Black. 24 • • . .fl.e6 White was threatening to win by 25 'i'a3 %k2 26 ifb3 e2 27 � . Therefore Black could not reply 24 . . . a6. It is true that now too White can win the enemy rook for a knight and pawn (25 'i'a3 Axd5 26 ifxb2 .i.xc4), but it is more advantageous to begin the destruction of the enemy pawns. 229 25 26 27 9'xa7 exd5 d6 .i.xd5 b5 bxc4 28 29 d7 9xc5 c3 occurred in a similar position in the game Spassky-Bronstein ( 1 956). How ever, subsequently · it was shown that this sacrifice is not dangerous for White. There was also another way: 29 d8'i' c2 30 'i'xc5. 29 30 • • • 9 c2 (1...0. -0 licl 30 d8'i' would also have won, but, as on the previous move, the chosen continuation is the simplest. 30 31 32 33 ltd8 .i.f8 lhc2 :bs ires 1!t'xb8 Black resigns Game 3 24 Botvinnik-Medina Now it turns out that Black cannot advance his c-pawn, since his d6 pawn will be en prise. And with the pawn at c7 it is more difficult for him to gain counterplay on the queenside with . . . a7a6 and . . . b7-b5 . OZvmpiad, Tel Aviv 1964 King's Indian Defence 1 2 3 4 5 6 c4 l£lc3 e4 d4 f3 il.eJ l£if6 g6 d6 .*.g7 0-0 e5 9 10 11 The flexibility of the King's Indian Defence is vividly demonstrated by this and the previous game. In No. 323 Black initially played on the queenside, where as here he is aiming to seize the initia tive in the centre and on the kingside. 7 d5 l£ih5 According to the Encyclopaedia, 7 . . . c6 creates equal chances. 8 1id2 f5 8 'ib4+ leads to more complicated and unclear play, intending if 9 g3 to carry out the combination 9 . . . tbxg3 I 0 'i'f2 ltJxfl · 1 1 'i'xh4 t2Jxe3, which .i.d3 tl:\ge2 lLld7 lLldf6 a6 If l 1 . . .fxe4 White would not have replied 12 fxe4 on account of 12 . . . l2Jg4, but 12 t2Jxe4 tbxe4 13 i.xe4. 12 exf5 Another possibility is examined in the Encyclopaedia by Lilienthal and Florian: 12 �b l , and if 12 . . . b5 1 3 cxb5 f.xe4 14 tbxe4 t2Jxe4 15 fxe4 lDf6 16 bxa6 tbg4 17 lDc3 , which also favours White. 12 13 . . . ... tl:\gJ . gxf5 The same position was reached in the final, 2 1 st game of the Botvinnik-Tal 230 It is possible that Black decided to defer this sacrifice by one move, but that he overlooked his opponent' s cunning reply. Return Match (No.294), the only differ ence being that instead of 'it>b 1 White has played .i..d3 , and this is probably in his favour. In .view of the threat of 14 etJxf5, the subsequent development of events is forced,· and an ·'evaluation of them was given in the notes to White's 1 3th move in the afore-mentioned game. 13 14 15 16 ltlxh5 l£\xf6+ 11c2 f4 fxe3 V xf6 The e3 pawn does not present any serious danger to White, and for the moment he can pay no attention to it. Of more significance is the possibility of occupying e4 with his minor pieces and of restricting Black's king's bishop. 16 • . • 17 lhe4! 17 18 ... <i>bl .i.f5 .i.xe4 19 .i.xe4 b6 20 21 l:tdel 11d3 .tf6 21 22 23 24 25 :xe3 :e2 'ii'cJ 1iet 25 26 gJ Ignoring the threatened discovered check, since after 1 7 . . . e2+ 18 J:.d2 'i'xh2 1 9 l:r.xh2 e l 'i'+ 20 J:tdl 'i'e3+ 2 1 'it>b 1 it i s doubtful whether the position of the black king can be defended. And . . . e5-e4 is no longer possible. Black is obviously pinning his saving hopes on the presence of opposite colour bishops, but it has long been known that they often increase the chances of the attacking side. 11h6 Black can defend his e3 pawn by . . . ..if6-g5, but for this he should have moved his king in advance to h8, as will subsequently become clear. Now 2 1 . . . .i.g5 can no longer he played on account of 22 h4 .i.f4 23 g3 (23 . . . ..txg3 24 :egl ), but if instead of the pointless move 1 9 . . . b6 Black had played 1 9 . . . �h8, he would have re tained the possibility of this defence. For the reasons explained in the previous note, it was essential to play 1 6 . . . e4 ! , after which the black bishop would have come alive', and the a 1 -h8 could have become a base for potential threats to the white king. As for the conversion of White's material advan tage - his extra pawn, this would have entailed technical difficulties. <i>h8 il.g5 .i.f4 .:n This completes the queen manoeuvre, necessary for driving back the enemy bishop. 23 1 g7 1' .i.h6 27 h4 a5 J:g6 40 h6 41 .i.e4 Black resigns. If 4 1 . . J �e6 (or 4 1 . . . Such a move is the best admission of the lack of any prospects for Black. 28 29 30 31 l:tgl l:teg2 a3 g4 �g8 'iff8 ltfg7 Only in this way should the kingside pawns be advanced. The conceding of the f4 square is no longer significant. 31 32 33 34 35 g5 :g4 1i'e2 .idJ l:gg8) there would have followed 42 ..txh7 ! 'it>xh7 43 g6+ �g8 (43 . . . �xh6 44 g7 and 45 �h l +) 44 g7 etc. International Master Medina was quite a good tactician, but he lacked subtle positional understanding. i.f4 'l'c8 fid7 �b8 c6 Game 325 Letelier-Botvinnik O(vmpiad, Tel Aviv 1 964 . Black exploits the very first oppor tunity to gain some counterplay with this move. But in the process his pieces are diverted away from the kingside, which White promptly exploits. 36 37 38 39 dxc6 h5 .i.e4 '9xc6 l'r.bg8 1ffc5 llf8 3 9 . . :xg5 40 .i.xg8 'i'xg l + 4 1 l:txg 1 ltxgl + 42 �a2 would also not have saved the game, altl1ough the resistance might have �een prolonged a little. . i..d5 Slav Defence 1 2 3 4 5 d4 c4 ltlcJ cxd5 �fJ �f6 c6 d5 cxd5 �c6 It was of no significance, of course, that in a training game against Petrosian (see Volume 2) I pl,ayed first . . . e7-e6 and . . . tiJc6 only on the 6th move, while against me Citrone (No. 268) went 5 .i.f4 before ctJc3 . J 232 � · �i 6 .i.f4 castling (on either side) is . inadvisable. In passing, the attack on the .b2 pawn is renewed (it was dangerous ..to play 1 3 . 'i'xb2 inunediately because of 14 ltJb5). e6 ti)b5 In Grune 268 White played 7 .i.g3 (there ltJc3 and . . . e7-e6 had not yet been played). There it was also shown that the attack on the queen, which in any case is aiming for b6, is pointless, and that· 8 .1e5 should have been preferred. 8 9 .. 1fb6 . .i.h4 .i.g3 hxg3 13 :ct b4 14 15 t;)g1 15 16 17 ti)a4 b5 18 .i.d3 . . • g4 Intending to transfer the knight to f4. a3 Indirectly defending the b2 pawn, since 9 . . 'i'xb2 is not possible because of 10 ltJa4. 9 10 11 12 14 A risky continuation, weakening the queenside pawns. h6 g5 ti)xg3 .i.g7 a5 11fd8 t'£Je7 Black defends against the threat of 1 3 ltJxg5 and strengthens his control of e5 . This is less natural than 13 .i.d3, as Petrosian played. However� after 1 3 .i.d3 'i'd8 14 ltJh.2 h 5 Black also achieved a good game. If 18 Wxg4 there would have followed 18 . . . e5, forcing the opening of the position, which is highly unpleasant for White, since his king is still in the centre. Now, however, this advance is not so dangerous. 18 19 20 21 13 @fS! The king occupies its most favourable position in a situation where dxe5 ti)e2 e5 .i.xe5 1fd6 h5 <i>f1 Black declines the pawn sacrifice, since after 2 1 . . .�xa3 22 tL'lb6 :bs 23 ltxc8+ and 24 tiJd7+ he loses material. 233 22 23 24 1fb3 lldt l£lf4 28 gxf4 1h:f4 The main defect of� White's position is the fact that his rook at hl is shut out of the game. Therefore, so as not to perish under the fire of an enemy attack, he seeks salvation in an endgame. b6 .i.b7 l:th6! This rook manoeuvre� with a king that has given up the right to castle, was one that I tried back in 1 944 in my game with Kotov (No. 1 3 9), and subsequently in the afore-mentioned training game with Petrosian. 25 29 30 31 32 �gl The whole point is that the h5 pawn is immune. 25 :xh5 loses immediately to 25 . . . .i.xf4, while after 25 tiJxh5 d4 26 exd4 .i.xd4 Black's pieces are dangerously trained on the enemy king. The following variation, for example� was possible: 27 �g l .ltxf2+ 28 <itixf2 'ii'd4+ 29 <t>f°l tiJd5 and White has no defence. 25 26 • • • 1t'e3 fxe3 .i.fl liel 'ifxe3 l:tc2 tl)f5 lDgJ �cs .ile2 It was again dangerous to take the pawn - 26 tiJxh5 in view of 26 . . . d4 27 exd4 .i.xd4 28 lDf4 l:txhl + 29 �xhl .t.xf2 . But all the same tl1e opening of the position is inevitable. This knight manoeuvre decides the outcome. lDxfl 33 l:h2 llf6+ 34 <t>xfl White resigns. If 3 5 �gl there follows 3 5 . . . g3 , and after the rook moves it is mate in a few moves. Game 3 26 Aloni-Botvinnik OZvmpiad, Tel Aviv 1964 King's Indian Defence 26 27 exd4 d4 .i.xf4 This was my seventh and last Chess Olympiad. After losing the match to Petrosian, I, naturally� played on board two for the USSR team. Although all 234 Since White cannot castle immed iately, this move, creating a threat to the e4 pawn, suggests itself. the points in a team are added together, on board two the responsibility is less, and some liberties can be taken. And against Aloni I was able to play a •jolly' game with Black. 1 2 d4 c4 ltlf6 c5 3 4 5 6 d5 ltJcJ e4 h3 g6 d6 i1..g7 10 11 8 12 13 flide5 White does not achieve anything with 1 3 'i'xd6, when there can follow ' 1 3 . tDxf3+ 14 gxf3 thd4 or 13 . . . ttJxc4 immediately. 0-0 e6 13 • . • dxe5 1 3 . . . tDxe5 14 'i'xd6 ltJxc4 1 5 i.xc4 (or 15 'i'xc5 15 . . . 'i'xc5 16 ii.xc5 %tfc8) 15 . . . Jl.xc4 16 lUd 1 would have led to less interesting play. I was very curious to see how a player with such an attacking style would try to create an attack in such a 'reinforced concrete' position. 14 la.ad 1 15 .i.d3 llid4 dxe6 After this exchange Black no longer has any opening difficulties. He gains a tempo for the development of his queen' s bishop and quickly concludes the mobilisation of his forces. 8 tl)f3 was essential. 8 9 0-0 ltJxe5 . . This move is premature. Here 6 i..d3 is more consistent, but Aloni was never strong in opening theory. i.e3 thc6 ltld7! The manoeuvre . . . tDf6-d7-e5 leads to the elimination of the weak d6 pawn and to the creation of an almost sym metric position. Black is obviously aiming for sharp play, for which he chooses. one of the most complicated Indian set-ups. 6 7 ifd2 i.e2 lbtJ .i.xe6 1f a5 My reckoning is justified. White avoids the quiet 1 5 lLJd5 with equal 235 play. Then 15 . . . 'ii'xa2? is not possible on account of 16 i.xd4, followed by 1 7 � a l 'i'b3 1 8 �a3 . Instead of this Aloni prepares the exchange of the dark square bishops - 16 .ilh6 (the immed iate 1 5 i.h6 was not possible in view of 1 5 . . . l!Llxe2+ 16 tLlxe2 'i'xd2 17 .ixd2 i..xc4). It is hard to say whether my opponent overlooked the loss of a pawn or deliberately sacrificed it. It was most probably the latter, since his subsequent play is clever and energetic. 15 16 ... b4! Threatening by a4-a5 either to win a pawn, or to create a strong passed c pawn. Black decides that the only thing is to begin counterplay on the kingside. 21 22 23 ... exf5 a5 24 .tlb5 f5 i.xf5 bxa5 The last chance to win the game, a chance based on psychology: suppose that the opponent should wish to regain one of the sacrificed pawns as soon as possible? .i.xhJ! After 16 i.xd4 exd4 17 gxh3 dxc3 Black would have had both a material, and a positional advantage. White temporarily sacrifices a second pawn, but seizes the initiative. 16 • . • cxb4 16 . . .'i'xb4 1 7 l:lbl 'i'a5 18 l:.xb7 did not appeal to me - the b-file would have remained in White's possession. 17 tLld5 il.g4 It would appear that Black is merely aiding the development of White's initiative. Later I incorrectly concluded that 1 7 . . . ild7 should have been played, but I overlooked that then White would have replied 1 8 a3 ! with a very active game ( 1 8 . . . 'i'xa3 ? 1 9 .txd4 exd4 20 :a l 'i'b3 2 1 :ib l ). Now, however, Black is able to parry the threat of a2-a3 . 18 :!bl 'ifd8 Otherwise 19 a3 would have been unpleasant. 19 20 21 i.xd4 llxb4 a4? exd4 b6 And that is what happens - Aloni does not find the strongest move 24 :b7 ! , after which Black could have hardly hoped to win. After both 24 . . . �xd3 25 'ilxd3 lif7 26 !txf7 <l;xn 27 'i'D+ 'it>g8 28 c5 !k8 29 c6, and 24 . . . 'ib4 25 g3 'i'g4 26 'iti>g2 ! with the threat of 27 tl:Je7+ and 28 :ht White would have been out of danger; more over, Black himself would have had to seek a way to draw. Now Black can double rooks on the f-file, which should deciqe the outcome. 24 25 236 'i'xd3 i.xd3 flh4 26 g3 This weakening . was not necessary� White decided to free his rook from having to defend the. f2 pawn. 26 27 • • • l:tel 'il'bl lle4 This move is White's only chance the black queen has no good move (i(, for example, 29. . . 'i'h3, then 30 ti:Je7+ <bh8 3 1 ctJxg6+ hxg6 32 lih4+). But it is not obliged to move ! 29 30 31 31 'iti>b8 'ifhJ+ d2! This is the end - the d2 pawn is so 32 33 34 ifg4 A desperate attempt to change the inevitable course of events. After 27 �g2 l:lf7 ! (this is where the absence of the white rook from b7 tells!) followed by . . . llaf8 White's position would have become hopeless. 27 28 29 The advance of the central pawn is a bad sign for White. Meanwhile, it was as a result of the move 2 1 . . . f5 that the blockade of this pawn was lifted. fxg3 <i>hl .&txg3+ 1fxg3+ dJ! This move, as the late grandmaster Tolush liked to say, cuts off the infantry from the tanks: the queen is temporarily shut out of the game. llle7+ 1fe1 <i>gl strong, that no counter-threats by White are able to help. 35 36 lllxg6+ 1i'h4+ hxg6 The tragedy for White is that after 36 l::th4+ 'it>g8 ! the e6 square is defended and he himself is mated. 36 <it>g8! . White resigns, since 3 7 ifxh3 leads • • • to mate after 37 . . . d l 'i'+ 3 8 <iifi>h2 1!f2+ 3 9 'iti>gJ 'ii'g l + 40 @h4 .if6+ 4 1 J:.g5 ifxg5 mate, while if 3 7 :lb l there follows 37 . . . 'ib3 ! There is an amusing story associated with this dynamic game. In 1 973 during a tour of West Gennany I gave a lecture in Seigen (where in 1 970 the Chess Olympiad was held). I was invited io a local computing centre. There the chess players decided to test my memory and they showed me a demonstration board where the final position from this game had been set up. I sensed that the position was very familiar, but initially I could not remember it. There was a simple explanation for this: against Aloni I was Black, but when I had to look at the position from White's side, it was only · with difficulty that I recognised it. A curious illus tration of the peculiarity of a chess player' s thi�ing. 237 mistake 7 d5. Things would have been more difficult for · )\T,hite after 1 1 exd5, since Black woufd ·have gained the opportunity to develop his bishop with gain of tempo a_t f5: Game 327 · Y anofsky-Botvinnik O�ympiad, Tel Aviv 1964 Pirc-Ufimtsev Defence 1 2 3 4 5 e4 d4 ttlcJ f4 ttlf3 d6 ttlf6 g6 i.g7 0-0 This position occurred later in the game Gipslis-Botvinnik (No.334 ), where White played the most energetic move, 6 e5. The continuation employed by Yanofsky is less dangerous for Black (although it is reco1mnended by the Encyclopaedia). 6 7 i.d3 d5 ttlc6 11 12 13 This merely aids the development of White' s initiative, since he gains the opportunity to undermine the centre by . . . c7-c6. In my view 7 ..te3 is stronger, and 7 e5 is also good. 7 8 ... ... aJ 14 15 16 17 ttlb4 9 10 11 11 xd3 ttJxd5 19 ttlxd3 cxd5 White no longer has his powerful pawn centre, and in addition Black has the advantage of the two bishops - such are the consequences of the positional l1e1 cJ ii'dJ 'ifc2 llac8 a5 i.b5 a4 Depriving White's queenside pawns of any mobility. 18 i..e3 1!f a6 c6 After 9 dxc6 bxc6 10 h3 ttJxd3 1 1 cxd3 Wb6+ Black has no difficulties (Lancaster-Hartman� 1 968). ttlxd5 1'b6+ .i.d7 The b2 pawn is, of course, immune ( 1 3 . . . ..txb2 14 l1b l ) . 0-0 A game Mikenas-Etruk ( 1 968) went 8 ..ie2 a5, and White did not gain any advantage. 8 9 1ixd5 'Wfi>hl i.d4 Ah6 There is a noticeable similarity of ideas employed by Black in this case and in the game Bemstein-Botvinnik (No. 160). The main difference is that here the queens remain on the board almost to the end, whereas there they were soon exchanged. 238 20 9'f2 A tempting, buf incorrect paw:Q. sacrifice. It is true that after 20 ttJg5 e5 2 1 .i.e3 .i.xg5 22 fxg5 .i.d3 White would have lost a pawn without any compensation, but if he had played 20 f5, there would still have been a lot of play in prospect. 20 21 e5! .txf4 This forced abandoning of the attack is equivalent to capitulation. 28 29 30 31 lDxfJ 9b3 :gt 1'xf3 ..tc6 ltt7 At least White is consistent. He would not have achieved anything after 2 1 'i'h4 e5 followed by . . . f7-f6 . 31 .i.xfJ This way, rather than 3 1 . . . 'i'e2, when White still had an active possibility (3 2 tDg5). f6! 21 A worthy reply! By advancing . . . e7e5, Black consolidates his advantage. 22 23 24 25 exf6 1ifh4 i.e3 l1xe3 e5 It.ti i.xe3 :c4 In this way Black eliminates the f6 pawn, since if 26 'i'g5 there follows 26 J:tf4 . l:txt'6 26 'i'h6 27 lbg5 :c7 Eliminating White ' s last hopes of a successful attack on h7 in view of possible counter-threats of mate, e.g. 28 &th3 ilc6 ! 29 llg l 'i'e2 (30 CDxh7 iixg2+ 3 1 l::txg2 :n mate). 28 :o . . 239 32 33 gxf3 'ifg4 33 34 35 'l'xf3 .rldl 36 37 38 :txd6 c4 lid5 llxb2 11.bJ ltxa3 40 lld7+ <i>e6 'fid3 It was not possible to save the pawn: 33 :n e4. After the passive defence of the pawn (35 . . J:If6) it would all have been more complicated (36 l:td5). Temporarily White is now three pawns down, and it is time to lower the curtain. <i;f7 39 l:td8+ lial+ 41 llxb7 White resigns (his last hopes, associated with time trouble, having been dashed). .. This was my second meeting with the Canadian grandmaster, who in fact received this title during the Olympiad. In Groningen (1946) at a critical point of the tournament I lost to him, although in the first half of the game I had held an obvious advantage. Game 328 Botvinnik-Gligoric Griinfeld Defence lDf6 g6 d5 ..i.g7 5 6 7 0-0 c5 9 • . . .ig4 Of course. The resulting great com plications are advantageous to Black, since his opponent's lack of develop ment is bound to tell, and also the knight at g5 is badly placed. 10 f3 10 11 .i.gJ Even now it was still possible to give preference to the quieter 10 .i.e2. Olympiad, Tel Avi\, 1964 1 d4 2 c4 3 lLJc3 4 ..i.f4 5 e3 already been played (without these moves, tt:Jg5 or lbcl4 is not bad). White also chose a poor continuation in Game 105 - 9 'iia4, removing an attack on the central d5 pawn. The best would seem to be 9 �e2, but then 9 . . . CDe4 is possible, when Black has active play for the sacrificed pawn. e5! I simply mixed up the move order, playing 5 e3 instead of 5 t2Jf3 , thereby allowing my opponent to choose a v�riation, the advantages of which for Black I had demonstrated nearly 30 years earlier in my games with Tolush (No. 97) and Ragozin (No. l OS). l:icl dxc5 ..te6 Against Tolush I employed the less convincing 7 . . . 'ira5, but with Ragozin I played the same way that Gligoric does here. 8 lLJfJ 9 .li)gs �c6 I made this . move automatically. forgetting �at e2-e3 and . . . tbc6 h�d 11 d4 All in the same energetic style. White now thought it best to choose th� continuation with the exchange of queens, so that the position of his king, caught in the centre, should not be so dangerous. 240 12 · fxg4 13 1ixd8 · . dxc3 :rx:d8 The author. The author. Author lovely photograph. First attempt as a typi st. · With G l i goric, 1 970. Celebrating the 5 0th anniversary of the first U S S R Championship. Botv i n n i k and Taimanov. B otvinnik - S myslov, 1 966. O ' Kel l y, Flohr and B otvi n n i k in Varna, 1 962 Before the match. B otvin n i k and Petrosian, 1 963. At the 1 95 8 Olympiad. The only game with Fischer, i n 1 962. In the electrical engineering laboratory, 1 962 . At the dacha, 1 96 3 . I n the Mercedes M useum with Yakov Estrin, 1 97 8 . Oberhausen, 1 96 1 . Before a game with Alexander Tolush, 1 965 . The 1 964 Olymp i ad . The match against Switzerland. The author. Autographs. B otvi n ni k ' s famous chess school, 1 963 . On the right is the young Anatoly Karpov. The chess school with the young Garry Kasparov, 1 97 5 . At the 1 962 O l y mpiad. The author, at home The author The author The author At the dacha, with the fami ly. With FIDE Vice-President Il makunas, 1 96 1 . With a group of amateur players from the Armed Services, Moscow. Analysing with S mys lov. B otvinnik and Tal . The Belgrade Tournament of 1 969. S i m ultaneous display in Brask. The author. After a game with Tol ush, 1 965 . Caricature. The USSR team - winners of the 1 95 8 O lympiad . Lecture in the Polytechnic Museum. Analysing on a pocket set. B otvinnik - Petrosian, 1 963 . During the 1 964 O l ympiad. In the thermal waters of the Ri ver Paratunka (Camchatka), 1 968. Duri ng the l ast tournament of the author ' s career, w i th Spassky, the i nterpreter and h i s wife. The author. The Chess S chool, 1 986. B otvi n n i k and Euwe. After a s i multaneous d isplay against the B ritish Junior Team. 14 l:xc3 A waste of precious time. Black h6 have won by 23 . �e6 (not 23 . . . %td2 24 ilfl) with the threat of 24 . . . htd2 25 .i.f2 liJd4. could For the moment White has a material advantage, but how much worse his pawn formation is than the opponent' s ! In addition Black's pieces are better placed, and it is no surprise that soon he 24 25 Now regains everything. 15 16 17 18 lt)f3 %let J\e2 ltJd4 ltJe4 ltlxc5 e4 driving ltJxd4 .i.xd4 .i.e3 .i.e5 ..txf4 its in strong .let �f4 .i.xf4 c5 It is essential to activate the bishop! 28 29 cation is his best way out of the current exd4 b3 l:tdl White at least succeeds the bishop from 25 26 27 28 By giving up a second pawn, White 18 19 20 21 h4 l1b3 position. at least achieves the exchange of another pair of minor pieces� simplifi situation. . . lld2 l:c3 All the same the pawn cannot be saved, and for the moment, exploiting the fact that d2 is occupied by the black rook, White brings his own rook into play. 29 30 lba2 ltb2 l:txdl+ 21 Of course, @fl not 2 1 . . . ltld3+. This knight must be exchanged for the bishop 31 at g3 . With the opposite-colour bishops �increase even in the endgame. 22 23 .i.xdl .i.e2 l:ld8 a5 h5! first Probably the then remaining, Black' s initiative will good move made by White in this long-suffering game. Since after 32 hxg6 fxg6 Black's pawns will 24 1 be somewhat devalued, the exchange on h5 is forced, enabling White to get rid of his doubled pawn. 31 32 gxh5 33 g4 gxh5 </;g7 If 32 . . . f5 there could have followed 33 c6 bxc6 34 :xc6, and White's pieces are activated. Thus, White has managed to prevent the coordination of the e- and f-pawns. 33 34 35 � 'it?g2 41 .*.g5 ltbl+ il.d2 In this way Black wins another pawn, and it was, of course, hard to resist this temptation. Meanwhile, at g5 the bishop was ideally placed, in particular eliminating the possible threat of g4-g5, whereas its position at d2 subsequently makes it hard for Black to convert his material advantage. 36 37 38 39 40 Ac4 c6 :xc6 .:t.a6 � which move I had sealed, and they themselves replied: · .'Of course, 4 1 g5 after all, it leads to a draw. ' It was uncomfortable for me to admit that I had taken a different decision. Incident ally, in the course of the analysis it transpired that the 'correct ' move 4 1 g5 would have lost, and that the move sealed by me was the only possible one . . . <it>e4 The point is that, if the black rook were to occupy d4, controlling the important a4, d l and f4 squares, White's position would become desperate. In the adjournment analysis a number of subtleties were discovered, and when the game was resumed both players acted quickly and confidently. 41 42 �4 :eS+ �e6 First Black must drive the enemy rook from the sixth rank. e3 bxc6 l:bb3 llb8 l:td8 43 l:ta8 The ending with opposite-colour bishops after the exchange of rooks is lost, due to the presence of the e6 pawn (as will be evident from subsequent comments). 43 44 45 <&t>e4 <!i>dJ D.f6+ lif2 �6 If Black penetrates with his king to g5, at the same time establishing his rook at f6, he will win easily by then taking his king to f2 . 46 Here the game was adjourned. My team colleagues and trainers asked me 242 l:th8 �g7 An enforced return. After 46 . . . @g5 47 l:tg8+ ! �4 (47 . . .�h4 48 g5 !) 48 l:lh8 Black would not have achieved anything. · - 47 :aS Now the exchange of rooks after 55 . . :es+ 56 �fl .:e6 57 l:I.xf7 IU6+ 58 l:xf6 �xf6 59 1-dl Cit>e5 60 h4J ! il.cl (or 60 �d4 6 1 g5 hxg5 62 h6 �d3 21 63 .i.b5+) 61 .i.c2 1 ! <it>d4 62 g5 hxg5 63 h6 i..b2 64 h7 leads to a draw. The idea of this manoeuvre (.idl-a4-c2) was suggested by Geller� it is possible only because of the unfortunate position of the bishop at d2. If instead Black defends his ii pawn - 55 . J:tf8, then 56 �f3 f5 57 l1g7+ �6 (or 57 . . . <ifi>h4 58 gxf5 .:txf5+ 59 'it?e4 and the black king is cut off) 58 :g6+ �e5 59 g5 hxg5 60 l:txg5, and the passed h pawn gives White some hopes of saving the game. However� after prolonged reflection Black took a different decision� which came as a surprise to me. ll.f4 A little trick: after 48 :a6 l:tf6! 49 1:ta8 �e6 the white king would b_e cut off from the klligSide. 48 l:b8 l:t,f6 49 'it>e4 l:tc6 . ... Black's first attempt to find a wimring plan has not succeeded, and Gligoric returns to the initial position. Another subtlety: what if White were to play 50 �f4, and then by 50 . :c4+! ( 5 1 .i.xc4 e2+) Black were to advan tageously transfer Iris rook to t11e fourth rank? . . 50 51 l:ta8 l:la7 . <itf6 Only in this way can tl1e black king be prevented from going to g5 Both 5 1 �h8 �g5 and 5 1 �f4 llc4+! were bad for White. . 51 52 �4 52 lte6+ . . • ilb4 In my analysis I had thought that the manoeuvre of the bishop to g5 was not possible. 53 c.t>f3 Ae5 <li>f4 .id6 li..e7 .i.d6 The bishop has to be freed from having to defend the a5 pawn. 54 55 56 In this position I was afraid qf 52 .. J:td6 53 �e4 l:d8 ! !, when White is in zugzwang (54 .i.c4 l:e8+ 55 � .fl.e7 56 l:.xe7 or 54 l:tb7 l:ta8). His last chance wou ld have been 54 .:ta6+ �g5 55 lla7 ! , 'ifi>f3 <itf4 By repeating moves Black gains time on the clock. 'iftg7! 57 �f3 A new zugzwang position, which I did not discover during my analysis. I thought that to 57 . . . Ji...e7 I would be able to reply 58 'it?f4, but now it is White to move, and he has no useful reply! After thinking for some 20 minutes I worked out one lengthy, seemingly 243 , drawing variation. I distrusted it, but since it led to a draw, how could I reject it? So, I decided to allow the black bishop to gS and to attack the fl pawn there was nothing else. complicated ending. · In this position, I have to confess, I was hoping to save the game: 69 . . . <i;O 76 Ji.di+ @f4 7 1 .i.c2 �g4 7 2 �e2 �:xtis .7 3 f6 ! But after the game the clever Geller nevertheless found a win for Black: 69 . . . i.f6 ! ! 70 �e2 .id4 7 1 � <ittg5 72 'it>e2 Aa.7 73 � i.c5 ! 74 �e2 'it?xh5 75 f6 'it>gS 76 f7 �4 ! 1��1t �t� - - 58 .i.c4 �·=:%.;:: i..e7 M& 59 -*.e2 .ild6 This now seemed strange to me . . . Surely Gligoric could see that he was not prohibited from playing his bishop to gS? As the reader will see, I was superior to my opponent in subtlety of analysis� but Gligoric was a very resourceful practical player. 60 61 .lc4 .le2 J.e7 <t1"6 Yes� Black did not see 6 1 . . . ..igS ! 62 Jtc4 �f8 ! ! 63 :txf7+ <ates 64 :rs (64 J:ta7 �d8 is hopeless for White) 64 . . . J:lxfS+ 6S gxf5 <i;e7 66 .ib3 �f6 67 .ic2 @es 68 �e2 <itf4 (but not 68 . . . �d4 because of 69 i.a4 'it>c3 70 f6, with a theoretical draw, since the dark square bishop plus h-pawn do not give a win) 69 <i;fl ! However, it was not only Gligoric who did not completely understand this ·;(.;,.�;,; � 4��/, if3.@ Ji&. �!ttf.$. z�� �WJt� \01 m Already threatening . . . �6-g5. 62 �4 �� • .i.b4 Another transparent trap - 63 l:Ia6+ J:te6 64 l:lxe6+ fxe6 ! 65 l2txe3 a4 would, of course� lead to an easy win for Black. 63 il.c4 l:e6 Again a two-move trap: 64 .i.xe6 e2 65 i.c4 (with the seemingly terrible threat of 66 l::txf7 mate) 65 . . . .td6+, and White must resign. But now Black loses his f7 pawn. and the position becomes drawish. 64 'it>f3 l::tc6 65 66 :xf7+ .ib5 �e5 l:.c3 Here I saw perfectly well that by playing 6 7 g5 ! hxg5 68 h6 e2 + (Black has nothing else) 69 �Ke� l:h3 70 h7 I could attain a dearly drawn position, but fate intervened: ·1 decided to play for 244 a win. This was ridiculous, of course, but for the · moment White still has a :.: draw. 67 68 69 70 l:.h7; l'th8 :eS+ l:te6 .tf8 .i.g7 <i>d4 This is already dangerous. The simplest way to draw was 70 l:e4+ ! �c5 7 1 i.. e2 i..d4 72 :te6. 70 ... l:tb3 Black misses the strongest contin uation 70 . . . .i.eS ! with the threat of 7 1 . . . l:ic2. Now, however, with 7 1 .ta4 ! na3 72 Ji.di i.e5 (otherwise the e3 pawn is lost) 73 l:.xh6 White could have attained the position that occurred in the game, but with the significant difference that the black rook would be passively placed. 71 .i.e2 .i.e5! simple reply 72 . . . :bs 73 <itiig2 �f8 74 i.f3 , since in my calculations 1 made two successive moves for Black (72 . . J !b8 and 73 . . . llfS+). Filled with horror, I began looking for other possibilities, and I decided first of all to drive the black king away from the e3 pawn by :e8-d8+. but this time I made two successive moves . . . for White .. 72 :es After this obvious mistake it is time for White to resign. To be fair, it should be mentioned that even after 72 l:.xh6 Flohr also pointed out the variation 72 . . J 1b7 ! ! 73 <it?g2 I!f7 74 .1i.f3 J:.a7 ! 75 :b6 a4 76 l::tb2 a3 77 l:ta2 .itf4, after which it is doubtful whether White could have defended successfully. 72 ... ltb6 72 . . . l:.b8 was simpler. 73 74 75 76 77 78 <it>g2 ltb2 .i.f6! �1 iic8 lta2 lic6 ..tg5 �e4 l:ic8 :rs a4 White resigns This sometimes happens, and not only in chess: you pass through hun dreds of dangers, but in the end you stumble on easy ground . . . In this fateful position. where for the last move before the time control I had only some 3-4 minutes left on the clock, I was all set to take the h6 pawn, when I suddenly imagined ·that after 72 :gxh6 · Black would win . a .piece by . . . �b8-f8f2+ . . . I was simply unable to notice the 245 Game 329 Ciocaltea-Botvinnik O(vmpiad, Tel Aviv 1964 Caro-Kann Defence 1 2 e4 d4 c6 d5 3 4 lDcJ tDxe4 opponent. 1 1 .i.e3 or 1 1 'i'f3 was preferable. dxe4 i.fS After the 3rd game of the 1 958 Return Match (Smyslov-Botvinnik) it was no secret that in the Caro-Karm Defence I normally chose this continuation. 5 6 7 8 9 lDgJ h4 tDte2 lDf4 ..lc4 11 l:lg8! This relieves Bla6k of any difficulties and allows him calmly to continue his development. Since 1 960 . such a method of defending g7 has become standard in the given situation. Black, of course, will castle queenside, but White too has no other option. .i.g6 h6 lDf6 .th7 e6 12 13 14 i.f4 lDxf4 11d2 14 15 0--0-0 .lxf4 lhbd7 In anticipation of Black's queenside castling, the white queen should not have remained on the d-file. 1 4 'i'f3 was preferable. This position occurred twice in my first match with Tai ( 1 960). In the 5th game (No.275) my opponent played 1 0 'i'e2, and in the 9th (No. 277) - 1 0 0-0. The Romanian grandmaster (at that time still a master) chooses another move� to free his queen from the defence of the c2 pawn. 10 11 i.bJ lDtb5 i..d6 A similar sortie (only, with the other knight, which was nevertheless better) was made by Tai in the 5th game of our match. Therefore the antidote that I found then also proves effective here� it should also have been known· to my 1lc7 0-0-0 16 . . . c5 is now threatened, and White is unable to prevent this attack on his centre. 16 17 lDdJ 1if4 cs White has nothing else; now he gets an inferior ending with an isolated pawn in the centre. · 246 17 . . . cxd4 18 'ifxd4 19 .. .iff4. tDb6 If 1 9 'i'e3 , then 19 . . . tbg4 20 'iie2 etJxf2 ! (2 1 'if xf2 �xd3 or 2 1 lLlxf2 'i1xg3 ). 19 20 • . • 1!fxc7+ 27 . . . J::[h4 could also have been played inunediately. but Black is aiming at a more important pawn -:- b2. 28 .i. xd3 :xdJ cxd3 <t>d2 . 28 . • • llb4 Now 29 . . . a4 is threatened, and if 29 d4 there follows 29 . . . ttJg4+. In addition my opponent had hardly any time left on his clock. The exchange of queens is faint consolation for White. 20 21 22 23 �e3 After 28 i.c4 Black would .have had to be satisfied with the h-pawn. <tJxc7 ltxd3 fld8 White resigns. Game 3 30 Botvinnik-Larsen Noordwijk 1965 Queen's Gambit It is hard to believe that in such an ending it is possible to win in just six moves. To intensify the pressure in the centre, Black transfers his knight from b6 to c5, and his rook will occupy the strongpoint at d4 . 23 24 25 26 1'c1+ l:kJ �e2 ttlbd7 �b6 ttlc5 lld4 It transpires that White cannot defend his h4 pawn (27 l:.c4 :xd3 28 ltxc5 �xb3 or 28 . . J:xg3), which means that it has to be advanced. 27 h5 ,:/ 1 2 3 4 c4 ttlcJ d4 cxd5 4 5 6 7 Ji.gs eJ ifc2 8 9 .i.dJ ttlge2 e6 d5 ttlf6 White is happy to go in for the Exchange Variation, which is by no means so peaceful as it seemed to theoreticians in the past. exd5 c6 ll..e7 For not the first time I was not in a hurry to develop my king's knight, in order to decide this question later, depending on circumstances. 7 0:...0 a5 247 ltlbd7 I employed the more usual 9 tt:Jf3 in Game 342. 12 13 14 0-0 l:ladl a6 b5 .t.n This plan of playing the bishop to f2 is rather unpleasant • for Black. I employed this manoeuvre in the afore mentioned game with Pilnik. 9 • • . h6 Theory disapproves of this move, and rightly so. The drawback to the pawn advance is not only that it weakens tl1e kingside, but also that Black drives t11e bishop to a more favourable position. However, even t11e approved 9 . . . :es followed by . . . lDf8 does not give Black an equal game, e . g. 10 0-0 liJf8 1 1 l::tab l (No.208). 10 .i.h4 :e8 10 . . . tbe8 too has been played, but also without particular success. 11 f3 This plan, where White immediately takes control of the e4 square, was one that I first employed in 1 952 in games with Pilnik (No.203) and soon after wards with Keres (No.208). Thanks to the fact that tl1e bishop is already at h4. it can immediately be played to f2 for the defence of the e3 pawn. 11 14 15 16 17 .i.f5 lLlg3 a3 lhb6 .i.f8 .i.b7 Black ' s avoidance of the exchange demonstrates his desire to complicate the game. c5 The usual continuation. After f2-f3 has been played, Black can attack the d4 pawn without fearing the exchange on c5, since then the a7-gl diagonal is weakeneq. c4 A risky step. Larsen was apparently afraid that after the manoeuvre of the bishop to f2. White could exchange on c5, but this did not present a serious danger to Black. Now, however, the e3-e4-e5 advance is unavoidable, and White gains good attacking chances on tl1e kingside. 18 19 e4 ilh3 g6 There was no point in playing 1 9 .i.xg6 fxg6 2 0 e5� if only because of _ 20 . . . <Ji>h7. 248 19 a5 This was the idea of 17 . . . .ilb7. At the given moment tl\e b5 pawn is immune, since the knight at c3 has to defend the e4 pawn, and Black is able to ·begin counterplay on the queens.ide. 20 21 e5 li)ce2 21 22 f4 25 a."tb4 The exchange on b4. is now necessary; White will need the a�file. 25 axb4 b4 At present it is unfavourable for White to exchange on b4, as Black can exploit the opening of the rook' s file. iDh7 26 li)xf5 26 27 28 i.xf5 ltlg3 To continue the attack involving the advance of the f-pawn, White is pre pared to sacrifice a piece. 22 Black, naturally, aims to weaken the enemy pressure by exchanging bishops. Ac6 Black avoids winning the a3 pawn, fearing f4-f5 . However, this could also have occurred now, even though he is threatening to win the exchange. For example, 23 f5 .i.a4 24 'ibl .i.xdl 25 fxg6 with a strong attack. However, I came to the conclusion that after 24 . . . lt:Jg5 (instead of 24 . . .llxdl ) 25 £xg6 (or 25 .i.g4 �xd l ) 25 . . . ti:Jxh3+ 26 gxh3 'i'd7 Black's chances were better than after the continuation in the game. l:al ifbt .ia4 f5 Trying to hamper White's pawn offensive. 29 il.xd7 30 1ig6+ �xd7 1!fg7 31 32 'flc6 lhal :xal 32 ... 1i'f7 33 :a7 ltJxe5 30 . . .�h8 was somewhat better. After 32 "i'xd5+ 'ilf7 33 'i'xf7+ �xfl the black pieces would have become a little more active. . 23 24 gxf5 1fe7 i.d7 To all appearances, Larsen had plan ned a combination, which he in fact carries out on the next move. The oversight he makes in so doing is easily explained by his severe time trouble. 249 3 3 .. J:le7 would also have lost quickly to 34 :xd7 :xd7 3 5 e6, but 33 . . . tl:JbS was more tenacious. However, after 34 . 'ib6 ttld7 (34 ... Wxf4 3 5 'i'g6+) 35 'i'a5 ! l:e7 (there is no other useful move) 36 f5 (36 . . . tLl.xe5 37 dxe5. and the white rook is defended by both queen and bishop) even so Black cannot avoid defeat. 34 dxe5 42 43 44 45 lhf5 <ifi>fl lhe6 b5 .ci4 But not 45 tDxd4 because of 45 . . . l:.d5. After establishing his knight at e4, White then drives the black knight from e6, after which the game is decided. 'if'e6 45 46 47 48 49 50 51 If 34 . . . 'ifxf4 White has not only 3 5 'i'xe8, but also 3 5 ilg6+ with mate next move. 35 "Ifxe6+ :xe6 36 37 'ifile2 g3 �d3 lhd6 l:lc6 cl lhe4 rs lhf6+ lCid5 e6 lla6 l:l.c7 'ifilh7 lhd8 �h6 l:tb7 lhc6 Now if the knight retreats there follows e6-e7 and e7-e81i'. 51 52 lhe5+ �xd4 Black resigns Game 3 3 1 Trifunovic-Botvinnik Noordwijk 1965 Pirc-Ufimtsev Defence The impression is that Black's two connected passed pawns are just as dangerous as White's. However, it soon transpires that the decisive factor is the inferior position of the black king. 38 39 40 41 bxc3 bxc3 .i.xc5 llal lhcS lhf8 .teJ .tc5 1 2 3 4 d4 e4 lt)fJ cJ g6 Jl.g7 d6 Cautiously played. Usually White chooses the variations with 4 ttlc3 or 4 c4 and 5 lL:ic3 . Trifunovic was quite well placed in the tournament, and therefore he did not object to a draw. It only remains for White to bring his king into play; he, naturally, ignores the h6 pawn. ·· 250 4 5 6 lhbd2 .lc4 7 0-0 lhf6 0-0 lhc6 es 8 dxe5 dxe5 8 ltJxe5 ;#as the easiest way to equalise, buf ;I avoided simplification, which with Trifunovic is especially ' dangerous' - it is always liable to result in a draw. . . . 9 him some compensation for his doubled pawns. llel 9 i.b5 was more active, and in the event of 9 . .i.d7 White could have achieved the desired exchanges: 10 i.xc6 .i.xc6 1 1 tbxe5 .txe4 12 ltJxe4 CDxe4. . . 9 · • • • a6 Eliminating the possibility indicated in the previous note. 10 11 a4 tDn lDb5 'ii x dl 12 :xdl .i.g4 Black had no grounds for avoiding this exchange (now he advantageously activates his queen's bishop). 13 14 15 gxf3 .i.xf4 $.xf3 tDf4 Too straightforwardly played. A post for Black's queen's knight is vacated in the centre, from where it will restrict White's bishop, and the f4 pawn will limit the freedom of the white knight. For the moment there was no objection to 15 <t>h2. 15 16 17 lDd2 exf4 ltab8 .i.e2 Now Black establishes his pawn at b5, after which it will be hard for White to approach the centralised black knight. From this standpoint 1 7 a5 was prefer able. 17 18 19 axb5 l:la6 19 20 ihbJ b5 axb5 White drives the knight to where it is itself aiming. 13 lDe5 b3 A move for which my opponent should definitely be reproached. White not only loses time, but also weakens his h-pawn Meanwhile, after 13 lDe3 i.xf3 14 gxf3 the prospect of invading with his knight at d5 would have given . 25 1 20 . . . b4 could have led to the exchange of knights. At a moment when the white knight cannot reach c4 immediately, this ad vance suggests itself. Now White is forced to advance his c-pawn, which activates the black bishop. 21 c4 llfd8 22 :ds c6 23 24 25 llxd8+ f!b6 c5 l:lxd8 26 27 .D.xb4 ltld2 Black tempts his opponent with the possibility of attacking the b4 pawn. �.f8 .. lta8 The b4 pawn is exchanged for the b2 pawn, which is clearly to Black's ad vantage, since he will be able to activate his rook. .:!a2 %tb8+ lhc4 30 31 .. <it>g2 32 lL\xe5 • l:tal+ l:lel Now White can no longer reach an ending with opposite-colour bishops. White decides to retain his b2 pawn, to avoid increasing the activity of the enemy rook along the second rank. .i.xc5 <l;g7 l:tb7 In chess too . it can happen that the 'better' is the enemy of the good. 2� 30 tDxe5 .i.xe5 3 1 l::tb 7 suggests itself, when if 3 1 . . .llxb2 3 2 l:txb2 i.xb2 the opposite-colour bishops allow White to strive for a draw, while after 3 1 . . . :a l+ 3 2 'iitg2 l:.cl 33 b3 llc2 34 .ic4 .id4 3 5 llxf7+ �h6 36 �xf4 he even stands better. In aiming to force Black to ex change knights, so that the bishop should go to c4 with gain of tempo, White overlooks that it is his bishop that he will have to give up for the knight. The time has now come to exchange a pair of rooks. 27 28 29 Ad4 29 30 After 32 i.fl tt:Jxf3 or 32 . . . �xfl 3 3 'i&?xfl lDxc4 34 l:.b4 lDd2+ 35 @e2 .ic5 the outcome would not have been in doubt. · 32 33 • • • lbe2 ltlxc6 After 33 CiJxf7 :xf2+ 34 �h l <at>f6 it would have been harder for Black to convert his advantage. 33 34 35 36 So, White has SUl-�eeded in attacking the opponent 's centralised· knight, which �bl lL\d8 libs llxf2+ .lxb2 <il>h6 Totally bad was 36 iiJx:f7+ �h5, when the black king will take part in the attack on its · opponent. If instead 36 J:.xf7 l:txf3 37 <ifrg2 :g3+ 38 'it?h2 l:te3, then Black 's material advantage is decisive. 252 52 h6 f3 White resigns 36 37 .. ·:b7 38 . ·l:tb5 <liJg7 <liJh6 f6 39 40 41 'l;g7 Botvinnik-Donner .i.e5 Noordwijk 1965 Queen• s Gambit , In this way Black has eliminated his last problems - with time .pn the clock. lDti+ �b7 lDd8 Game 3 32 � 1 2 3 4 5 6 c4 � �f3 e3 cxd5 d4 lDf6 e6 d5 c5 exd5 thc6 So, by transposition of moves a position from the Tarrasch Defence has been reached. 7 42 lhe6+ 42 43 44 45 46 47 48 lhg7+ lhe6+ %1.xh7 <liJg2 �2 h4 49 50 51 h5 l:g7 : � . <i!i>gt .i.e2 Before compensating for the loss of the f3 pawn by the capture of the h7 pawn, White aims to drive the enemy king to the other wing, but even this does not help. <liJe8 <i!i>d8 �c8 %lxf3 :g3+ g5 :thJ 7 During my adjournment analysis I established that this move is stronger than 48 . . . g4, when there follows 49 l:.g7, and White easily gains a draw 23. g4 flh2+ g3 • • • cxd4 Quite a good method, employed many years ago with reversed · colours by Alekhine. Now the white knight at d4 will be no better placed than at f3 , and Black eliminates the potential possibility of dxc5, which in some cases would lead to White gaining a tempo. 8 253 lhxd4 .i.d6 9 0-0 10 b3 0--0 A sensible move, although b2-b3 has also been: played after the preliminary 10 lbf3 a6. Then Black can immediately develop his queen's bishop at g4. 10 • • • a6 Of course, 10 ... 'i'c7 1 1 tlldb5 ! i.xh2+ 1 2 <iti>hl \ib8 cannot be recom mended for Black, since after 1 3 f4 his bishop ends up in difficulties. However, Gufeld made an interest!ng suggestion 10 ... i.e5. in order after 1 1 .i.a3 to equalise by 1 1 . . . tDxd4 12 �xd4 i.d6. 11 12 13 14 15 .i.b2 llcl .i.f3 1fxd4 'ii'd2 2 1 l::th l tllf5+ 22 'ittf4 - to try and calculate such a position at 'the board is more than dangerous. On practical grounds White chooses a quieter development of events. :e8 .il.d7 lhxd4 .ie5 .i.g4 16 17 18 llfdl gxf3 <iti>g2! Jl.xfJ 11fd7 'i'f5 19 tDe2 .ixb2 20 21 'ifxb2 �d4 l:tac8 1fe5 A vain attempt to build up an attack on the king, which is irrunediately par ried. However, even as the game goes, Black achieves the exchange of bishops and could later have exchanged rooks, which would have given him hopes of a draw in the endgame with queens and knights. The point is that the weakness of the d5 pawn is compensated by the weakening of White's kingside. And yet, as will be seen from what follows, the move played is a loss of time. If 19 . .'i'g6+ 20 t:Dg3 h5 2 1 i.xe5 lhe5, then 22 h4 or 22 <it>h l h4 23 t:De2, and the laright reaches the central d4 square, securing White a positional advantage. . A very cunning move, which forced me to think. Indeed, it was not easy to decide on 16 i..xg4 Jl.xh2+ 17 <it>xh2 ttJxg4+. After 1 8 <tigl 'i'h4 1 9 l:tfel 'ifh2+ 20 <iti>fl iihl+ 2 1 <iti>e2 'i'xg2 22 lbdl tiJh2 Black already has two pawns for the piece �ith a continuing attack, while if 18.:: <iti>h3 'i'g5, then, for · example, 1 9 1id4 'i'h5+ 20 <it>g3 lDh6 Instead of this Black should have returned his queen to d7. to exchange all the rooks (cf. the note to Black's 1 8th move). Now, however. even the ex change of one pair of rooks would lead to White seizing control of the open file. 22 23 :tc2! .:.dct lLJd7 l:l.b8 Best - at least the b7 pawn will be defended. 24 ifc3! The white queen i�r transferred by the shortest way to the kingside. 254 parried all the threats, and there is nothing for White to do but await a propitious moment. 24 25 • • lLJf8 1if6 • it'c7 Without the queens Black would not be able to hold out for long. 26 27 28 lLJe6 fxe6 �ed8 ilg3 tDxe6 Ilc7 Establishing control of all the squares on the seventh rank, so that White should not be able to double rooks there. 29 h4 29 <t>hl �h8 ! 30 :g l l:tg8 was unclear, but 29 f4 ! (29 . d4 30 .l: lc5 d3 3 1 :lg5 g6 32 'i'h3 or 30 . . . h6 3 1 :es with the tlueat of 32 �xe6) followed by 30 'i'g5 would have led to � highly favourable rook ending. After the move in the game the latent threat of ilh3 is no longer there. 32 33 e5 A1c5 30 'i'g5 was essential. Black, of course, avoids the obvious trap 30 . . . d4 3 1 exd4 exd4 32 llf5, and White loses his advantage. 30 31 • • • h5 h6 lid6! The outcome is that Black has 255 d4 lbd4 Of course, not 33 ...exd4 in view of 34 :tf5. 34 35 1fg3 ltc4 :r4 Already White has to parry the threat of . . . l:.f5-g5 (35 . . .l:tf5 36 l::t g4). 35 36 .. 29 30 'i'g4 exd4 . • • :xr4 :rs exf4 Here Black offered a draw, but White nevertheless decided to see what the remaining moves to the time control would bring. 37 38 'i'g4 llb7 b6 After 3 8 iig6 Vxg6+ 3 9 hxg6 :t"6 40 %k8+ :f8 White would have been unable to go into the pawn ending, but now he is threatening 3 9 it'g6. 38 • • • Jlti 3 8 . <iith 7 was simpler. In the queen ending, which Black decides to go into, he encounters definite difficulties. . . 39 l;[xf7 @xf7 The king aims to approach closer to the queenside, to stop the white pawns if necessary. 45 46 47 a4 @e2 1fd5+ 11h1+ h5 @h7 The king again has to move away from the queenside, since after 4 7 . . . @t'8, by exploiting the open position of the enemy king it would easier for White to queen his a-pawn. 48 How should the diagram position be evaluated? It would seem to be time to agree a draw .. After all, based on general growids, Black even has the better pawn structure. But if one 'digs' deeper into the position, other arguments can be found. For example, for his h5 pawn White can win the enemy pawn at a6. Then he has the chance of creating an outside passed pawn. after which pawn endings will be in his favour. As for the blockaded pawns at f2, f3 and f4, they defend the white king quite well against checks and restrict the black queen. Then even the appearance of a passed h pawn will not prove so dangerous. In short. White should play on. 40 41 42 43 44 1ifd7+ 11fc8+ ifxa6 <ltfl @g8 @h7 1!fg5+ ..xh5 1!fd3+ Now Black cannot reply 44 . . . 'i'g6 because of 45 'ire4 ! 'ii'xe4 46 fxe4, when White already has one passed pawn (e) and will soon acquire another (a), which will ·d.ecide the outcome. 44 <it>g8 256 "ifdl! Exploiting the fact that Black cannot exchange queens (his king would be unable to stop the a-pawn), White drives his queen to a passive position. 1i'h2 48 A positional error - the queen is bad ly placed here. Any other move would have given Black better chances. My opponent did not fully appreciate how dangerous his position was. He simply made a move that defended the f4 pawn and ensured the advance of the h-pawn. 49 b4 h4 And at the decisive moment Black overlooks a clever trap. 50 ilfl ! Now the outcome is settled, since 50 . . . h3 5 1 a5 bxa5 52 bxa5 iig2 53 a6 leads to the loss of Black's last trump his h3 pawn. Hpever, there is no other way of releasing the queen from impris onment� and my opponent decides on a desperate step - to include the g-pawn in the play. 7 8 bxc3 d3 9 e4! tnc6 d6 Missing the last chance for . . . e5-e4. In Botvinnik-Basman eflastings 1966 /67) after 8 . . . e4 9 l'Dd4 exd3 10 exd3 LDxd4 1 1 cxd4 d5, although White's position was preferable, Black's resources were far from exhausted. 50 g5 bxa5 51 a5 g4 52 bxa5 53 a6 Black resigns. After 53 . . . g3 54 a7 g2 5 5 'ib 1 + and 56 a8'i' he can even acquire a second queen, but he is mated inuuediately. See what events are poss ible in a seemingly drawn queen ending. ' Game 333 Botvinnik-Langeweg European Team· Championship Hamburg 1965 English Opening �f6 e5 .i.b4 1 2 3 c4 4 5 i.g2 �f'J 0-0 5 6 . ·. • 0-0 :eS ..txcJ tt)cJ g3 • • In the notes to Game 3 22 it has already been mentioned that this move would seem to be the best reply. In the afore-mentioned game White played the weaker 5 a3 (losing a tempo), but 5 e4 is also quite ppssible. . W First 6 . . . e4 is goo� ,:Black has no , reason to avoiq �,� lgical move, now · or subsequently. r·· ···�· After stabilising the position in the centre, White will now prepare f2-f4, which further suggests itself for the reason that Black's king's rook has moved from f8. 9 Ci)e7 After 9 . . . il.g4 followed by the exchange on f3 White also retains the advantage, since he later advances f2-f4. 10 11 • lhh4 lt)f5 lDg6 . This move is part of the manoeuvre lDf3-h4-f5 (and then perhaps to e3), hindering . . . d6-d5 and ensuring the advance of the f-pawn. The exchange l l . . . .ixf5 12 exf5 is wtfavourable for Black, since his b7 pa�p is left en prise. Therefore he bfocks itf ad¥aDce the long · diagonal. ::: � �\ 257 11 c6 12 Ir.bl d5 12 . ..txrs 13 exf5 li:Je7 is still un it is not possible to retain the material advantage. 19 20 21 . . favourable because of 14 Ibb7, but the natural move in the game is also a positional mistake. White gains the opportunity to undouble his pawns, after which the c3 pawn will take part in the battle for the central squares. 13 14 cxd5 c4! • • • dxe4 dxe4 .ixf5 This guarantees Black against the appearance of the enemy knight at d5 or d6, but now White's king's bishop comes into play. 16 17 18 exf5 i.xb7 •xd8 l%d7 :cs cxd5 Black has an unpleasant choice. If he advances his d-pawn. f2-f4 will become very strong and White will have the possibility of playing .ta3� c4-c5 and lbd6. The exchange chosen by my opponent gives . White the d-file and strongpoints at d5 and d6. 14 15 il.eJ .i.fJ g4 t'jje7 llb8 l:lexd8 White is temporarily a pawn up, but Despite White's two good bishops, it is not easy for him to strengthen his position. First he must defend his f pawn, with a kingside pawn offensive in mind. In addition, when the c-file is opened� he will be able to gain control of it. 21 22 23 24 25 l:tb8+ lhc8+ l:tcl 26 27 .i.e2 h3 lbc4 l'lc8 ti)xc8 ti)d6 e4 · l'k5 Now the f5 pawn is defended by the rook and g4-g5 is threatened; if 25 . . . .:ie7 White has the unpleasant reply 26 lk6. h6 Played ·a la Capablanca' - taking account of the opponent's lack of useful moves, White patiently waits, in order to embark on activity .in a more favour able situation. The immediate aim of both sides is the activation of the kings. 258 3 4 tZ)cJ g6 f4 This is the most unpleasant contin uation for Black. 4 .. .i.g7 tZ)tJ 0-0 . It is well lmown that here and on the next move . . . c7-c5 is premature. 5 6 e5 In Oaine 327 White played 6 i.d3 . lLlfd7 6 I did not even have to compare the virtues of this move with 6 . dxe5, which I had to reject, since the captain of the Moscow team had informed me that I had to win this game. .. 27 28 29 �g2 �gJ lDfe8 <t;f8 �e7 One of the black pawns. a7 or e4, is bound to be lost - this cannot be prevented. 30 31 32 :a5 !te5+ i.b5 lDc8 c-.t>d8 It is more useful first to force favourable simplification. 32 33 :xe7 34 ii.xe8 35 <ifilf4 36 i.c5 Black resigns: the be defended, threatened. 7 h4 The logical continuation. The open ing of the h-file will be especially dangerous for Black, since his king's pawn defences will become less secure, and also there are not so many pieces defending his king. l!e7 <l;xe7 'ifi>xe8 <ifild7 e4 pawn cannot and 3 7 .i.f8 is also Game 334 Gipslis-Botvinnik USSR Trades Union Spartakiad Moscow 1965 Pirc-Ufimtsev Defence 1 . e4 2- d4 d6 . lLlf6 7 • • • l£lb6 I played this so as not to allow the white bishop �o go to.·c4, and to �ve the possibility after,. tbe �x��e on g6 of recapturing with the·· f�pawn,·· Theory is 259 probably correct in stating that 7 . c5 is more energetic. Tiris opinion is based on the game Padevsky-Matanovic ( 1 966), which continued 8 h5 cxd4 9 'ti'xd4 d.xe5 1 0 1if2 ·e4 ! 1 1 ltJg5 lLlf6 1 2 hxg6 hxg6 13 1ih4 1"d4 ! (or 1 3 lLlcxe4 Ci)xe4 14 l£ixe4 'i'd4). 12 . . 8 9 10 11 h5 hxg6 .tel l£ig5 fxe5 If 1 2 g4 exd4 13 lDa4, then Black defends as follows: 1 3 . . . ..te6 ( 1 4 ctJxe6 'i'd5). 12 � . · h5! Black supports the position of his bishop at f5 and simultaneously sets up a barrier on the h-file. White, naturally, cannot reconcile himself to this. . .i.g4 fxg6 dxe5 13 14 A critical point of the game. The position is so complicated that it is hard to express a definite opinion on it. Subsequently an exclamation mark was even (although temporarily) attached to 1 1 tllg5 in theoretical guides, but to me it seemed that 1 1 fxe5 tllc6 1 2 .ie3 was more dangerous for Black. White's 'energetic' move has a typical psychological chess context: he thought that he had to act with the utmost speed. .. g4 .i.xg4 .. hxg4 While White was controlling the c4 square, he should have completed his development: 14 ..te3 , then 'i'd2 and 0-0-0. 14 15 ... l£ie6 lbc6 Black ' s position appears hopeless, but this is all merely a mirage. 15 • • • 1t'd7! Sacrificing the exchange, Black com pletes the mobilisation of his forces and seizes the initiative: 16 l£ixf8 l:xf8 17 .i.xf5 'tlfxf5 . White, o f course, declines such a 'gift'. 16 17 18 ..txf5 l£ixg7 .i.eJ llxf5 <;j;xg7 If first 18 i.h6+ <&t>g8 and now 1 9 .i.e3, then this makes things easier for Black: 1 9 . . t:Dc4 20 'i'e2 ctJxe3 2 1 'i'xe3 'i'xd4, and White does not have a queen check at h6. . 18 19 11 ..tf5 Black is obliged to retain this bishop for the defence of his kingside. After l l . . .i.xe2 1 2 1fxe2 tDc6 1 3 dxe5 tl:ld4 14 'ifd3 he would have had problems. • • • ltJc4 l£ixb2 'ife2 During the game I was afraid of the afore-mentioned check ( l 9 . . . ll'lxe3 20 'i'xe3 ilxd4 2 1 'i'h6+), not noticing that after 2 1 . <it>f7 22 1i'h7+ �e6 23 11xg6+ Black replies · 23 . . . @xe5 and remains a pawn up with a safe position 24 260 . . Therefore I was forced to take the b2 pawn, in order at any cost to prevent White from castling queenside. 20 d5 This move is much more tempting than 20 'i'b5, but . . . also hmch weaker. After the queen sortie there could have followed 20 . . . ltJxe5 2 1 'i'xb2 (2 1 dxe5? 'iixb5 22 ttlxb5 l:xe5) 2 1 . . .ttlc4 22 'ifxb7 ttlxe3 (22 . . . 'i'e6 23 lDe4) 23 'i'xa8 ttlxc2+ 24 @d2 'i'xd4+ 25 'ittxc2 .:t.fl+ 26 <it>b3 11b6+ 27 <itic4 ii'a6+, and it all ends in perpetual check. 20 21 ... i.d4 :xe5 After 2 1 dxc6 iixc6 22 .i.d4 2 5 'if xhl + 23 �d2 Black would have won by 23 . . . 'i'h6+ (24 'itte l 1i'h4+), but now his position seems critical. 21 ... ttlc4! This seemingly impossible move enables Black to bring his knight into play, and with equal material (knight and two pawns against a rook) to retain a positional superiority. 22 23 i.xe5+ ith4 ttl6xe5 l:th8 26 1 It is essential to exchange a pair of rooks, in order to ensure the safety of the black king. 24 llxh8 But not 24 l:xc4 lDxc4 25 'i'xc4 because of 25 . . . :hl+. 24 25 26 . 'i'h2+ 'i'f4 . • <i!i> xh8 cl;g7 . This accelerates White's defeat, although after 26 'ilg3 'iff5 27 0-0-0 'i'f6 28 l:thl b5 29 'i'h3 /£jf7 Black's king is safe, and the threat of . . . b5-b4 is hard to parry. 26 27 28 lDdl 'ifd4 'If h3 1ig2 <i!i> g8 Even an attack with a queen and one knight can be very dangerous; here White has to defend against a queen and two knights! 29 'lff2 \ihl+ 11'e4+ 30 <itie2 31 @ft 1i'hl+ 32 <itie2 •xd5 33 lDcJ 1t'd2+ White resigns The need to defend the d4 pawn hinders White's development and forces him to reject the more natural 1 2 'ife2. Game 3 3 5 Botvinnik-Tolush Moscow v. Leningrad Moscow 1 965 12 ll Nimzo-Indian Defence 1 2 3 4 5 c4 �cJ d4 el �ge2 .i.b7 liadl ifil6 e6 .i.b4 c5 Now a variation that can be found in all the opening guides is reached; it is considered to lead to an equal game, although this is perhaps a somewhat premature conclusion. 5 ... d5 Black can consider 5 . . . cxd4 6 exd4 d5 7 c5 ttle4 8 .id2 ltJxd2 9 'i'xd2 b6 10 a3 .i.xc3 1 1 lDxc3 bxc5 12 dxc5 a5 with the positional threat of . . . a5-a4 (Averbakh-Panno, 1 95 8). 6 al .i.xcJ+ 6 . . . cxd4 7 axb4 dxc3 8 lLlxc3, which occurred in my game with Najdorf (No.247), favours White. After 6 . . . .i.a5 7 dxc5 dxc4 8 'ifxd8+ Black's chances are also worse. 7 li:)xcJ cxd4 Regarding 7 . . . b6, see Game 233. 8 exd4 The attempt by White to avoid the creation of an isolated pawn by 8 'i'xd4 proved unjustified (Fine-Reshevsky, 195 1 ) . 8 • . . dxc4 13 14 15 White was now threatening c4-c5. 9 10 11 12 .i.xc4 .i.eJ 0-0 1ld3 lDc6 0-0 b6 . . • l:De7 When I played 1 3 :ad l , I reckoned with the possibility of 1 3 . . . ttla5 1 4 i.a2 'i'c8, when I was intencling to choose 1 5 .i.g5 ! .i.a6 16 'i'h3 i..xfl 1 7 ..ib l with a dangerous attack for the sacrificed ex change. The game continuation is also not altogether successful for Black, since it ignores the positional threat of .i.g5, enabling White to retain control of the central d5 square. Two other variations do not have the drawbacks listed: 13 . . . 'i'd7 14 .i.g5 ltJd5 15 lDxd5 exd5, although after 16 i.a2 followed by ..tb 1 White neverthe less retains the initiative, and, perhaps the soundest - 1 3 . . . h6 followed by . . . ttle7 or . . . lDd5. .i.g5 f4! li:)g6 This is why tbe king's rook had to remain at fl ! WJU.te makes use of an obvious way of developing his attack, 262 on the next move his rook would be captured with a double check, but he was no longer able to deviate from what he had intended. one which I employed back in 1 936 in my game with Vidmar (No.84). He has to hurry, before Black manages to play . . . h7-h6. h6 15 20 21 22 %bf8+ :n+ lllxb2 �xf8 �es 23 il.e6 lld8! 24 25 d5 l:.fi Jlc8 25 26 27 28 29 dxe6 l:bg7 l:xa7 h3 But not 22 ... cJi;e7 because of 23 l:tf7+ cJi;d6 24 lDb5+ cJi;c6 2 5 l:tc7 mate. The best defence: Black forces the advance of the d-pawn (weakening it) and brings his rook into play. Even so, the invasion ·of the seventh rank by the white rook is bound to decide matters. 16 f5! This interposition refutes Black's plan (16 ... hxgS 17 fxg6 clearly favours White). Now not only the a2-f7 diagonal is opened, but also the f-file. After 1 6 .i.xf6 'ifxf6 17 f5 'i'g5 the prospects would have been more favourable for Black than those that await him in the game. 16 17 1s li.xf6 :xrs exf5 11fxf6 llJf4 Probably the losing move. After the exchange of queens it is true that Black avoids the threatened attack, but the resulting ending is hopeless for him. The best chance was 1 8 . . 'ifc6 ! , forcing 19 l:ld2 or eise 1 9 i.d5 11i'd7. . 19 20 lixf6 tDxd3 :xf7 What can Black do? After 20 l:txf7 2 1 �xd3 he remains a p_awn down. Only here, apparently, did Tolush notice that . . . Jlxe6 :d6 llxe6 lDd3 llJf4 Black forces further simplification, pinning his hopes on the drawing potential of a rook ending with a small number of pawns. If . 29 . . . l:tg6 (with the threat of . . . t'.Df4), there would have followed, as in the game, 30 .:ta4. 263 30 31 32 J:la4 &Dxe2 D.a7 tDe2+ ltxe2 It was hard to deny myself the pleasure of again cutting off the king on the back rank. After all, this is a no less significant factor for achieving success than an extra pawn. � 32 33 Jib7 D.e6 Or 3 3 . . J:lb2 34 11h7 (but not "34 a4 because of 34 . . . ltb4 ). and White acquires two connected passed pawns. 34 35 36 Cifi>f2 <li>gJ l:lf6+ D.g6+ <it>g8 37 38 39 40 41 g3 �g4 'ifn's �g6 <ifih5 l:r.c6 <bh8 :cs+ Jlc6+ <i>f3 This makes things easier for White. Black should have continued giving checks while the pawn was still at g2. 41 • • • Game 3 36 Yudovich-Botvinnik Moscow Team Championship 1966 Pirc-Ufimtsev Defence 1 2 3 4 �g8 264 e4 d4 lDcJ f4 g6 d6 c6 There is perhaps no point in making this move before the development of the black knight at f6. 4 5 As soon as the g2 pawn has moved to g3 . the white king is free to advance. The sealed move·. Black resigned, without resuming. The winning plan is not difficult: White places his pawns at a4, g4 and h4, and his rook at a7, after which there is no defence against the threat of g4-g5 h6xg5, h4xg5 followed by l:la6 and a4-a5. JJ.g7 .i.eJ It is already not easy for White to find a sensible continuation. At any event, in the ga91� he did not succeed in , so doing. Possi8l�5 ttJf3 should have been preferred, although then Black has the quite favourable reply after 6 �e3 'ifb6 7 Wd2 ti::\d7 (but not. 8 . . . 'i'xb2 position is quite secure. 5 into consideration. 5 ... 5 . . . �g4, and .txf3 8 gxf3 9 �b l) his a4 also came lib6 In the game Kupreichik-Spassky ( 1 98 1 ) 5 . . . ttJd7 followed by a queenside pawn offensive by Black led to equal play. 6 :bl Not too aesthetic, and it also denies White the right to castle queenside, but what else can be suggested? Exploiting the fact that e4-e5 is ruled out due to the bishop at e3 being un defended, Black immediately makes a pawn thrust in the centre, creating strong points for his pieces. 6 ... 8 9 10 11 gxf5 lhdf6 lhh6 Preparing a possible g2-g4. 11 Jid7 f5 Weaker is 6 . . . e5, when White advantageously replies 7 ttJf3 . But now, instead of this move, the bold 7 e5 dxe5 8 fxe5 i.xe5 9 tDf3 was better, when White has sufficient compensation for the sacrificed pawn - a lead in develop ment and the initiative. The modest continuation chosen by White deprives him of any opening advantage. lD d7 ! 7 lDf3 Black successfully resolves his most difficult problem - the development of his queen' s knight. He directs it to f6, and the king's knight, conceding this, its customary position, successfully comes into play via h6. In the new situation 8 es can no longer be approved (8 . . . dxe5 9 fxe5 t'.LJxe5 I 0 ttJxe5 i.xe5 1 1 tt:Ja4 'if a5+ ), but also the exchange on f5 will merely assist Black's seizure of the e4 square. exf5 .i.d3 1fe2 h3 12 13 0-0 lDg5 'ffc7 d5 14 .i.d2 'fid6 fa..1Jloiting the fact that White, in making a seemingly active move, has taken his laright to the side so that its centralisation (at e5) is temporarily no longer possible, Black sets about occupying the e4 square. 14 . . . t:De4 would be over-hasty because of 1 5 °ii'h5+. Therefore Black parries this future threat, while for the moment defending the e6 square. 15 16 �h2 lhe4 lt)f3 An admission of his error on the 1 3th move. Here 16 'i'h5+ is now pointless in view of the reply 16 . . . 'i'g6. 265 16 17 18 .i.xe4 lDe5 'ifg6 fxe4 .i.xe5! Black succeeds in establishing his 25 knight at f5, which practically decides the outcome. 19 tLlf5 19 26 27 28 11f2 After this Black' s attack develops of its own accord, but what was there for White to do? The impression is that his last chance was 20 g4 iDd4 2 1 'i'e3 (but not 2 1 f5 'ilg7 22 f6 'i'f7) 2 1 . . .ttJf3 + 22 1:.xf3 , but the trouble is that after 22 . . . exf3 he cannot play either 23 f5 i.xf5 , or 23 'ifxf3 flxc2. 20 21 • • • h5 28 29 30 31 32 l:.g8 hxg3+ d4 ti)xg3 'i' xc2 l:lxg7+ l:lxg7 White resigns Liberzon-Botvinnik h4 Moscow Team Championship 1966 Again threatening 23 . . . e3 (24 i.xe3 'i'g3+). 23 24 il.xg3 %%.b4 ltxg3 Game 3 37 Threatening to win a piece by 22 . . . e3 . l:r.gt e3 llag8 Now the attack comes from all sides: vertically, diagonally and horizontally. b4 21 ltJe2 'ii'fl g4 After . . . d5-d4 it would all the same have been impossible to defend the g pawn, so White is not risking anything. White 's counterplay is not only far too late, but also essentially harmless. 22 ..i xc6 · The only result of the opponent's pseudo-activity is that Black's bishop has reached the necessary diagonal. Now it only . remains to open this diagonal and to double rooks. dxe5 Now the f4 pawn will merely cramp the white pieces, and so 1 9 fxe5 l'.Df5 20 'i'f2 h5 2 1 lLle2 would perhaps have been slightly better. 20 bxc6 French Defence .i.el b5 1 2 266 e4 d4 e6 d5 3 lllc3 The threat of d4-d5 is averted, and Black can consider himself to be out of danger. ttit6 1 1 lllxd4 l l . . . llfxd4 1 2 ifxd4 i.xd4 1 3 l:.xd4 ..td7 was good enough for a draw, but why not play on, if the position allows this? ti.Jf6 In this team competition I decided to try a continuation that I had not previously employed. dxe4 .i.g5 lDxe4 i..e7 gxf6 .i.xf6 f5 llJfJ Although 7 . b6 is certainly sounder, 4 5 6 7 . . 8 liJc3 Correctly played. If 8 . c5 White intends 9 d5. . 8 9 ifd2 . .i.f6 c5 Now 1 0 dxc5 or 10 d5 would suit Black. In the latter case he would not continue 1 0 . . . exdS on account of 1 1 liJxd5 .i.xb2 1 2 llb 1 i.g7 1 3 .i.b5+ and 14 0-0, but 10 . . . e5 followed by . . . e5- e4. However, after 10 .ib5+ White would probably have caused his opponent to regret his experiment. Subsequently he can no longer hope to retain an opening advantage. cxd4 10 0-0-0 12 .i.b5 13 lllxf5! i.d7 But now the exchange - 12 . . . i.xd4 would no longer work because of 1 3 i.xc6+ f bxc6 1 4 1i'xd4 'i'xd4 1 5 :t.xd4, when White has an opportunity, after establishing his knight on c5, of gaining some advantage in the endgame. the text move is condemned by theory, perhaps without sufficient grounds. An elegant piece sacrifice, but it leads only to an equal game. 13 exf5 14 1'd6! Only this quiet move confirms the correctness of the sacrifice. Mention can be made of Game 347, in which Liberzon played in similar style. 14 Jle5 15 l::the1 There are many threats: 16 .i.xc6, 16 :xe5+ and even 16 f4; to parry them, 267 Black has to leave his bishop at d7 en 21 22 23 prise, but . . . not for long. . 15 16 . <it>bl! • . 'ifg5+ 23 24 25 0-0-0 • . . ifxc7+ 1ic7 <i;xc7 li)e2 With one move White defends the second rank (25 . . J:td2 26 'ifi>c l) and the g-file (25 . . . �g8 26 l::tg l). Xtxe5 25 26 27 28 29 17 �b8 1!fxc6 Since White is not intending to avoid exchanging queens, it would have been simpler to do this immediately. White too has to find the only move. He would have lost after both 16 'ii'd2 1Wxd2+ 17 l:txd2 0--0-0 18 .i.xc6 .i.f4, and 16 J:td2 0-0--0 17 l:txe5 .i.e6. 16 17 .i.xc6 1fe5+ l:.gl f3 �xgl b3 ltg8 htg4 :xgl+ 'it>d6 <i;e5 i.e6 The whole point is that after 17 . tLlxe5 1 8 t°tld5 Black is mated. . . 18 1lc5 18 19 20 tbxdl t'tlc3 And now the knight is pinned, and White avoids major loss of material, giving up only his extra pawn. ltxdl+ �d8 Although Black's kingside pawns are badly placed. his active king, together with the greater mobility of his bishop compared with the knight in a position where there are pawns on both wings, give him at least equal chances. Of course� not 20 tLle3 (20 . . . 'i'xe3 2 1 'i'xe3 l:td l + 2 2 1i'c l :ixc l + 2 3 ®xc l tDxe5). 20 21 • • • Wfxg2 :et The rook at e5 was not very suitably placed for the CQming endgame� also, the back rank has to be defended. 268 30 31 32 33 @cl .i.d5 <i;d2 b5 f4 a3 <i;dJ . Draw agreed 14 Game 338 Amsterdam 1966 King's Indian Attack 1 2 3 g3 ltlfJ i..g2 d5 c6 i.g4 A comfortable method of develop ment. However, if Black exchanges his bishop for the knight at f3, White' s advantage of the two bishops may tell. 4 5 d3 b3 ll'ld7 .i.xfJ 14 15 .i.xf3 • • l:.ac8 c5 • .tel Now Black' s bishop will occupy a passive position right to the end of the game, and his weakness on the light squares will become very acute. 5 . . . il.h5 leads to more complex play. 6 1fc2 Not the best ( 1 4 l::tb l was stronger), since now Black could have replied 1 4 . . . a5 with the threat of 15 . . . axb4 16 cxb4 .td4, which would have restricted somewhat the actions of. the white pieces. . However, my opponent was enticed by another, unsuccessful idea. Botvinnik-Szilagyi e5 ltle8 thd6 16 17 b5 tLlc4 18 i.g5! 6 . . . e6 7 liJd2 liJgf6 (7 . . . .id6 followed by . . . t:[je7 is also possible. as in a game Gufeld-Vasyukov, 1 972) would have led to something resembling a Caro Kann Defence. 7 8 ltld2 e4 lDgf6 dxe4 An essential exchange: in this way Black restricts the activity of White's king's bishop. 9 dxe4 .i.c5 Played in the spirit of the open games. 9 . . . Jl.e7 was more circumspect. 10 11 0-0 c3 'ffe7 After this move Black· s queenside castling is ruled out because of 1 2 b4 i.b6 1 3 a4 etc. However, he should have castled kingside only after first playing 1 l . ..a5. 11 12 13 · b4 a4 0-0 .i.b6 :rd8 1 3 . . . a5 ·woulditave led to difficulties after 14 b*4§. 4lX:a.§'..fs .:ia3 . 269 Essentially the decisive move. Black cannot reply 18 . . . 'ifxg5 because of 19 tLlxd6 c4 20 'it>g2, and if 20 . . . liJf6 21 liJxc8 ktd2 22 'i'c l (2 1 �adl :bs 22 ttJxc4 is also possible) . If instead 1 8 . . . tiJf6, then 1 9 ttJe3 . Consequently, he is forced to weaken the important a2g8 diagonal. 18 19 : . , ;: - f6 i.eJ.'" i :/• . lflxc4 • . .• Submissively · · ·· (and ·.· without any compensation) allowing the bishop onto tl1e important diagonal. 1 9 . . �h8 was more tenacious. . 20 21 22 23 .*.xc4+ as l:lfdl 1f a2 �b8 il.c7 thf8 lhd1+ After the exchange of rooks it will be more difficult for Black to defend the g8 square, but the exchange could have been avoided only by going totally onto the defensive, which� of course, would have proved unsuccessful. 24 25 26 :lxdl l:hd8 a6 iid8 .i.xd8 White waits. If 27 'iWq2 Black would have defended by 27 . . . 'i'd7, but after the obvious 27 iieZ (with the threat of i.b3 and 'i'c4) he would have had to play 27 . . . f:De6 . (28 .i.b3 lDc7 ! 29 'i'c4 'i'e8), when 28 �g4 CiJc7 29 'i'c8 soon concludes matters. 27 28 . . 'ifd7 thg6 • 1ie2 Now White finally gains the oppor tunity to transfer his queen onto the a2- g8 diagonal, which leads to a rapid finish, but 28 . . . ti.Je6 would not have delayed the end for long. 29 30 31 .i.bJ 1ic4 1ff7 1£Je7 h6 <it?b7 32 33 34 ii.c4 h4 'i'e8 lid6 'fidl If 3 l . . .'i'xb5. then Black loses a piece: 32 Vf8+ �h7 33 'i'xd8 'i'xb3 34 ifxe7. White frees his queen from having to defend the a-pawn. In addition, in the endgame the weakness of the a7 pawn may tell. White's position is of course won, but it is still not so easy to find a concrete way to win. 26 . • • b6 34 After 26. .. i.b6 White wins inuned iately by 27 'i'd2 ! (with the threat of 'i'd5-g8 mate). 27 <li>g2 35 270 f5 In time trouble Black hastens his inevitable defeat (White was threatening Jtf7. h4-h5 and .1Lg6+). exf5 ·· l£ixf5 36 i.g8+ Transposing into this variation of the Black resigns: he is mated in two Sicilian Defence is risky for White. 7 i.d2 and then 8 e3 was more circum spect, and 7 g3 also had its.advantages. 7 moves. • Szabo--Botvinnik Sicilian Defence c4 8 lt)f3 d4 incomplete, this active play looks reck less. But even without this move� as was shown 4 e3 is more accurate, when after the game (Olympia4 Black plays . Alexander 1 954 ), . . . lllli6 and when . . . f7-f5 his position is secure. The whole point is that instead of i..e2 White has played favourable for Black. cxd4 lhc6 lDxd4 lDc2 by Botvinnik 4 . . lDf6 5 d4 the play develops similarly to Grune 309. but in a version more 4 5 6 h4 With the mobilisation of the forces c5 g6 .i.g7 iDcJ .ltlh6 • tage by 8 exf5 .ixf5 9 it)e3 tt:Th.6 1 0 g3 0-0 1 1 .ig2 1'd7 12 tiJxf5. Amsterdam 1966 1 2 3 4 • 7 . . . f5 is less good; in the game Stein Honfi ( 1968) White gained an advan Game 339 · tDc3 . If his king.' s bishop were already at e2, then 7 . . . tbb.6 would be refuted by the well-known reply 8 g4 . 8 9 During the past decades this manoeuvre, suggested by Rubinstein, has undergone a reassessment. h5 f5 fxe4 It is generally considered undesirable to retreat the knight, especially when the bishop at g7 is not blocked. Therefore 6 e3 should have been preferred. 6 Here or d6 • . . on tl1e following move 6 . . . i.xc3+ came into consideration, after which it is hard to decide which are more important tlle defects of White's doubled pawns, or the weakness of - Black's dark squares and the fact tl1at he no longer has the two bishops. In the game Taimanov-Kupreichik 10 ( 1 974) 10 e4 d6 1 � tDe3 b6 was reached· ,, e4 · a balanced position hxg6 A further error ( 1 9 · thxe4 after 7 bxc3 'i'a5 8 .i.d2 ttJf6 9 f3 llJe5 was correct) - the op�ning� of the h-file is useful only to �lack. 27 1 The attacking ii.xh?; ·.t.xh6'::: r1·:hxg6 hxg6 . . attempt 10 12 11'cl is refuted primarily by 12 . . . .i.d2+, �hen White loses the exchange. 10 11 • • • 1£ixe4 hxg6 .if5 Black mobilises his queenside pieces will all possible speed, in order after castling queenside to take control of the open rook's file. 12 /l)cJ Black also retains the advantage after 12 lllg3 'i'd7 13 liJxf5 gxf5 (14 'ilh5+ 00), whereas the tempting 12 . . . •as+ 1 3 .i.d2 'l'e5+ 14 .te2 'ilxb2 would have allowed White to seize the initiative: 15 �xf5 lDxf5 16 llxh8+ .i.xh8 17 l:lb 1 . 12 . . . 'ilas 13 .i.d2 1le5+ 15 'lra4 lhg4 18 lDcdl 18 19 20 11fa5 :ct It would appear that the queen can be pushed back by 18 .�edl , but then 18 . . . .i.xc3 ! , while if . 1 8 lDxg4 there follows 1 8 . . . 1Wxg4. l£id4 l:tbl White would not have been helped at all by 20 ii'xa7, which also does not even threaten perpetual check. 20 21 • . • lhe5 Wc7+ The threats of 2 1 . . . llld3 mate and 2 1 . . .iff3 (and then 22 . . . 'i'e2 mate) do not leave White any other choice. 21 22 23 If 13 . . . 0-0--0 there is the unpleasant reply 14 lLJci5 . 0-0-0 14 /£ie3 llld5+ it)xf4 <l;xc7 'it>d7 g5 The exchange of rooks is un avoidable (if 16 l:tgl ·'i'h2), after which the black rook reaches the first rank, and the outcome is decided. 16 17 �7. .. , , .llxh8 1lb5 !txh8 'lff4 ; , �JJ�:tf.:l;�:·"·":<fi!(�:�} t i «� 8 ·i� �� ·�- � • �� � The queens have disappeared, but the attack has not died away. After 24 i.c3 gxf4 25 .i.xd4 l2Jd3+ 26 <ltd2 �xcl. 27 i..xg7 llxfl 28 'it>xcl e5 Black is effectively a rook up. White resigns. {�J/ . y. ·;t ��i · ·r· �t�w1/'}1'. Y. 272 no means better. But there are also no other expedient suggestions, so that 8 . . . .llb4 certainly has to �e condemned. Game 340 Botvindik-Pomar Amsterdam 1 966 Slav Defence 1 2 3 4 5 6 c4 c6 d5 cxd5 lLlf6 ltlc6 i.f5 lLlcJ cxd5 d4 lLlfJ .lf4 At one time this symmetric way of developing was considered to be the most comfortable way to draw. Then came a period when preference was given to the more cautious. 6 . . . e6. Now it can again be acknowledged that 6 . . . .if5 leads most simply to equality. 7 8 e3 .i\b5 e6 8 ... .tb4 i.xc6+ 0-0 bxc3 bxc6 i.xc3 White can hope to gain a sufficient advantage only by establishing his knight at e5 � then his extra development tempo will tell. 8 . ..liJd7 .is undoubtedly stronger. Black need not fear 9 'ifa4 in view of 9 . l:lc8. when the acceptance of the sacrifice (10 i.xc6 l1xc6 1 1 'i'xa7) is dangerous for White on account of l l . 'i'c8. 1fa5 9 lLle5 . . . . 10 11 12 l:.c8 In the 1 1th game of our return match ( 196 1 ) Tai played 1 2 . . 'i'xc3, but after 13 'i'cl he had to go into an inferior ending (No.290). The continuation chosen by Pomar, · although recom mended by the experts in its time, is by . 273 13 c4! 13 14 ... g4! This pawn, after reaching c5, will be highly unpleasant for Black, but it also cannot be taken, since then he immed iately loses the exchange. Evidently he should have made the prophylactic move 1 3 . . . h6. 0-0 White' s plan is simple: the pawn is advanced to c5, and then by f2-f3 control is established over the e4 square, the bishop at g� is exchanged, and the b-file is occupied. Black utilises the only possibility of countering this plan. 14 15 . . . .lg6 lLle4 Now the exchange on g6 has its dan gers for White, since the black knight at e4 and then at c4 will be impregnable. 16 11 cs f3 �d2 .:n For the reason given in the previous note, weaker is 1 7 1i'e 1 f6 18 tt'lxg6 hxg6 1 9 l:tf2 ltlc4. 17 18 19 20 li)xc4 .i.d6 e4 27 . . . 'i'xc3 28 ktxc3 lld7 29 l:ib3 , or an irresistible attack on his king. Black chooses the second way, since it would appear to be less obvious. In addition, for the moment the material balance is restored and there are more pieces remaining on the board, and hence there is a greater possibility of something unexpected. li)c4 dxc4 l:lfe8 Here we can take stock: the c4 pawn is cut off from its main forces and the white bishop is cramping the actions of the enemy rooks, whereas the black bishop is shut out of the game, and an attempt to activate it leads merely to the opening of the f-file. Even so. my opponent goes in for this continuation: after all. he has to do something! 20 21 22 1ic2 fxe4 f5 fxe4 1fa3 23 24 l:tel :g2 1i'b3 l:tcd8 An important interposition. The black queen, so active just before, is now seriously restricted. 28 It is possible that Black was now intending to sacrifice the exchange. but even this could no longer help. l:tg3 1fxc4 11c3! 1ixa2 :g2 1fa6 Of course, not 28 . . . 'ifa4 because of 29 :.a1 1fbs 30 l:.b2. A t)'pical example of how useless is the activity of one piece. 25 26 27 27 28 'ifh6 'if d2 Pomar now faced an unpleasant choice: a hopeless endgame after 29 h4 lld7 30 31 h5 rlal il.f7 After 29 . . . h5 there could have followed 30 1i'f3, if there is nothing better, while if 29 . . . h6 30 h5 .th7 3 1 g5 . The e4 pawn no longer needs to be defended. and White exploits an oppor tunity to drive back the enemy queen even further. 274 31 11i'c8 32 1ff3! Game 341 This move resolves any remaining doubts. 32 33 Botvinnik-Zuidema Amsterdam 1966 Grtinfeld Defence Vd8 g5 1 2 3 4 5 lDf3 g3 .i.g2 0-0 c4 lDf6 g6 i.g7 0-0 d5 6 7 8 cxd5 d4 dxc5 lDxd5 cs Had he wished. Black could still have maintained the symmetry. 8 e4 is hardly stronger, but it leads to more interesting play (cf. Game 170). · The advance of the kingside pawns forces Black's capitulation. 33 34 h6 g6 e5 35 i.xe5 llb7 3 5 'i'f4 and 36 lie5 was threatened. Or 35 .. Jhd4 36 .l:txa7 l:td7 37 :xd7 'Wxd7 38 'iWf6 ifdl + 39 'it>h2 'i'h5+ 40 @g3. 36 37 38 iff4 38 39 40 41 cxd5 l:ta7 c6 c7 lie7 i.d6 Black resigns :n d5 This breakthrough attack. 8 9 c6 ti)a6 Theory gives preference to the popular 9 ltJgS, but to me the move played seems more thematic, since sub sequently the black c-pawn may be both a target. and a barrier to its own pieces. 9 10 lDbd2 bxc6 a5 i.b3 concludes the Mate is unavoidable (even if the queen is given up). Another. perhaps even more active plan for White consists in the rapid · 275 transference of his pieces to the queenside. For instance, 10 ttJci4 i.b7 1 1 lDa3 'i'b6 12 tl'lb3 llfd8 1 3 ii.d2 'Wic7 14 4Jc4. 10 • • • 'i'c7 A poor move� in some cases White's queen' s bishop will occupy f4 with gain of tempo. 1 0 . . �b8, while suffering from the same drawback, would also have had an advantage - an attack on the b2 pawn. For example� 1 1 lbc4 liJc5 with an equal game, in Boleslavsky' s opinion. Other possibilities were 10 . . 'i'b6 and 10 . . . li:Jc5 1 1 tiJc4 i.a6 12 lbfe5 'JJic7 1 3 tl.\xc6 i..xc4 (or 1 3 . . . e6) 1 4 .i.xdS il.xdS 15 'i'xd5 :res 16 'i'xc5 'i'xc6 17 'ifxc6 l:hc6, when Black has definite compen sation for the sacrificed pawn. . . 11 • • • · · ··· · : 17 e4 18 .i.f4 �b6 Preventing (which · could have fol lowed after l 7., .tl)f6) 1 8 i.a5 I:txd l + 1 9 l:!xd 1 . when 1 9 . . . .txe4 fails to 2 0 1ixe4 'ifxa5 2 1 1ib7 26, while after 1 9 . . . 'i'c6 20 tiJd6 :lf8 2 1 �e5 White' s initiative increases. 1fc6 Black decides to give up the ex change for a paw� after which the placing of his pieces is improved, he occupies the centre, and the battle flares up with renewed strength. The follow ing line was unpromising: 18 . . l:.xdl + 1 9 l:txdl 'i'c6 20 liJd6 l:f8 2 1 ctJe5. . 19 20 21 22 a3 Securing the position of the queen at c2 and restricting the mobility of the knight at a6. 11 If 16 . . . fS, then 17 e4 would have been even more unpleasant. �d6 :xd6 1fxe4 tl)gS :xd6 1fxe4 .txe4 If 22 l:.adl Black has the adequate defence 22 . . l::tf8 . . 22 c5 • • • i.d5 Here the pawn becomes more active, but it takes away another square from the ill-fated knight at a6. However, Black also does not escape from his difficulties after l l . . . tt.Jc5 12 'i'c2 liJe6. 12 13 14 tl)c4 'li'c2 �dl .lb7 nac8 14 e4 was premature: 14 . . . tiJb6 1 5 .i.f4 li'd7 (or 1 5 . . . \ic6) 16 l:.ad l 'i'e6. 14 15 ... i.d2 rl.fd8 tl)b8 Black naturally tries to bring his unfortunate knight into play, but now White provokes a weakening of the d6 square. 16 i.h3 e6 Black has two active bishops, the white rooks are separated, and at first . sight it appears that White's position has become critical. 276 23 A careless move, which makes it harder to win. 34 i.a3+ should have been played, to retain the bishop and force the enemy king to retreat to f7. lhxe6! A continuation that White had to have in mind when he played 17 e4. The whole point is that the black knight at b8 is left undefended. After the ex change of one pair of rooks, the other white rook invades the black position, which is of great importance. 23 ... 34 fxe6 Or 23 . . . .i.xe6 24 :xe6. 24 :xd5 25 .i.xc8 26 . .i.xb8 27 l:tbl! exd5 tbxc8 i.. xb2 27 28 29 30 .i.xa3 c4 .i.f8 .i.d6 :xa7 :.a6 ffiff!2. WP l:i1e7 ct;n <li>e6 .i.g7 '%\'.!{ • �d7! 35 36 37 .i.xe7 :a7+ .:c7 r;l;xe7 <it>f8 37 38 ... <it>e2 c3 38 39 40 <ia>d3 g4 Again 3 8 g4 should have been played. Neither White nor Black had yet appreciated all the subtleties of this endgame. It is this that explains the errors committed. Now if necessary the white king can stop the passed pawns, freeing the other pieces from the need to do this. 30 31 32 33 • The proximity of the time control has its effect. The immediate g3-g4-g5 was stronger. with a guaranteed win. White must occupy the seventh rank with his rook as soon as possible. llb7 .i.e5 <ilifl • Black is threatening by 3 5 . . .tt:lc6 to restrict the white pieces and to begin advancing his passed pawns. Therefore White is forced to exchange his bishop. h6! d4 r� %2/ lff. {'.�;i �;�;{<i�f���:/%.��,:;,; ?. i d •@;jg_t: .:?§J;g 34 f4 f1t1l�;���:r��,}''-B'fil 40 • • • i.f6 Black misses the only saving chance, which was 40 . gS, when after 4 1 f5 .i.f6 .. 277 it would have been much more difficult (if at all possible) for White to breach the opponent's defences. 41 h4 c.fi>g8 Of course, not 4 1 .. . .ltxh4 42 ®xd4, when the white king has freedom of action. 42 43 �e4 g5 . hxg5 'ittd3 llc6 �e4 .i.h8 @g8 �h7 50 f5! gxf5+ Or 50 . . . d3 5 1 <it>xd3 gxf5 52 g6+ <tlg7 53 �xc3 . 51 �xf5 i.g7 Again the pawns have to remain where they are: 5 1 . ..d3 52 g6+ '1;;g7 5 3 :xc3. 52 lte8! ! This fantastic move leads to a new, but more subtle zugzwang. After 52 . . . .th8 53 g6+ <tig7 5 4 ilc8 Black is bound to lose material. But the advance of either of the pawns also leads to its loss. The rook has to be on the e-file, from where it can threaten both the black king and the pawns. 52 By giving his opponent the move, White forces the black bishop to occupy the comer square . . Then the king is also driven onto the rook's file. which allows White to create mating threats. .tk7+ l:r.c8+ Now the time for decisive action has arrived. .ll. g7 hxg5 This was an IBM tournament, which later became a traditional Dutch festival. Flohr and I were living in a motel by the way out from Amsterdam to The Hague (this building has now been demol ished). Not far away was a modem church, in front of which was an auto mated bell resting on four columns. It was in this church that we played. Since the game was wrlinished, I also had to analyse it in the 'tournament building', since the dinner break was only short. Apparently I was helped by God him self: in some forty minutes I completed one of the most subtle analyses in my life. If 43 . . . h5 I had prepared the follow ing variation: 44 ltc6 ®fl 45 ®d3 i.h8 46 l:tc7+ �gs 47 l:tc8+ 'lt>h7 48 :as �g7 49 l:.a6 i.h8 50 f5 gxf5 5 1 l::th6+ <&t>g8 52 �xh5 il..g7 53 g6 etc. , while if 43 . . <it18 I would have played 44 l:c6 �fl 45 h5 ! 44 45 46 47 47 48 49 • • • d3 If 52 . . . c2, then on the 54th move the rook retreats to the second rank, instead of the third. 278 53 54 g6+ l:[eJ ' 'it>b6 .i.d4 Or 54 . . . �e5 55 .:t.xe5 <Ji;g7 (otherwise 56 !:te7 and 57 l:th7 mate) 56 l:e7+ �g8 57 �6. l:.ab 1 and a rapid pawn offensive on the queenside. However, White decided to make use of an old idea of Marshall, which he employed in 1 925 against Rubinstein in Moscow. c2 55 llxd3 </iig7 56 ltb3+ 57 l:b7+ Black resigns (he loses his c2 pawn). But about half an hour later someone suggested that the ending with the pawn at g6 was drawn. I merely replied that I had once seen Smyslov win such an ending against Simagin. A little more time passed, and everyone calmed down. In an endgame book the way to win was found. Game 342 11 Botvinnik-Robatsch Amsterdam 1965 Queen's Gambit 1 2 3 4 5 6 7 c4 lt.Jc3 d4 cxd5 i.g5 e3 1ic2 tDf6 e6 d5 exd5 c6 Jle7 8 9 .i.dJ lllf3 ttlbd7 12 0--0 It has long been known that 7 . . .tbe4 is unsatisfactory for Black because of 8 .i.xe7, when if 8 . . . 'fixe7 9 ctJxd5, or 8 . . . �xe7 9 tbxe4. In Game 3 30 White developed his king's knight at �?. 9 10 11 0--0 :e8 ltJf8 ltael The more usual plan involves 1 1 . . . ll:le4 A later game Keene-Csom ( 1 97 1 ) went 1 l . . . g6 12 ltJe5 tbe6 13 .i.116 lt.Jg7 14 f3 .tf5 1 5 i.xg7 .i.xd3 16 ltlxd3 ¢lxg7 1 7 'ii'f2 with quite good prospects for White. i.xe7 During the game I came to the conclusion that the familiar idea of 1 2 1'.f4 f5 1 3 ltle5 followed b y f2-f3 would not give White anything in view of the reply 12 . . . .i.fS ! Therefore there was nothing else, other than to play as in the Marshall-Rubinstein game. 12 13 14 .txe4 l£id2 9xe7 dxe4 Black faces a difficult choice. Rubinstein played 14 . . . f5 and after 15 f3 exf3 16 tLJxf.3 i.e6 1 7 e4 fxe4 1 8 l::txe4 he ended up in a difficult position. Robatsch thought for about a hour, but chose a move that was hardly any better. 279 Black is unable to intensify the attack on the d4 pawn, since his knight cannot be .released from the pin. 22 14 • . . • • . fxe4 b6 For the moment the e4 pawn is immune: 15 . . . .i.a6 is threatened, when Black wins the exchange. 15 1!fa4 f5 But this is inconsistent. If Black was intending to make this move, there was no reason for him to weaken his queen side by . . . b7-b6. Now 1 5 . . . b5 16 'i'a5 would not have improved his position, but some practical chances were offered by a pawn sacrifice: 1 5 . . . .i.b7 16 t2Jdxe4 c5. 16 f3 The capture of the c6 pawn was ruled out by 16 . . . .i.a6. 16 17 . • • /i)xfJ exfJ .i.b7 This move loses two tempi, since the bishop has to guard the f5 pawn, as soon becomes clear. 18 19 /i)eS 11'c2 1fe6 .i.c8 Black's hopes of easing his defence somewhat after 23 tDxe4 'iid 5 are refuted by a reply that comes as a surprise to him. 23 23 e4 lidl 1i'b3 1i'd6 llle6 • • • a5 Black can also, of course, not pay any attention to the opponent' s threats, but to what does this lead? 24 tDxe4 Black resigns. If 24 . . . 1i'd5 there After 1 9 . . . g6 20 e4 Black's kingside would have become irreparably weak. 20 21 22 IUi! This activation of the rook im mediately decides the outcome. After 23 . . . lDg5 White wins by 24 t2Jxe4, while if 23 . . . ltJxd4 24 %lf8+ or 23 ... l:tf8 24 :lx:f8+ <2i>xf8 25 ttJxe4 'i'd5 26 'i'f3+. Nothing is also changed by the attempt to divert White from his attack: 23 . . . e3 24 ltJe4 'W'd5 25 'i'xe3 t2Jxd4 27 26 thf6+ gxf6 27 'i'g3+, and mate in two moves. follows 25 I:txg7+ ! �xg7 (25 . . . ltJxg7 26 tt:Jf6+) 26 1i'g3+ � 27 :n + with unavoidable mate. 280 This decisive is hard to approve. Smyslov was probably ex1Jecting the reply 9 gxf4, when there would have followed 9 . . . d5 10 cxd5 ttJxd5 1 1 !Dxd5 'i'xd5 with a good position for Black. However, White, of course, avoids this thereby placing his continuation, opponent in a difficult position. 8 . . . tiJh5 came into consideration. Game 343 Botvinnik-Smyslov USSR Team Championship Moscow 1966 English Opening 1 2 3 4 c4 lLlcJ gJ i. g2 e5 d6 ll:lc6 ll:lf6 Usually in this opening (the Closed Variation of the Sicilian Defence with reversed colours) Black develops his king' s knight at e7. 5 d3 g6 If 5 . . . i.e7 White could have contin ued 6 :b l and then b2-b4, as I played against Kostjoerin (Leipzig 1 960). 6 f4 White makes this move unusually early� which has a certain point. If he first developed his king's knight at f3, he would then have to waste a tempo to revert to the chosen plan. i.g7 6 0-0 7 lLlfJ 8 . 9 .i.xf4 .i.g4 Instead of this, 9 . .. d5 was hardly suitable on account of 1 0 cxd5 lDxd5 1 1 tDxd5 Wxd5 1 2 'ii'd2 ! , when Black has difficulties over the defence of his c7 pawn, e.g. 12 . . . iDd4 13 <it>hl ! However, 9 . . . h6 suggests itself. 10 1i'd2 l:te8 With the f-file half open it is risky to move the rook from f8, since the f'7 pawn may come under attack. 11 :.aet :.bs Such a move is, of course, a waste of a tempo, although it is made so that the knight at c6 should not have to cover the b7 pawn against a possible attack along the long diagonal. 12 0-0 .ig5 Creating a highly unpleasant pin, which it will not be easy to get rid of. If, say, 12 . . . Wd7, then 13 .i.x:f6 .i.x:f6 14 lLlg5, and White sets up dangerous pressure on the f-file. Therefore Black decides to eliminate the knight at f3, so that it will be unable threaten the f'7 pawn. 12 13 8 exf4 . .. ..fl.xfJ i.xtJ lLle5 The subtle point of Black's move is that now he is controlling g4, and there fore if 14 tiJd5 he can reply 14 . . . l:e6 ( 1 5 i.g4 is not possible). Had he played 28 1 1 3 . . . CDd4. then after 14 tLldS the threat of 1 5 i.rl6 i.xf6 16 tLlxf6+ 'I'xf6 17 i.c6 would have been highly un pleasant. 14 1if4 .i.xh6 1ixh6 .i.xh6 c6 Gaining control of the d5 square, freeing the rook from having to defend the b7 pawn, and also threatening a queen check at b6 - what more can be ex-pected from one move? 17 18 <ifi>bl 1id2 b3 d4 e4 22 <ifi>gl 22 23 :e2 <tig7 1ih8 l£ieg4 9'b3 h6 The pressure on the f-file has become so dangerous, that the attack on the knight at f6 can be panied, only by sacrificing a pawn. The simple 15 .i.g2 was threatened, and 14 . . . l:.e6 would have been met by the decisive 15 .i.d5. 15 16 18 19 20 21 Covering in good time the rook at fl, and also the g3 pa� which would have been lUlder attack after 22 . . . l:.h8. White does not exchange his bishop for the knight at g4, which would merely strengthen the position of the other knight on this square and deprive him of his bishop - an important defensive piece. But when the knight at f6 is diverted to another square, this ex change may prove useful. %le6 Although it is not very pleasant, White has to retreat his queen, allowing 18 . . . <itig7 followed by the activation of the black pieces on the h-file. To begin advancing his central pawns (and this is White's only active chance) it is neces sary that the queen should subsequently defend the d3 square and the rook at e I . 282 • • • lth8 Now it is not apparent how Black can strengthen his attack on the king, and he is still a pawn down. His attempt to win the central e5 square for his knight leads merely to simplification, after which White's advantage becomes obvious. 23 24 25 dxc5 lDd5! c5 dxc5 rook) 32 d6 White wins easily. His king should be at gl, whereas in the game it moves to g2, and the battle is greatly prolonged. By threatening to win the exchange with 26 tDf4, White forces the exchange of all the minor pieces. 25 26 .i.xg4! tDxd5 A tactical subtlety, as a result of which only the heavy pieces are left on the board. 26 27 28 cxd5 1fc3+ 28 29 �d4 1fxc5 After 29 l::te7 ltf8 30 'i'xc5 @g8 White's second rank would have been insufficiently well defended. 29 30 31 'if xd4+ 1ixe2 <it>g8 :n Stronger (as shown by Averbakh) was 3 1 h4, when after 3 1 . . . l:.hS (otherwise Black cannot activate his �g2 32 33 d6 \tel+ It was not yet too late to repeat the position by 32 :n 'i'e2 and to make the accurate move 3 3 h4, but the truth was still not apparent to me. 1fxg4 :xe4 This is the whole point. 2S . . . @gS loses immediately: 29 !txe4 Wxe4 30 :te 1 'ifd4+ (otherwise 3 1 :es+) 3 1 'i'xd4 cxd4 3 2 :es+ <j;g7 3 3 ltxh8 @xh8 3 4 d6. Therefore Black is forced to block the check with his rook. 31 32 llh5 33 1i'e6! A s�btle move, the strength of which was underestimated by White when he played 29 'i'xc5, and later when he was considering 3 1 %tf2. Now he cannot play 34 d7 because of 34 . . . 11fd5+!, when he loses his passed pawn. This is why the white king should have been at g 1 , which would have been achieved by 3 1 h4. Had Black sought salvation in a queen ending, then after 33 . . . l:tf5 (instead of 33 . . . We6) 34 :xrs gxf5 3 5 d7 'i'e2+ 36 �h3 'iibs+ 3 7 'i'h4 he would have had to resign. 283 34 l:td2 1ifd7 35 36 37 38 �gl lk2 :.c7 1!ff4! b6 . <it>h7 'iff5 <&tg3, and the problem is solved very easily. Forcing the exchange of queen, since 38 . . . 'irb l + is hopeless in view of 3 9 :!c l . After the exchange of queens White gains a big material advantage in the rook ending. The passed d-pawn guarantees him a win. 38 39 40 41 gxf4 .:xa7 d7 <it>d8 43 Black wants to eliminate the possible operation 44 d81ll+ <it>xd8 45 l:xf7, but he leaves undefended the d6 square, which his rook will be unable to control. 43 . . . J:ld2 was more tenacious. 44 <l;e4 l::th5 Black attacks the h-paw� while continuing to defend the fifth rank. If 44 . . . ltd2 there could have followed 45 h4. lli'xf4 cilg7 ltd5 <it>t'6 45 46 a4 <l;e5 llxh2 Again the best. After 46 <i;d5 l:e2 Black could have continued to resist, whereas now. in view of the threat of 4 7 <t1"6, he has to agree to tlle exchange of his g7 pawn for the d7 pawn, i.e. go in for tlle continuation that he avoided on tlle 43rd move. 46 47 48 49 Here the game was adjourned. After it was completed. Smyslov suggested that 42 a4 was the simplest way to win� but White preferred to improve the position of his king (in accordance with Capablanca' s principles). I rejected 42 l:tb7 because of 42 . . . <it?f5, when the black king is too active. 42 <it>f2 <lJe7 43 <i;f3 Best. After 43 <it>e3 f5 it would have been hard for White to penetrate witl1 his king to h4. But if now 43 . . .f5 44 d8'if+ l:txti <it>d6 '1;e7 <i>xd8 %lb2 The white pieces dominate the board. The loss of the b6 pawn is unavoidable, 284 e.g. 49 ... �c8 50 @c6 <it>d8 5 1 !td7+ ©c8 52 l:td3 :f2 53 l:d4. 49 50 • • • l:te7+ 7 <i&>e8 <i!i>d8 Or 50 . . � 5 1 l:te3 :f2 52 �e4. . 51 52 53 54 55 l:td2+ l:tb7 �c6 11�4 l:.xb6 l!xf4 @e7 llb8+ a5 Black resigns 7 Game 344 USSR Team Championship Moscow 1966 English Opening c4 lDcJ ti:)f3 • • • c5 In the event of 7 . . . d5 8 cxd5 ltJrg4+ 9 ®g2 1'f6 1 0 1ie2 � (or 10 . . . l£ld3 1 1 h3 ttJgf2 1 2 'i'e3) 1 1 ifxf2 �h3+ 12 ®g l ttJf3+ 1 3 'ifxf3 'i'xf3 14 i.xh3 White · has three pieces and a pawn against a queen, which constitutes a clear material advantage. In the game the play becomes positional in charaeter. In this case the two bishops and the possibility of exploiting the half-open f-file make White's position preferable. However, Black still has adequate resources for counterplay. Botvinnik-Keres 1 2 3 e4 It is essential to forestall . . . d7-d5, after which the position would be opened and the white king's situation could become dangerous. lhf6 e5 8 9 d3 h3 d6 Since 3 g3 c6 was a favourite contin uation of Keres, White decided to take a different course. 3 4 • • • lhc6 .i.c5 gJ An interesting possibility, but it would seem to lead to a difficult game for Black. The well-tried 4-:-. d5 is sounder. . 5 lhxe5 A well-known exchanging combin ation, which probably favours White. 5 . • . i.xf2+ If Black simply plays 5 . . . tt:Jxe5 6 d4 'JI e7, then 7 dxe5 (but not 7 dxc5 tbf3 mate) 7 . . . ltJg4 (or 7 . . . 'flxe5 8 .ltg2) 8 e3 tt:Jxe5 9 Ag2 (9 . . lDxc4 10 'i'g4) with a slight but tangible advantage to White. . · 6 �xf2 ltlxe5 285 9 h5 A nervy move. Simpler was 9 . . . 04>, as Uhhnann later played (against G. Garcia in 1973), gaining a satis factory position. 10 i.e2 ltlh7 11 'it>g2 It was dangerous to play 1 1 i.xh5 t'.Dg5 with an ·attack for Black. h4 11 A dubious undertaking. It is true that Black gains the g5 square for his knight, but the h4 pawn will constantly need defending, making it hard for him to castle. 12 13 14 lDg5 .i.d 7 g4 i.e3 'i'd2 A routine move. With 1 4 'i'g l followed by iff2 White could have hmnediately switched his queen to an active position, from where it would have simultaneously controlled ... b4 lDe6 kingside, White takes measures to make tDg5 because of 23 bxc5 dxc5 24 i.a4+ b5 25 �xb5 (25 . . . tLlxb5 28 26 .ixb5+ '3;e7 27 .i.xg5). a5 22 a4 A highly risky step. In the forth it also impossible on the queenside. • • b6 . The acceptance pawn course, sacrifice of the unfavourable coming attack on the black king, the open b-file will be used by White. temporary ( 1 5 . . . cxb4) was, because Jl.c6 Black was unable to carry out the planned manoeuvre 2 1 . . . ..i.xdS 22 cxd5 Seeing as it is hard for Black to castle 15 @h2 21 three important squares : d4. h4 and f7. 14 15 21 Preparing an attack with g4-g5 on the black king stranded in the centre. of of 16 t'.Dd5. when White gains control of the Perhaps things would have been a little easier for Black after 22 . . . i.xd5 23 cxd5 central d4 square and easily wins back tDg5. 23 24 his pawn. 16 17 18 An :abl l:thfl @gt unsuccessful .i.c6 i.b7 ltX6 regrouping: bxa5 lla7 bxa5 1ff2 Defending the weaknesses on the kingside. 25 the 0-0 g5 The only move. If 25 . . . fxg5, then 26 knight was better placed at e5 than at i.xd4 cxd4 (26 . . . ltJxd4 27 ..th5+ !) 27 d4. 19 20 lDd5 i.dl lDcd4 f6 'i'f5� and White' s attack is irresistible. 26 Black intends to exchange bishop for knight ( . .. .ixd5), and for this he secures a base at g5 for his knight from e6. g6 . f5 Overlooking a veiled tactical blow by the opponent. 26 . . . .ieS 27 'i'g2 f5 28 exf5 tbxf5 2 9 ife4 was more tenacious, 286 other than my opponent). However, on that move I made a decisive oversight. And so, from the sixth round of the Team Championship, here is my en counter with the challenger. although even t11en Black's position was probably lost. . 1 2 3 4 5 6 7 e4 d4 tt)cJ tt)xe4 tt)gJ h4 ltlf.l c6 d5 dxe4 .i.f5 .i.g6 h6 In Games 275, 277 and 329 one can follow the course of events when White develops his king's knight at e2 (or h3). Here 7 f4 is also played. 27 l:tb8! Black resigns. After 27 . . . 'i'xb8 28 7 8 9 'i'xh4, mate is inevitable. h5 .i.dJ lild7 .i.h7 9 .i.c4 e6 10 'ife2 tiJgf6 leads to an equal game. Game 345 9 10 Spassky-Botvinnik USSR Team Championship .Moscow 1966 Caro-Kann Defence There are. of course, always two participants in a match for the World Championship: the champion, and the challenger. One has been victorious in a match, while the other has overcome all the grandmasters in the qualifying com petitions. It is the dream of every player to meet them at the chess board, and this is what happened with me in the USSR Team Championship. I ·will give one of these games. In the second, with World Champion Petrosian (on the 1 1 th day of uninterrupted battles�. I had even more grounds for winning {the way to win on the 34th move was pointed out by none • • . .i.xd3 Vxd3 This variation of the Caro-Kann Defence was a favourite· weapon of Spassky. Tastes differ, but does a con tinuation deserve such attentio� when over a period of roughly 50 years only two insignificant changes have occur red? These are that they began advancing the white pawn to h5 (instead of leaving it at h4) and after queenside castling they began placing the queen at e2 before playing �b l and c2-c4. I forgot about this latter innovation and carelessly played . . . �gf6 too early (instead of . . . 'i/c7). After this Black got into difficulties, since White was able to establish his knight on e5 . However, this entire variation was quite well known to me: back in 1 928 I 287 happened to be the commentator on a game Grigoriev-Panov, where this same opening occurred. li)gf6 in tt:Jc3 f6 1 9 exf6 i.xf6 2 0 'i'c4 Wb6 2 1 b 4 tLla6 2 2 ttJe4 h e eacountered definite difficulties. This was :one . of the reasons why I chose a different continuation. In addition, it involves a cunning positional idea. 16 f4 c5 17 c4 tllb4 18 ·i.. xb4 18 19 20 21 .U.xdl lDe4 illd6+ Forcing events. White cannot allow . . . c5-c4 (this manoeuvre in fact occur red in the afore-mentioned Grigoriev Panov game) . 11 i.. d 2 Here 1 1 .i.f4 has also been played, but then 1 1 . . . 'i' a5+ 1 2 .id2 'JJkc 7 was found, leading to the same position that is reached in the game. Later White tried to improve on this variation with 1 2 b4 'i'xb4+ 1 3 c3 'ib5 14 c4 ifa5+ 15 ii.d2 'i'a6 16 0-0 (Velirnirovic-Hort, 1 97 1 ), but it is not clear whether his positional advantage compensates for the sacrificed pawn. 11 12 13 14 0-0-0 1fle2! lDe5 'ffc7 0-0-0 e6 tDxe5 Events develop differently after 14 . . . ctJb6 ( 1 5 ..ta5 l:t.d5), as in a game Romanishin-Bagirov ( 1 978), with quite good play for Black. 15 dxe5 The simplest, . but perhaps not the strongest continuation. After the black pawn takes up position on b4, it will be hard for White to find a comfortable shelter for his king. For this reason, if the queens are on Black will always gain counterplay. 1 8 <t>b 1 should have been preferred. lDd5 1 5 . . . ttld7 would seem to be more natural, but not long before our game, Petrosian played · this against Spassky, and after 16 f4 i..e7 17 ltle4 ttlc5 1 8 l:txdl+ cxb4 i.e7 'ita>b8 At first sight White has an over whelming advantage, but this hardly accords with the truth. · The c4 pawn is 288 Spassky played so quickly at this poin� that two obvious conclusions suggest themselves: he judged the now inevitable exchanges to be in White's favour, and at the same time he was afraid of giving the impression that he had not seen Black's last move. These factors probably prevented him from considering the most sensible decision and returning his knight to d6. Now White has to play an endgame a pawn down. weak, and the position of the knight at d6 is insufficiently secure. 22 �xf7 22 23 24 25 lbd6 g3 'ifg4 If 22 g3 Black could have replied 22 . . . f6, while now his pieces are activ ated as a result of the exchange of the ii pawn for the f4 pawn. Although White has simultaneously attacked two pawns, both threats are easily parried. 25 .. 'ifd7 26 'ii?b t Ji.gs 27 ltlb5 • 26 fixg7 is not now possible (26 . . . i.g5+). In addition. Black is intending to play 26 . . . 'ii'a4 and 27 . .. i.g5+. 28 29 30 ifxdl <3i>xd1 llxdl 'ifxdl+ .i.e3! 31 32 'it>e2 b3 .i.cl .i.b2 33 34 35 lbd6 0e4 g4 .i.xe5 'ii?c 7 35 36 37 'ii?d3 cxb5+ White has to be ultra-attentive. For example, 3 1 b3 .if2 32 g4 JL.c5 33 <it>e2 a6 34 tiJd6 .txd6 35 exd6 b6 36 'it>d3 <3i>b7 37 'ii?d4 <it>c6 38 'ii?e5 'it>d7 would have led to a lost pawn ending. Of course, not 26 . . . l:tf2 because of 27 ilxg7 1Wa4 28 .:l.cl ! This move does not yet spoil anything, but there was no need for it. Now White loses his e5 pawn, but Spassky' s intuition did not betray him, and he has every reason to hope for a draw. After 35 ttJc5 .ltxg3 36 tDxe6+ <3i>d6 37 tlJxg7 White loses his knight: 37 . . . <it>e5 38 tDe8 29 <it>e6. 21 28 • • • <i>c2 nn The rook cannot be taken because of mate (28 �xfl 'i'd3+). 'ii?c6 b5 Sooner or later this capture would have been forced. Since White's king could not leave d3 because the exchange . . . b5xc4, b3xc4 would be unfavourable for him, he was only able to move his knight. But then Black, using zugzwang, 289 would reach the central e5 square with his king. 37 <it>d5 If 37 . . . <itixb5 38 lDd2 <irc6. then 3 9 'ili>e4 ! and � ' and White h3s created an impregnable fortress. Therefore Black pins his hopes only on the activation of his king. As for the b5 pawn. for the moment it need not be taken, since it is of no great importance. • 38 • e3 , and defend the b4 pawn. Then his king will be able to head for e5. • 41 42 43 44 ft)f.2 ft)g4 lt)f2 li)g4 .i.g5 .i.f4 .i.d6 45 46 47 48 49 b6 ttlxh6 lLif5 cii>e 2 li)b4 gxh6 e5 e4+ �e5 <it>d4 The other defensive plan involved 44 ti:)e4. After 44 . . . .i.:f8 45 lDt"2 JJ..e7 46 lDe4 <it>e5 4 7 tt:)f2 lt>f5 48 <ili>c4 <ili>g5 49 ti:)d3 �xh5 50 liJxb4 g5 5 1 a4 g4 52 �d3 White would probably have gained a draw, but 44 . . . .i.c7 45 iLlf2 JJ..b6 46 �e4 <it>eS 47 liJd2 �4 48 <it>c4 <it>e3 49 lDfl+ (or 49 �b l .i.a5) 49 . . . <iire2 50 liJg3+ <it>f3 5 1 QJf1 .i.c7 ! would have secured Black a win. Therefore Spassky's decision to exchange the kingside pawns is well founded, even though it allows Black's passed pawn to advance. .i.c5 44 g5 After the game Spassky pointed out a more convincing drawing possibility: 3 8 <it>e3 il..c7 3 9 �f3 ! @d4 4 0 lDt"2 <it>c3 4 1 <it>e2 <it>b2 42 <ifi>d3 43 <it>c2, and there is nowhere for Black to break through! However, it would seem that White' s plan of exchanging the kiugside pawns should also have led to a draw. 'iti>xa2 · 38 39 40 ltlxg5 ltle4 hxg5 .i.f4 .i.b6 There now begins a prolonged manoeuvring battle witl1 tl1e bishop against the knight. Black' s aim is to move his bishop to squares from which it can both prevent the knight check at When the game was adjourned, I analysed only the variations given in the 290 previous note. Therefore Black's later play was not always consistent, but this should hardly have affected the result. 50 �f5+ 51 �h6 <"bd5 50 . . . <&ttc3 did not achieve anything because of 5 1 lDg3 . This move looks dubious. but how can it be refuted? After. 5 i . . . �e6 52 ttJg4 @f5 53 %6+ 'it>f4 54 &Df7 1l.e7 55 l!Dh6 the white knight cannot break free, but there is also no way for Black to win it ! 51 52 53 54 55 56 �g4 �f2 lDdl �f2 lbg4 57 itlf2 58 59 60 61 ttlg4 ttlh2 lbn lbd2 This position reached once . . . Titis position occurred . . . has too il..e7 .i.g5 <ii>d4 .i.cl 'iit>d5 .i.g5 already .i.f6 has been 62 ttlc4 62 63 <ittd t <it?c3 Ji.d4 64 <i>e2 e3 65 ttla5 Apparently� only this ill-fated move leads to defeat. Aft.er 62 lbfl <&ttc3 63 CDg3 �b2 64 �d 1 ! �xa2 65 'it>c2 e3 66 lDe2 the b5 pawn would have saved White, since Black does not have . . . a7a5-a4. But now the knighfat c4 will no longer be able to return to e2, and the game is decided. • • • Tlueatening 64 . . . �d3 . Black is intending . . . �c2-blxa2. If instead 65 tDxe3 .i.xe3 66 �xe3, then the pawn ending is hopeless for White: 66 . . . 'it>b2 67 'it>d3 'it>xa2 68 'it>c4 'it>a3 . already 65 66 ttlc6 67 ll\e5 68 lbd3 White resigns: the inevitably queen. .fi.d4 .i.c5 c.ird4 i.b6 <ii>b2 Ji.c5 'it>xa2 il..e7 b4 pawn will Game 346 Botvinnik-Balashov Hastings 1966167 Nimzo-Indian Defence 1 2 3 4 5 6 7 c4 lDc3 d4 e3 i.d3 a3 i.xc4 lDf6 e6 .i.b4 0-0 d5 dxc4 .i.d6 This position occurred twice in the World Championship Return Match of 29 1 196 1 (Games 285 and 286)� There it was mentioned that 7 . . . ilxc3+ is bad for Black. 8 f4 After 8 ttJf3 lllc6 9 b4 e5 10 i..b 2 White's position is considered to be slightly preferable. although Game 286 showed that Black can successfully fight for equality. In the present game White tries to test another possibility, relying on the inexperience of his young opponent., but this proves to be a psychological miscalculation. can hope for some advantage, since in the endgame his king will be better placed than the opponent's. But for the moment the initiative is with Black. 11 12 tLlfJ lhd8 .i.d7! The threat of 13 . . . :cs is highly unpleasant, and White must be attentive. 13 ii.d2 14 tLle5 %%.c8 The only way! After 1 4 i.d3 i.xe3 White would simply have lost a pawn. 14 c5! 8 Threatening 9 . . . cxd4� when Black will be better mobilised. Therefore White must not avoid the possible exchange of queens. 9 10 dxc5 b4 i.xc5 10 11 . 'l'xd8 i..b6 A tempting. but risky move. The b4 bishop. c3 laright and e3 pawn are insuf ficiently well defended� which is subtly exploited by Black. 10 �f3 was correct. • . This. strictly speaking� is the idea of the entire variation. Theoretically White 292 tllg4 This simplification eases White's de fence. However, even after the strongest continuation 14 . . . ..ie8 ! he would have maintained the balance (in the only possible way!): 1 5 'it>e2 ttJc6 16 ltJxc6 .i.xc6 17 :hg 1 i.xg2 1 8 .i.xe6 ! 15 16 17 18 tllxg4 lDe5 ll\xd7 �e2 :ixc4 llc7 tllxd7 It has become clear that Black's initiative has petered out, and it would have been simpler for him to accept the inevitability of a draw. 18 • • • tDt'6 : �ct 19 20 . 21 l!hdl Ael lid8 !lcd7 t'.Dd5 22 23 t'.Dxd5 l:txd5 r!xd5 l:xd5 24 a4 . Black would not have achieved any thing with 2 1 ... lLlg4 on account of 22 1.lxd7 l:Ixd7 23 t'.LJa.4 ! , but' the exchange of knights leads merely to a loss of time. 2 1 . . .l:txd l was the simplest. 23 . . . exd5 was more circwnspect, but Black, naturally, did not want to give himself an isolated pawn. It transpires that White' s opening idea has in the end proved successful - his king is more active than the opponent's. 27 28 29 30 e4 b5 i.d2 lia8 lld7 · i_e7 .i.d6 b6 Now Black is left with a weak pawn at a7, which, with the bishops on the board, makes the endgame hopeless for him. In the rook ending after 30 . . . .i.c5 3 1 .te3 .i.xe3 3 2 ct>xe3 b6 3 3 axb6 axb6 34 l1b8 White also should have been able to win, but more accurate play would have been required than in the bishop ending. 31 32 33 34 a6 .ic3 h3 llc8 i.c5 <li;e7 <Ji;d6 f6 24 An obvious move, but most probably t he losing one; it was all due to a tactical oversight. By 24 . . . l:.d7 25 a5 .W.c7 Black would have retained a defensible positioD:- for the moment not allowing the enemy rook into his rear. 25 26 bxa6 29 i..a 5) White's pressure would have intensified 26 ·<MT a5 l:lc8 .td8 Only here did Balashov notice that after 26 . . . a6 27 b5 (27 . . . axb5 28 a6 293 34 lk7 An oversight in time trouble, but defeat was already inevitable. 35 36 37 38 39 e5+ lig8 Jhg7+ klg8+ llh8 Cit>d7 f5 <it>d8 Ci&>d7 Jl.e7 Black tries to activate his rook, which leads to the loss of his a7 pawn. 40 41 42 ltc2+ Ad4 lla2 @dJ lla8 Black resigns Game 347 Liberzon-Botvinnik Nloscow Team Championship 1967 Sicilian Defence 1 2 3 4 5 6 7 e4 lhfJ d4 lhxd4 tLlc3 .i.g5 1i'd2 c5 d6 cxd4 ltlf6 ltlc6 e6 h6 11 12 12 game of my match with Bronstein ( 195 1 ), but after a disaster against Keres in the Alekhine Memorial Tournament ( 1 956) I again decided to test this variation in a team match. This is all described in more detail in the notes to Game 1 84 (with Bronstein). Axf6 gxf6 0-0-0 In 1 956 Larsen chose against me a move that is less dangerous for Black 9 l:dl (No.243). 9 10 • • f4 • .i.d7 In a game with me in 1 952, Suetin developed his light-square bishop at c4 (No.2 1 5). I first employed this move in the 6th 8 9 @bl J..e2 a6 h5 In my game with Bondarevsky ( 1 9th USSR Championship, 1 95 1 ) I tried to manage without this move, and wrongly, since the white bishop should not be allowed to go to h5. However. it is quite possible to defer . . . h6-h5 until White has played .i.e2. • • • '8'c7 Against Keres I chose 12 . . . 'i'b6, but he retreated his knight - 1 3 ltlb3 , and the queen turned out to be badly placed at b6. 13 %lhfl il.e7 The most accurate. In the event of queenside castling it would have been easier for Wnite to find a plan. In particular, he has :f3-h3, attacking the h-pawn. Black's aim is for the moment not to determine the position of his king, since, for example, after the exchange of queens his king may come in useful in the centre. 14 15 16 llf3 !fxd4 ltlxd4 9'c5 1fd2 A situation that is normal in this variation: Black offers the exchange of queens. After all , his king is secure and · better placed, and the two bishops may 294 have. their say. For the same reasons, White, naturally, avoids going into an endgame. 16 17 • . . lleJ .i.c6 1la5 Now Black prepares to castle long. which was unfavourable immediately ( 1 7 . 0-0-0 1 8 l2Jd5). However, White skilfully prevents the enemy king from moving to the queenside. Therefore 1 7 . . . h4 should possibly have been considered. 18 19 • • ltd8 • -*.c4 Planning at the appropriate moment to attack the e6 pawn by f4-f5 . Black, of course, takes prophylactic measures. .. 19 20 21 9'e2 .i.a2 22 f5 .i.d7 1lc8 A forced (sooner or later) retreat. After 2 1 liJd5 exd5 22 exd5 �f8 23 l:he7 Black would have won by 23 . . . ilg4. 21 <i>f8 Somehow Black has just managed to complete the development of his pieces. Not without reason, White avoids 22 e5 fae5 (but not 22 . d5 23 f5) 23 fxe5 d5, after which his bishop at a2 is shut out of tl1e game. The manoeuvre that ·he has begun leads to the winning of the h5 pawn. . . 22 23 . • • 1:1.hJ W'e5 .i.d8! After 23 . . h4 24 'ii'f2 Black would have lost a pawn without any compen sation. Now, however, the similarity of the ideas in this game (although here the queens are still on the board) with No. 2 1 5 becomes apparent. . The same position has been reached as in my game with Bronstein. More over, White both times chose the same move order, and it was only Black who slightly varied his replies. 18 aJ! This, it turns out, is what my opponent had prepared. Bronstein played 1 8 �f3, allowing Black to castle queenside, which is now ruled out after 1 8 . . . 0-0-0 1 9 b4 1Wb6 ( 1 9 . . . 'i'xa3 20 ttJd5 'iia4 2 1 l2Jb6+) 20 ttJd5 exd5 2 1 exd5 .1xd5 2 2 l:txe7 White has an obvious advantage. The Encyclopaedia thinks that 1 8 f5 should also be considered. 24 25 26 27 28 l;Ixb5 'i'xh5 1fh6+ fxe6 ifh7+ 28 29 liJe2 �xh5 .i.a5 c:J;;e7 fxe6 I think that 28 'ifg7+ <it>d8 29 ttJe2 would have caused Black more prob·· lems, since 29 . . . 'i'xe4 leads to mate after 30 'i'f8+. 295 �d8 f5 34 The point of the manoeuvre begun with 23 . . . .td8 ! ; 30 · exf5 does not work because of 30 . . . 1ixe2 . After 30 it:)g3 f4 3 1 ltJe2 i.c6 3 2 'i'g8+ 30 rJ;c7 the active placing of Black's pieces compensates for the lost material. However, my opponent finds a clever combination. which places Black on the verge of defeat. 30 lLld4 34 . . . .tt.c7 seems forced, but then after 35 'i'g5+ Black has to part with his bishop. But he has a possibility of 'selling' it more dearly. 34 i.xe6 llfl fxe4 .i.xe6 ifxe6 1i'e8 36 . bxc3 Again the only defence. If 3 3 . . . �e8 34 'ii'h5+, and White regains the piece, while retaining all the advantages of his position 32. 33 . . . 'ife7 is not possible: 34 �f7 'iie8 3 5 'iih4+. and mates. i.cJ! Since 3 5 bxc3 loses the rook (3 5 . . . ifb5+ and 36 . . . 'ifxfl), Black forces the enemy queen to abandon the critical g7 square. Here the continuation of the attack by 3 5 'ifxb7 does not achieve anything after 35 . . . .i.d4 36 c3 i.xc3 ! (but not 36 . . . .i.c5 3 7 b4 i.e3 38 :n, and Black is in zugzwang) 3 7 bxc3 I:txc3 . All the same. Black has at least equal chances. and White has to restrict himself to regaining the piece. </;c7 35 1i g5+ 35 . . . 'i'e7 was not possible due to 36 :tf8+, while if 35 . . . �d7 White could have replied, as in the game, 36 bxc3 . There is no other move. After 30 . . . 'i'xe4 3 1 it:)xe6+ jLxe6 32 %.:.xd6+ Black would have had to resign, whereas now he is threatening by . . . d6d5 to consolidate his position . 31 lLlxe6+! 31 Perfectly logical : the bishop covering the black king is exchanged, and White seizes control of the seventh rank and creates mating tlueats. 31 32 33 1Vg7 �b8 Despite being a pawn down, the weakening of the enemy king' s position gives Black sufficient counterplay. 296 37 1id5 38 · :et 39 1Wd4 :c6 1fh8 any opening surprises, but, alas, he does not succeed in this. This leads immediately to a draw, but after 39 Ihe4 Wxc3 40 ifd3 llb6+ or 39 'i'xe4 'i'xc3 40 'ile8+ ct/a7 4 1 'i'e3+ �b6+ 42 �c l 'i'h8 Black's chances would have been no worse. 39 40 ltxe4 40 41 42 l:le8+ Jle7+ • • . 1H'xh2 Or 40 1Yxe4 lhc3. 3 4 5 6 7 it.gs lbbd2 J.h4 .i.g3 8 9 e3 .i.e2 9 10 hxg3 Q)f6 d6 h6 g5 ll'lb5 This standard manoeuvre had already occurred in a training game of mine with Petrosian (cf. Volume 2), and also in Games 268 and 325, but the first main and exact predecessor was the game Petrosiari-Botvinnik (1 9th USSR Championship, 1 95 1 ). In the resulting simple play, great significance is acquired by the posit ional subtleties, typical of this variation. With these my opponent was probably not familiar. e6 A not altogether successful move. In the afore-mentioned 195 1 game Petrosian developed his bishop at d3. 1ixg2 <t/c7 <iltb8 The only way! 42 . .. �c8 would have lost to 43 'i'h8+, and 42 . .. <it>d8 to 43 J;txb7. ttlxg3 g4 <l/c7 l::te8+ Draw agreed 43 Game 348 Levit-Botvinnik USSR Spartakiad, Moscow 1967 King's Indian Defence 1 2 3 d4 g6 .i.g7 ttlf3 cJ White deviates from the more com plicated continuations, in order to avoid 297 This move suggests itself, although it seems rather risky. By continuing now 1 1 ill g l ! h5 12 .i.d3 followed by ltle2 White would still have retained a satisfactory position. 11 White is unable to find a clear plan, and so he makes useful moves. For Black it only remains to activate his king's rook. lhb4 After this 'natural' move White' s king's knight is shut out o f the game, and Black's chances (in both the middlegame, and the endgame) become preferable. 11 12 13 14 15 11c2 0-0-0 <i>bl .:ct l:lh6 22 b5 lt:\c6 \ie7 .i.d7 a5 Black' s entire plan of development., including this move signalling the start of an offensive against the enemy king, · also occurred then, in 1 95 1 . Black, of course, does not intend to castle on the queenside, where his king might come under attack. <ifi>t'8 16 .i.b5 Here the king feels quite safe, better than after castling on either side. As for Black' s king's rook, in time it will come into play via h6 . 17 1'a4 lia7 18 c4 lLlb8 19 J.. xd7 1ixd7 Preventing b2-b4, just in case. Black does not object to the exchange of the light-square bishops. A perfectly justified offer to exchange queens: in the endgame Black has clearly the better chances. 20 1fa3 <it?g8 21 22 tbe4 l:tbdl a4 On prophylactic grounds Black removes his king from the a3-f8 diagonal. Here is the manoeuvre, predicted in the note to Black's 16th move, and typical of the plan chosen by him. 23 d5 24 lt:\cJ 25 26 lLlb5 dxe6 White is striving for the initiative and wants to complicate the play, but the main thing that this move does is to activate Black's bishop. In his time, Tarrasch. commenting on a similar situation, concluded his note with the words: 'mate on g2 (there Black had a light-square bishop - M.B.) will be inevitable'. Here too in the end a cat astrophe awaits White on the b2 square. 'i'e8 23 • • • lla5 Not fearing 25 b4, which would weaken still further the a 1-h8 diagonal. Now Black is intending to bring his knight into play (in this case the a4 pawn must be defended). 298 tba6 fxe6 27 35 36 37 38 tLld4 My opponent evidently realised that a fter 27 b4 l:txb5 ! 28 cxb5 'iixb5. his position would become altogether hopeless. 27 28 tLlb5 b6 d5 29 30 31 tLld4 tLle2 lLlf4 c5 lLlb4 :as . . • This breakthrough, activating Black' s pieces, enables him to create a decisive attack. 3 2 cxd5 exd5 3 3 ttJxd5 (or 3 3 lDf5) would be advantageous to Black in view of 33 . . ...e4+, and so he can unhurriedly defend his back rank. 32 33 lLldJ lhdJ liel f4 lhe4 i.d4 fif7 dxe4 e5! Now the doubling of the black rooks on the b-file is unavoidable. 39 40 fxe5 :b6 l:tet l:.ab8 White resigns Tarrasch was correct. But Black was also correct, when he drove White's king's knight to the edge of the board. Right to the end of the game it remained there as a static observer. Game 349 lDxd3 b5 Botvinnik-Polugayevsky USSR Spartakiad lvloscow 1967 King's Indian Defence 1 2 3 4 A little trick, consisting in the fact t hat White cannot reply 34 cxd5 due to the loss of his queen after 34 . . . b4. As a result the b-file is finally opened. 34 35 :d2 e4 bxc4 If 3 5 'iixc5 the simplest is 3 5 . . . a3 , sweeping away the last obstacles on the long diagonal, or else 35 . .. l1b8. _ c4 lLlfJ lLlc3 eJ c5 tLlc6 g6 I first employed this variation in the 4th game of my World Championship Match with Petrosian (No. 309). There Black prevented the advance of the d pawn to the fifth rank by 4 . . .tDf6 5 d4 cxd4 6 exd4 d5 (incidentally, this pos ition can also be reached in the Caro Kann Defence). In that game after 7 cxd5 tiJxd5 8 �3 ltJxc3 9 Ac4 Black had problems. 7 i..g5 lDe4 8 cxd5 ttJxc3 9 bxc3 'i'xd5 10 'i'b3 is also possible. Polugayevsky decides not to prevent the advance of White's central pawn, and the game transposes into a King's Indian Defence. 299 4 . 5 6 d4 .i.e2 .ig1 d6 In the given situation White' s king' s bishop i s better placed at e2 than at g2. 6 7 • • • d5 tLlf6 tLla5 7 l£le5 8 CLlxe5 dxe5 9 e4 would have promised White some advantage, but moving the laright to the edge of the board also has its negative points. . .. 8 e4 It would seem that, compared with the usual variations of the King' s Indian Defence� White has lost a tempo, since he has advanced his e-pawn is two moves instead of one. But how much time will Black have to spend, for his queen's knight to become a fighting piece! 8 .. 0-0 9 0-0 9 h3 also came into consideration, not allowing the enemy bishop to go to g4. . 9 • • • .i.g4 10 .i.e3 White is now intending to advantage ously continue 1 1 tbd2. Therefore Black must not forgo the opportunity to double the white pawns� although he pays for this a high price- he has to part with his light-square bishop. il.xfJ 10 e5 11 gxf3 A natural move. After f3 -f4 Black prevents the possibility of f4-f5, which could have become a constituent part of an attack on the kingside. Now, however, White exchanges his doubled pawn. . • • 12 f4 12 13 14 .i.xf4 .i.dJ 16 bJ 16 17 18 <ii>g2 The alternative plan - 12 dxe6 fxe6 1 3 e5 or 1 3 f4 - would also have left White with a slight advantage, but he prefers to retain his central d5 pawn, continuing to play for the exclusion of Black's queen' s knight. exf4 11'e7 tLld7 If 14 . . . tllli5 there would have followed 1 5 �d2 ( 1 5 . . . i.xc3 16 .i.xc3 tLlf4 17 �hl with an obvious advantage to White 33). tLle5 15 :c1 White does not fear the exchange of the lmight at es for his bishop at d3 the fewer the pieces on the board, the greater the significance of one of them being shut out of the game. h5 a6 .i.b1 If Black really doesn't want to eliminate the light-square bishop, how can White deny himself the pleasure of retaining it ! 300 18 19 . . 1le2 queenside merely frees White's hands in the centre. l:lab8 'i'd7 30 31 <irh2 tiJdJ t£id8 <"atrb8 3 1 .. .\Wc7 was more tenacious, although it would . not have made any fundamental difference. 20 .td2 With this move Wl1ite begins a systematic preparation of e4-e5. All his pieces, apart from his king. queen and I i ght-square bishop, will be aimed at the square that the pawn is striving to occupy. 20 21 22 23 24 �dl f4 h3 �f2 b5 tDb7 �g4 lDf6 :.be8 32 33 Black is forced to give up any plans or active play on the queenside, in order Io try and prevent the breakthrough of 1 he enemy forces. From now on the sights of both players will be mainly I rained on the e5 square. 25 26 27 28 29 :eel 'ifo :e2 ltfel .tc2 32 l!e7 :res �h7 .t.d4 b4 This strategy can hardly be correct. Nothing compelled Black to make this advance, and the closing of the 301 e5 White has carried out the advance of his central pawn, and his advantage becomes decisive. <i>hl 1rc7 dxe5 Typical of Polugayevsky' s style. He could still have stuck to waiting tactics and not forced the play. Now Black is bound to lose material. 34 35 fxe5 .l:xe5 35 36 ltxe5 .i.xe5 In 1 935, in a game against Lisitsyn in the 2nd Moscow International Tourna ment, I overlooked the possibility of such a manoeuvre (while the game was still in progress it was mentioned to me by Capablanca himself). But here I made use of Capa 's adVice. lbe5 lixe5 37 i.f4 ktf5 38 39 i..xc7 i.. xd8 llxfJ lleJ 40 41 42 43 44 <t>gl .i.dl ll)f2 �fl i.g5 lte2 :d2 ll:\f8 li)d7 :xa2 The resulting endgame is hopeless for Black. However, his position also crumbles after 37 . . . f6 38 i.xe5 fxe5 3 9 'i'e3 . In this way Black at least penetrates onto the second rank with his rook. If instead 39 . . . .fixh3+ 40 �g2 l:te3 4 1 � �h3, then 42 JJ..c7 followed by 43 lDxc5, and White wins easily. ll:\dJ i.e7 h4 �el .*.d6 ll:\b2! 50 51 52 53 54 55 l:tal ll:\a4 ttle5 i.xe5 fxe5 lLlxc5 as �d2 �a2+ <ite3 Black resigns Game 3 50 Taimanov-Botvinnik Black has a rook and two pawns against two bishops, which nominally is not so bad, but the weakness of the c5 pawn leaves him no saving chances. 45 46 47 48 49 50 This move, found in adjoununent analysis, decides the game. In view of the threat of 5 1 t£ia4 followed by 52 i.c7, the black rook has to leave the c5 pawn undefended. USSR Spartakiad A1oscow 1967 Grtinfeld Defence �g7 lla5 f6 1 2 3 4 <i;f7 d4 c4 lf)tJ lLlbd2 d5 c6 lLlf6 White is aiming to deviate from the well-trodden variations of the Queen' s Gambit. �g7 What can Black do? He is forced to admit that he has no useful moves. 4 • • • g6 Now something resembling a Grtin feld Defence arises. but in a situation that is more favourable for White, since, instead of being at c3, his queen's knight is at d2. Therefore the realisation of Makogonov' s idea, involving b2-b4, promises White more than in the conventional set-ups of this defence. ..tg7 5 e3 6 . .i.e2 7 0-0 8 b4 302 0-0 b6 The outcome is that the opening battle has gone in White's favour. 8 • ·• . .i.b7 The alternative 8 . . . c5 (Novotelnov Fbralidze, 1 949) led to _complicated play, and is also probably advantageous l o White. 9 10 i.b2 lLlxe4 lLle4 This exchange is of no benefit to White. The simplest was 10 c5 ( 10 . . . bxc5 1 1 bxc5 t'Llxc5 1 2 dxc5 il.xb2 1 3 .tlb l ), after which it is not easy for Black to find a successful plan. 10 11 dxe4 iLld2 But not 1 1 t'Llg5, which would have been answei:ed by l l . . . c5. with favour able prospects for Black. 11 12 13 c5 1ib3+ 14 d5 f5 iLld7 <t>h8 An amusing oversight, as a result of which White not only loses his advan tage, but even immediately ends up in a lost position. 14 . • . cxd5 303 15 .txg7+ 'it>xg7 Here my opponent now saw that his planned combination - 16 c6 .i.xc6 1 7 'iic 3+ would not achieve anything after 17 . . . l:tf6. After this, instead of winning a piece for two pawns, White would simply be two pawns down. He had to quickly adjust, in order to be just one pawn down. 16 .i.b5 bxc5 17 18 19 20 bxc5 'i'cJ+ iLlbJ :act :cs l:if6 �g8 a6 21 .i.xd7! Towards the end of this competition I was feeling rather tired, and so I avoided the more complicated 16 . . . liJf6 1 7 c6 �c8, which would evidently have been more consistent. As a result of Black' s tendency to simplify, White acquires certain equalising chances. White' s knight will be clearly stronger than tl1e black bishop, and this gives him definite compensation for the lost pawn. 21 22 :tfdl 11'xd7 llc6 23 lDa5 f4 Black .· is already forced to fight for the initiative� since in some cases White could have activated his rooks on the b file. But now the game transposes into a double rook ending with a minimal advantage to Black. 24 25 exf4 lDxc6 After 25 . . . 'i'xc6 regains his pawn. 26 27 28 11fe5 11fxe4 lbd7 llxf4 Jlxc6 26 ife5 Even here it was still possible to resist successfully with 3 1 l:r.d2. The move played allows the· exchange of the weak e4 pawn for the important g2 pawn. 31 32 33 · �xet lla4 %1.xel+ lk2 33 l:txe4 really was better, but my opponent did not want to admit to the inaccuracies committed. White 33 34 35 dxe4 llfxc5 fxe3 h4 e3 llxg2 h5! Now White has a limited choice: a hopeless rook ending, in which after 36 l:xa6 l:.g4 Black creates two connected passed pawns on the h- and g-files, or a pawn ending, where it appears that all is not yet clear . . . 36 ct>rt So, the choice is made - the pawn ending. 36 37 38 29 llxg4 'i&'g2 llg4 hxg4 II.et A dubious move. It would appear to lead to White ' s king approaching the centre with gain of tempo, but in reality the centralisation of the king weakens the kingside pawns� which inspires Black with new hopes. After 29 lhc5 llxc5 30 g3 �f7 3 1 l:td2 or 30 . . . :es 3 1 lla7 Black's winning chances would have been microscopic. 29 30 31 <t>ft lid4 . ltct · �7 38 304 g5! After 38 . . . �f6 39 Ciftg3 �5 40 e4+ Black would ·have been unable to count on more than a draw: (40 . . . �xe4 4 1 c.t;xg4 e5 4 2 �g5 � 4 3 �xg6 e4 44 h5 etc.). 39 Game 3 5 1 Botvinnik-Boleslavsky USSR Spartakiad Moscow 1967 English Opening h5 1 2 3 4 If 3 9 hxg5 <it>g6, and Black wins easily, while after 39 <it>g3 �g6 40 \�xg4 gxh4 4 1 @xh4 <it1'5 42 <it>g3 �e4 43 <£fi>f2 as White loses due to inevitable 1.ugzwang. 39 40 . • c1>g3 lDf6 e5 c6 d6 Regarding 4 . . . e4, see Game 288. 5 6 7 <l;g7 <i>h7 . c4 ltlc3 g3 ft)fJ Of course, not 40 . . . 'it;h6 41 <it>xg4, when it is not White, but Black, who is i n zugzwang (4 1 . . . as 42 e4 a4 43 eS). But now it is White who is in zug1.wang, and for the reason that the doomed g4 pawn deprives the white king of both important squares f3 and h3. .i.g2 0-0 d3 g6 .*.g7 - 41 42 <it>xg4 e4 <it>h6 This move was sealed by Wltite. The result of the game is so obvious, that I took the liberty of departing from the usual norms of behaviour and immed iately convinced my opponent that further resistance was hopeless. The point is that after 42 . . aS 43 a3 e6 44 a4 c5 4S <it>f5 �xhS 46 �xe5 g4 47 �4 <hh4 48 e5 g3 49 �f3 �h3 50 e6 g2 5 1 c7 g l 'i' 5 2 e8if 'i'fl+ White loses his queen, as I was 'taught' back in the 1 1th game of my 1937 match with Grigmy Levenfish. . White resigns. The variation employed by Boleslavsky in the present game makes it difficult for White to transpose into the King's Indian Defence, since after 7 d4 e4 8 tiJd2 d5 Black's e4 pawn is already defended. Even so, with 7 e4 followed by 8 d4 White could have achieved this. But was it really advan tageous to aim for the King's Indian, an opening of which my opponent had made such a subtle study? . . . 7 8 9 30S .i.d2 1fcl lDbd7 0-0 9 l:tbl , preparing b2-b4. possible. lie8 9 • 1. 0 • knight at IB, and in addition Black is behind in the development of his queenside pieces. is also • ihgS 13 A shrewd move. White provokes 14 15 . . . h7-h6, which, generally speaking. weakens Black's position, and at the given moment is simply not possible: after 10 . . . h6 11 l2Jge4 the d6 and h6 pawns are simultaneously attacked. and 1 1 . . . lL\xe4 12 lL\xe4 d5 does not save Black in view of 1 3 ti:)d6. 1.0 11 12 13 b4 cxd5 e4 16 ihfJ dxe4 lldl h6 dxe4 'ith7 This move begins a prolonged battle for the open central file and the d5 square. 16 17 18 ttlf8 d5 cxd5 'ifb2 .i.eJ i.d7 ile7 White 's pieces are better prepared for tl1e occupation of d5, than Black's for the manoeuvre of his laright to d4, and also White's entire force is much more active than the opponent's. 18 20 hJ tiJd2 22 23 'iib3 ::txd7 19 b6 .i.c6 fiad8 If 20 . . . ttJe6, then 2 1 ttJd5 CDxd5 22 exd5 e4 23 'ilfb3 is advantageous to White. ife6 21 ihc4 l:d7 This exchange is necessary, to retain control of the d-file. The only plan in this type of position. It unexpectedly turns out that it is not Black. but White who is playing the King' s Indian Defence with reversed colours. If now the black pawn adv ances to d4, the position favours White, as was demonstrated many times (with reversed colours, and hence for Black ! ) by Boleslavsky, by Geller, and b y Tal. Therefore my opponent exchanges pawns in the centre, but then the white knight at c3 proves more active than the 23 24 25 :dl l£ld5 lL\8xd7 ll.f8 So, the central d5 square has been occupied, whereas d4 is inaccessible to the enemy pieces, and this gives White an obvious positional advantage. Now Black has to parry the threat of 26 li:Jc7. 306 25 26 :cs .i.cl Since the · d4 square is completely secure, White has an opportwrity to I ransfer his bishop to a more active position, from where it will threaten Black's central pawn. i.xc6 and 34 'i'd3+, or 3 3 'ifxe6 fxe6 34 .i.xc6 l::txc6 3 5 :xd7+). 31 26 27 28 l£ie8 l£id6 i.b2 l£ixd6 Although, as a rule, exchanges in a cramped position ease the defence, here this exchange merely increases White 's advantage, since it gives Black new problems, associated with the dangerous position of his king. 28 29 i.xd6 f4 White avoids a cunning trap, typical of Boleslavsky 's resourceful style (29 0ixb6 lbxb6 30 1i'xe6 fxe6 3 1 .:xd6 11Jc4, with a material advantage for Black), and he begins play on the kingside, which was planned back on Lhe 26th move. 29 il.f8 Of course, not 29 . exf4 30 gxf4 i.xd5 3 1 ifxdS, and Black's position becomes desperate. .. 30 f5 1fe8 If 30 . . . gxfS 3 1 exf5 'i'xfS, then 32 t2Je3 and White wins (3 2 iie6 3 3 fxg6+ White also retains the advantage after this, but 3 1 l:Ifl or even the preparatory 3 1 a3 was stronger still. When the opponent is bound hand and foot, there is no reason to force events. 31 32 33 �fl J:tf2 fxg6 i.g7 By this point it had transpired that, for the victory of our team in the competition. it was evidently sufficient for me to gain a draw. This explains that I had to retain an advantage, just in case, but to avoid complicated continuations which would involve some risk. 33 34 35 <ifi>h2 bxa5 <ifi>h8 a5 It was simpler to play 3 5 a3, main taining control of c5 . White exchanges on a5. to create a weak pawn in Black's position. This pawn is won, but the opponent acquires hopes of a draw. . . . 307 35 36 i.a3 bxa5 il.xd5 37 1ixd5 37 ... lDb6 Peace was concluded on the directive of the two team captains. 38 39 40 ifxa5 · 1ic3 11f xa3 lDc4 lDxa3 l:a8 Botvinnik-Toran Draw agreed. Exploiting the very first opportunity, Boleslavsky takes play into· an ending with opposite-colour bishops. Game 3 52 Palma de Mallorca 1967 Sicilian Defence If now White were able to play his bishop to d5, he would gain winning chances. but Boleslavsky skilfully pre vents this. 41 42 43 44 1id6 flc7 1ib7 \id5 44 45 46 47 1ib3 1fdt lld2 lld8 llc8 ltb8 In a team competition, and especially at the end of the last round, the unwritten rules state that you cannot take an independent decision: to agree a draw or adjourn the game, in order in analysis (if it is required) to seek a way to play on. For the moment our captain was considering this question . . . l:d8 llb8 :ds 1 2 3 4 5 6 c4 e4 lDfJ d4 lDxd4 thc3 g6 c5 l£ic6 cxd4 lDf6 lhxd4 At one time this exchange was con sidered to be something of a revelation in this seemingly thoroughly studied variation. Persistent searches of forcing continuations were made, to try to demonstrate an advantage for White. Fortunately, I did not remember all these investigations, and I chose · an out wardly unpretentious variation, which nevertheless guarantees White a slight advantage. However, in our time it is almost impossible to think up something new at such an early stage of the game. 308 7 8 9 10 11xd4 .i.e3 i.e2 'i'd2 11 .i.xg4 mentioned game after 1 2 0--0 lic8 1 3 b3 b5 Spassky sacrificed the exchange - 14 CiJxb5 .i.xal 15 :x:al, but he lost his opening advantage, and the game quickly ended in a draw. d6 .ig7 0-0 lhg4 Only after the toumam�nt did I learn that this position had already occurred in the games Cardoso-Tai ( 1 958), Tal Gurgenidze ( 1 959) and Tal-Ghitescu ( 1 960 ), the last being one that I should have seen at the Leipzig Olympiad, but somehow it did not draw my attention. Black's other reply, 10 . . . .i.e6, was employed · against me by Matulovic in the second round of the 'Match of the Century' (Belgrade 1 970). After 1 1 f3 'ti'a5 12 tiJb5 'i'xd2+ 1 3 �xd2 tDd7 White could have gained the better ending with 14 CiJc7 l'tac8 1 5 l£ixe6 fxe6 1 6 :tabl followed by f3-f4. In all the above games where Black played 10 . . . l£ig4, White replied 1 1 .�d4. which led to some complications. By exchanging Black's knight, White deprives his opponent of any counter play. And again I was unaware that a few months earlier this exchange had occurred in the game Spassky-Petrosian ( 1 967). True, as the reader will see, there the idea of the exchange was completely different . . . To be fair, I should mention that, whereas is those years I was not familiar with the latest findings, my opponent was never strong in opening theory. 11 • .. .i.xg4 Naturally, not l l . . .i.xc3 1 2 'i'xc3 ,.txg4 because of 1 3 i.h6. 12 ..td4 This is now essential� since White must retain his knight. In the afore- 12 . i.e6 13 14 .i.xg7 0-0 'i!i?xg7 <it>g8 16 17 f4 llf3 f6 'itth8 • • Toran fails to find a good plan, but it is hard to say whether Black has one in this position. After, for example, 12 . . . i.xd4 13 'i'xd4 i.e6 14 0-0 'i'a5 1 5 l:fel .:tfe8 1 6 b3 (Keene-Schmid, 1 973) White's chances are better. Of course, 14 . . . ..i.xc4 was not pos sible due to 1 5 ifd4+, but 14 . . . f6 was definitely stronger, not wasting time on moving the king. if as 15 b3 Here the king feels safer; it does not have to fear either the opening of the g file, or, in the event of tactical compli cations, the capture by the knight of the e7 pawn with check. However, in play ing this Black admits that his 1 4th move was inaccurate. 309 18 D.el ktg8 Black, as after 23 :xc5 dxc5 24 lldl the invasion of the rook is unavoidable. :ac8 This too is not the best. 22 . . . J::t.c7 was essential, in order after 23 e5 dxe5 24 fxe5 to reply 24 . . . f5, and if 25 :edl :gc8, retaining, despite the passive pos ition, chances of a successful defence. If 19 f5 Black is ready to reply · 1 9 . . i.f7 . 19 20 :dJ lDd5 Strictly speaking, this is hardly the best plan - in the endgame it will not be easy for White to convert his positional advantage. �ut from the practical point of view this decision was virtually forced, if it is borne in mind that the Soviet grandmasters had arrived late at the tournament and that the present game began less than 24 hours after our plane touched down on the island of Mallorca, so that there was no question of acclimatisation and rest . . . 20 21 :xd2 llc6 22 . 'ifxd2 .txd5 23 24 25 e5 fxe5 dxe5 l:te6 � A little psychological trick, which the Spanish master did not see through. If White had played 25 l::td7, and after 25 . . . l:tb8 26 �' Black would of course have replied 26 . . . fS, when he would still have been able to defend his position. Therefore now too 25 . . .f5 was the correct reply. But the point of the trick is that the new position of the white king seems unfortunate, and Black decides to exploit this. 25 1:.f8 ' The threat of 26 exf6 is parried by 26 . . . :fxf6+, when White loses a rook. But in fact this pseudo-active move proves to be a waste of precious time. 26 22 l:lxd5 A somewhat unexpected, but quite justified decision. After the standard 22 exd5 Black would have withstood with out difficulty the pressure on the e7 pawn (22 . . . :tc7 23 :de2 ltg7 etc.). But now White is threatening by e4-e5 to advantageously open lines in the centre. For example, 22 . . J�c5 is unsuitable for 3 10 • • • J:ld7 In Games 3 16, 3 74 and 378 White did not hurry with the development of his queen's knight, and then brought it out at d2. Naturally, there the opening battle developed somewhat differently. Now if 26 . . . :bs White gains a decisive advantage by 27 exf6. There fore Black has to accept the temporary pawn sacrifice. 26 27 28 Cifi>e3 'it>e4 fxe5+ lib8 3 4 5 6 The active inclusion of the white king in the play makes the outcome obvious. 28 29 30 Cifi>d5 Jlxe5 �xd6 'it>xd6 � lld6+ exd6 J!Ld8+ 6 7 In this way at least the black rook breaks free . . . 33 34 'it>c7 'it>xb 7 g6 �g7 c6 White has not played this variation in the best way (to me it seems more rational to develop the light-square bishop at e2 rather than c4, as, for example, Ciric played against me in Beverwijk - Game 370). Now Black has the chance of shutting the bishop at b3 out of the game by . . . d6-d5. @g8 After 30 . . .I:ixe5+ 3 1 �xe5 �e8 32 @e6 the game would have concluded even more quickly. 31 32 ltJfJ ilc4 .i.b3 1ie2 0-0 lid2 r!xg2 The capture on a2 would not have changed anything. The outcome is decided by the white c-pawn. l:txh2 35 c5 ltc2 36 c6 37 b4 Black resigns. In view of the threat · of 3 8 :Lc5 he has to give up his rook ilmnediately. Game 353 Medina--Bot,1nnik Palma de Mallorca 1967 Pirc-Ufimtsev Defence 1 2 3 e4 d4 tDcJ d6 ltJf6 And this allows Black to exchange his queen' s bishop for the knight at f3 (in most variations of the Pirc-Ufimtsev Defence this exchange is necessary) with the gain of a tempo. If White had decided not to prevent the exchange by 7 h3 (whicl\ however, is not especially good for him). he should have castled immediately. Apparently my opponent 311 17 was already intending to castle queen side. 7 8 9 .i.g4 hJ "i'xf3 li.xf3 e6 It is not possible to play . . . d6-d5, but even the modest move of the e-pawn makes the position of White's king's bishop quite unpromising. 10 11 ii.gs .i.h4 17 18 18 19 20 tLlbd7 0--0-0 Very risky ! Black is the first to begin active play. Meanwhile, after . 1 2 0-0 White's position would not have given cause for alarm. 12 13 14 ife2 f4 1ia5 b5 This move weakens White' s which Black immediately exl>loits. 14 15 16 tDbl tDd2 b4 d5 c5 e4, • • . ihxf6 dxc5 After 18 e5 ctJh5 1 9 \in c4 20 g4 lDxf4 ! White would have been left with a broken position. h6 Th.is move has the drawback that at h4 the bishop will obstruct White' s pawn offensive o n the kingside. 11 12 .i.xf6 The threat of 1 7 . . . c4 was highly unpleasant, and it would still have been on the agenda after 1 7 e5 ttlli5 . There fore White simplifies the position, but after this exchange Black's l<lng' s bishop may become very dangerous. lL\xe4 ifxe4 dxe4 lL\xe4 'ifxc5 It has become clear that, despite the exchanges and the material equality, White's game is already lost, since he has nothing to counter the opponent's pressure on the a 1 -h8 diagonal. All this, incidentally, is repeated in the game Matulovic-Botvinnik ('Match of the Century', Belgrade 1 970 - No. 378). Black' s plan also involved a tactical subtlety, consisting of the fact that after 2 1 :ct? :ad8 22 l:Ihd l (or 22 l!xf7 :xn 23 'i'xe6, when Black has 23 . . . ilf5, if there is nothing better) 312 28 22 . . . �xd7 23 �xd7 'i'gl + 24 ld.dl i.xb2+ he gains a material advantage. My opponent noticed all this, but the most surprising thing is that after 40 minutes' thought he did not in fact think it possible to avoid this variation. The point is, apparently, that Black' s main tlueat, . . . 'i'e7-f6, is not so easy to parry. 21 22 23 24 l:td7 llhdl lixd7 <t>d2 ltad8 :xd7 1f gl + This is what White was apparently pinning his hopes on. If now 24 . . . .txb2, then 25 <it>e2 ! followed by l:d l, and White is only a pawn down with a comparatively satisfactory position. 24 • • • <it>d3 1fe2 26 27 11'f3 29 30 31 32 iffl+ If 26 'ltd2, as already mentioned, there would have followed 26 . . . .i.xb2. a5 1fc6 �e2 �1 :.c7 'iixb2 1!1'e5+ h5 White decides to exchange rooks, but this does not bring him any relie(. since . the rook at f8 is not participating in the attack. 32 33 34 35 'i'f2+ <t>e2 :cS iff4+ .i.d4 it'e3+ <ii?d l Alas, there is no choice . . . But after this accurate move White's losses are more considerable. After 25 �c I 'i'fl + 26 <t>d2 .ixb2 he cannot avoid being crushed. 25 26 1fb7 It only remains for Black to over come some technical difficulties - he has both a positional, and a material advantage. 35 36 .i.cJ lb:f8+ 'ita>xf8 White resigns In my game with Matulovic, which I have already mentioned, in one of the variations exactly the same mating position occurred. ifxf4 'iie5 Game 3 54 Botvinnik-Diez del Corral Palma de Mallorca 1967 King's Indian Defence 1 2 3 4 5 6 7 c4 lbcJ e4 d4 f3 il.e3 d5 t°Llf6 g6 d6 .ll..g7 0-0 e5 t'Lle8 The more usual continuations are 7 . . . lllli5 (Nos.294 and 324), if Black, as 3 13 in this game, is aiming for counterplay with . . . f7-f5, and 7. . . c6, when his plans also include active play on the queen side. 8 9 ild2 10 11 .i.d3 .i.c2 play with l l . . .fxe4 1 2 fxe4 lDf6 1 3 h3 lbhs 14 lDge2 'i'f6 15 <it>bl a5, but White retained some advantage. 12 13 f5 �a6 0-0-0 In beginning t11e manoeuvre of his knight to c5, Black intends to intensify the pressure on the e4 pawn. Therefore White must immediately bring his king's bishop into play, deploying it on t11e b l -h7 diagonal. ltlge2 exf5 .i.d7 This exchanging operation relies on the fact that it is unfavourable for Black to recapture with his bishop, since he loses control of the central e4 square, which will be occupied by the white pieces. 13 14 �c5 gxf5 h4? Not only preventing 14 . . . 'i'h4, after which it is not easy to defend the c4 pawn, but also aiming to disrupt the opponent' s defences by the further advance of the rook's pawn. 14 . . • a4 This attempt to create counterplay cannot be successful; it involves too small a force, but Black cannot just stand still ! 15 16 11 h5 b3 a3 b5 a5 This and Black's nex1 move are based on the conviction that his position on the kingside is perfectly secure, and that, wit11out losing time, he can prepare active play against White' s queenside castled position. Meanwhile, as the game shows, there were insufficient grounds for such complacency� so that he should have preferred l l . . .f4 12 .i.t2 g5, aiming to curb W11ite's initiative. Nine years later, in a game with the new World Champion Anatoly Karpov ( 1 976), Con:al tried to improve Black' s In this situation Black can of course make a temporary pawn sacrifice, but even so he will be too late with his 3 14 counterattack. But it is doubtful whether he had any other plan� his decision is forced. 17 h6! Driving back the bishqp and, even more important, depriving tl;ie knight of the g7 square, from where it could have defended the f5 pawn. 17 18 . 19 cxb5 lih5 ilh8 l:.b8 White is threatening to play 20 ltg5+ �fl 2 1 i.xf5 . The attack on the f5 pawn will now be intensified, in the end leading to White gaining control of e4. 19 20 .. g4 • i.xc5 lDe4 24 25 dxc5 lLlf6 lLlxc5 ltle6+ l:txb5 The advantage is most easily con verted in the endgame. 25 i. xe6 • • • 26 dxe6 9'xd2+ 27 28 29 30 lixd2 lDxf4 :l.xe5 :rs+ .J:te7 lib8 ltlxg4 lDf6 The tempting 26 . . . !:tdS would have been met by 27 'i'b4+ :e7 28 :xe5. Apart from Iris positional advantage, White is two pawns up. l:ee8 31 .id3 l:tf7 f4 20 . . . fxg4 would also give up control of e4, but additional breaches would appear in Black's defences. 21 22 having to play without his king's bishop. 23 %lg5+ 'i&>f8 c6 32 i.b5 33 34 35 36 37 i.xc6 l:tec8 llc2 </;e7 l:! g8 i.b5 <it>d8 l:tc7+ ftfi Black resigns Black has to give up another pawn, in order to at least disentangle himself a little. Game 3 55 Botvinnik-Bednarski Palma de Afallorca 1967 King's Indian Defence 22 . . . i.xb5 would have been decisive ly met by 23 l1g5+ �f8 24 tLlxc5, and 22 . . . J:.xbS by 23 ltJ2c3 :a5 24 lDa4. However, in such a position there is more than one way to win. After all, apart from Black's other problems, he is 1 2 3 d4 e4 c4 g6 i.g7 I avoided my opponent's tacit invitation to go into the Pirc-Ufimtsev Defence, which would have arisen after 3 15 As is usual in such set-ups, White does not object to . . . f5-f4. Then the pawn offensive by Black on the king side ( . . . g6-g5, . . . h7-h5 and . . . g5-g4) is a very remote prospect, whereas White's attack on the queenside (b2-b3 , a2-a3 and b3-b4) is of more conse quence. The exchange made by Black allows White to seize control of the central e4 square. 3 li1c3, since I did not object to playing White in the King's Indian Defence. 3 4 5 6 lhc3 .i.e3 d5 d6 lhd7 e5 The game inevitably transposes into an unfavourable version for Black of the Samisch Variation. This is how they played the King's Indian Defence several decades earlier. 6 • • • 12 13 14 15 a5 Forced ! The usual continuation here, . . . c7-c6 and . . . c6xd5, no longer works: the d6 pawn is en prise. 7 .i.d3 The bishop is aiming for c2, where it will occupy an ideal position, control ling the important squares b3 , a4 and e4. 7 8 9 .i.c2 f3 tDc5 lhf6 Thus the Samisch Variation has been reached by transposition. 9 10 11 12 1id2 lhge2 0-0-0 0-0 tDh5 f5 tDxe4 .i.xe4 lhc3 fxe4 lhxe4 .i.f5 �6 Carelessly played. 1 5 . . . 'i'd7, with the threat of 16 . . . lt1f6, was undoubtedly more sluewd. In this case White would not have achieved any real gains from 16 i.g5 i.f6 1 7 g4 .i.xg5 1 8 'ifxg5 .i.xe4 19 �xe4 liJf4 20 et1f6+ :xr6 2 1 1i'xf6 'ifa4, when for the exchange Black has the initiative. 16 .i.g5 White exchanges the enemy knight, after which he is guaranteed control of the central e4 square, and Black's king's bishop remains passive. 16 17 18 19 .i.xf6 �bl iie2 'ifd7 i.xf6 'fle7 i.g5 This attempt to activate the dark square bishop does not lead to anything, since it is only able to control squares that are of no great significance. White has only one concern - not to allow it onto the a7-g l diagonal. As regards his subsequent plan, it begins with the preparation of h2-h4, in order to further restrict the dark-square bishop and prepare the opening of the rook's file. 3 16 26 27 28 �xh7 l:.hl+ g4 <it>xh7 <it>g8 Now Black is forced to exchange his bishop, after which the knight gains for ever the key e4 square. If the bishop re treats (28 . . . .i.d?), then White's pressure increases (29 'i'h6). . 20 21 22 23 1flel h4 1fe3 h5 hxg6 111g7 30 31 32 il.e7 l:.ti Wf6 hxg6 Undoubtedly a mistake, leading to Black's defeat. He definitely had to re tain his h7 pawn; he should have played 24 . . . .txg6 (25 l:.h6 .txe4+ 26 tDxe4 Wf4), when he would still have retained the space necessary for manoeuvring. 25 :b2 l£ixe4 11'b3 .i.xe4+ Wg7 Provoking new weaknesses on the light squares. After 23 . . . g5 the weakness of the l ight squares in Black's position be comes more marked. 24 28 29 30 :h7 b6 :rs .i.d8 Black has ended up in a passive, but apparently solid position. Naturally, before the time control White does not undertake anything active. 33 34 35 36 a3 11'c2 <&ti>a2 1Fh2 'tif7 1f g7 ci;f7 I avoided playing b2-b4 : the opening of the position could have given Black counterplay. 36 3 17 1ta4 l:thJ <Ji>g8 37 38 39 1ihl llh2 @bl ltf4 :rs The only constructive decision that I took in time trouble was to take my king to e2, where it is safe, and will also defend the f3 pawn. 39 40 41 42 @c2 @d3 <i>e2 ltf4 :rs l:lf4 :rs The plan for converting the positional advantage had already been worked out in the interval, of course, but for the moment practical prudence required me to accumulate time for thought, in order to avoid getting into time trouble before the next control. 43 44 45 46 47 48 49 50 51 I[hJ b3 lth6 lth2 1i'g2 1fh3 1rg2 11ht D.h6 lif4 :rs llf4 :rs :.r4 :rs D.f4 :rs l::tf4 There is no point in waiting any more. Everything is prepared for zugzwang. 52 g5! This apparently risky move is de cisive, since Black is unable to prevent the activation of the white queen on the h3-c8 diagonal. 52 53 54 55 56 :rs 1fh4 11t'gJ 1fg4 'ifh3 l:tf4 :rs l:f5 Now the time has come for the decisive manoeuvre. Black is forced to move his bishop, which exposes his back rank. Then by ltJg3 White either wins the exchange, or, as occurred in the game, wins the queen and several pawns for rook and bishop. 56 • • . JLe7 But not 56 . . . .i.xgS because of 57 �gs l:xg5 58 'i'c8+ <l;f7 59 'i'd7+, 60 'i'd8+ and 6 1 'i'xg5, winning a rook. 57 tLlgJ :xg5 If 57 . . . e4 58 ll'ixf5, and although Black's queen penetrates into the white position, it is only able to give a few checks. 58 1ie6+ �f8 If 58 . 1ff7 there follows 59 l:.h8+. rJ;n 59 ifc8+ . . 60 l:h8 60 61 62 1i'xh8 1lxb8 1lxg3 <ifi>f2 Black resigns Threatening 61 'ife6 mate (60 . . . 'i'f6 6 1 'i'g8 mate). If 62 . . . l::t g5 there follows 63 'i'c8, and almost all the black pawns fall. 3 18 Championship (Hamburg 1 965), I very much wanted to win. Therefore I decided to continue for the moment the mobilisation of my forces, in the hope that possible subsequent inaccuracies by White would enable me to decide on a successful plan. Game 356 Gligoric-Botvinnik Palma de Mallotca 1967 Nin:izo-Indian Defence 1 2 3 4 5 6 7 d4 c4 ll'lc3 e3 ..td3 ll'lf3 i..d2 ll'lf6 e6 _.i. b4 c5 b6 .ilb7 . 10 The strongest move is considered to be 7 0--0 (cf. , for example, Game 3 17), but Gligoric employs his favourite manoeuvre, although here it looks artificial (White provokes the exchange of Black's king's bishop, without weak ening his pawns). 7 8 9 a3 i.. xcJ dxc5 Only later, when annotating this game, did I learn that 10 0-0 would have led to a position from the game Gligoric-Smyslov, played earlier the same year in Havana, in which after 1 0 tlJe4 1 1 i..xe4 i..xe4 12 ltld2 i..g6 Black's position was preferable. . . . 10 . . • bxc5 Although White' s forthcoming pawn offensive on the queenside seems quite threatening, Black is hoping for counter play with his central pawn majority and good queen's bishop. 0-0 i.. xc3 11 12 0-0 b4 ll'lbd7 The expected offensive begins. However, this move was more or less obligatory, in view of the possible advance of the black .pawn to a5 .12 11c7 . • • • 13 · .J:.ct lDe4 After completing the mobilisation of his forces. Black embarks on active play. If White replies 14 Jib2, then after 14 a5 1 5 b5 e5 both of his bishops will be insufficiently active. Therefore Gligoric decides to exchange his light square bishop, so as not to allow . . . a7a5 and to gain time for switching his other knight. to the queenside, which appears very tempting. . . . 9 • • • d6 Perhaps the immediate 9 l2Je4 is the simplest, but from my tournament position, and after that terrible defeat that I suffered in my game with Svetozar in the European Team ... 3 19 14 15 .i.xe4 liJd2 -*.xe4 when the exchange sacrifice leads to an obvious advantage for Black. 18 . • • e5 Since the exchange 1 9 bxc5 dxc5 clearly does not come into White' s plans (he i s intending to play b4-b5 followed by CDc6), his bishop is now securely shut out of the game, and Black can calmly develop his initiative on the kingside. 19 20 • • . /i)bJ Exploiting the fact that 16 . . . cxb4 1 7 axb4 'i'xc4 i s not possible because o f 1 8 Ji.xg7. 16 17 18 lDa5 1fg4 22 23 IUdl Ael �df8 .i.g6! This is also what Smyslov played in the afore-mentioned game. This non standard retreat simultaneously ensures Black's safety both on the kingside (the g7 square! ) and on the queenside (the b l square!). Here White should have dis played the necessary caution and not deprived his king of an important defensive piece - his knight. But Gligoric� who was always consistent in the implementation of his plans, here too remains true to himself. 16 f5 f4 With the terrible threat of 2 1 . . .f3 . lbf4 21 exf4 White seems to be quite alright: he is not threatened in any way, on the queenside he has the initiative, and the opposite-colour bishops guarantee him a secure position in the endgame. 15 b5 11e2 .n.ad8 /i)b6 23 • • • /i)cS! Effectively the decisive move. The base of the pawn chain c5-d6-e5 (the d6 pawn) is securely defended, the black queen is switched to the kingside, and White does not succeed in bringing his knight back to the defence of his king. 24 25 White is forced to remove his queen from the d-file. If 18 b5 there could have followed 1 8 . . . d5 1 9 tiJc6 /i)xc4 ! , 320 /i)bJ lLld2 1if7 e4 There is no satisfactory defence against the threat of" 26 . . . Ji.h5. With an exchange sacrifice . White tries to complicate the play, but in vain. 26 f3 was preferable. 26 27 28 thn. Wc2 llxdl Game 357 Botvinnik-Matulovic .1lb5 ..lxdl ltJe7 Palma de Mallorca 1967 King's Indian Defence Bringing the knight into play in the simplest way. The d6 pawn is immune, since in the variation 29 l:lxd6 lDr5 30 .6td5 e3 ! 3 1 f3 ltJd4 White.'s position cannot be defended. 29 30 31 i:tJf5 Vxf5 i:tJe3 ihxf5 l1xd6 1 2 3 4 5 d4 c4 d5 ltlc3 e4 lt)f6 c5 d6 g6 In my view, in this variation the fianchetto of White's king's bishop is less promising. 5 6 7 .i.g7 0-0 i.d3 h3 This modest move leads in fact to a new and original variation, which I had previously prepared in home analysis. However, this was the first time I had tried employing it. The next game on this same theme, with Kavalek (No. 372), was played a year later. 31 7 8 9 10 e3 • • . e6 exd5 :eS+ i:tJfJ exd5 .*.eJ Taking play into a prosaic endgame. 32 33 34 35 36 exf2+ ll8xf5 :xc4 b5 lla4 1f xf5 iLxf2 :d2 bJ h4 In this way Black wins another pawn, which is the quickest way of convincing the Yugoslav grandmaster of the futility of further resistance. 37 l:td8+ <it>h7 38 :d3 l:a5 39 llb3 , · c4 40 Ac3 lifxb5 41 lixc4 l:ha3 42 i.d4 a5 White resigns "�- ;. .. r ..'· ·. 32 1 �·() '·< . \·� .. , i:tJh5 It is doubtful whether 10 . . . l2Ja6 1 1 0-0 !Dc7 1 2 a4 ti::Ja6 1 3 :!c l ti::Jb4 14 ii.b l , which occurred in the games Balashov Stein ( 1 97 l ) and Balashov-Suetin ( 1 972) is any stronger for Black. In the first of these after 14 . . . a6 1 5 ilf4 Vlic7 16 ifd2 White had a clear advantage. The sharp move 10 . . . b5 leads to interesting complications, the simplest reply apparently being 1 1 cxb5. lLld7 11 0-0 It is doubtful whether the committing 1 1 . . .f5 was any better. 12 active enemy piece - the dark-square bishop, which also deptjves the black king of a reliable defender. 15 . ti)dxf6 • 16 The exchange of all the rooks will assist in reducing Black's activity, and will also make it easier for White to find the key to the enemy king's position. It was no accident that I chose this method of play, since I knew that Matulovic did not like defending passive positions, and I was hoping that imperceptible posit ional errors by Black would accumulate. i.g5 16 17 18 19 Black cannot tolerate this bishop here. 12 . f6 The most natural reply, but then the f-pawn has to be advanced further, and this restricts the bishop at c8 and weak ens the black king's position. . 13 • .i.d2 .i.gS ifxel ifd2 :et lixel+ i.d7 iff8 l::te8 Black has no right, of course, to leave the open file in White's possession, and no amount of counterplay on the queen side could have compensated for this. f5 20 21 22 23 White was threatening 14 g4. 14 • ktel i.f6 lbe8 a3 b4 bxc5 1ixe8 �g7 b6 The open b-file does not present any danger to White. Moreover, in some cases his queen may be able to invade the enemy position along this file. 23 24 gJ bxc5 h6 24 . . . !De4 would not have achieved anything for Black after 25 'i'b2. 25 'i'c2 1fic8 � 26 �h2 15 27 28 .i.xf6 So, the persistent forays of White' s bishop to g 5 have borne fruit. He has succeeded in exchanging the most 3 22 'i'd2 ltlgt <l;g7 White prepares f2-f4, which will further restrict the black bishop, gain control of e5, and, most important, 38 39 subsequently strengthen his attacking possibilities when the g-pawn advances. 28 29 30 f4 <it>g2 1fe8 Cj)fJ lLlh4 9'c2 I had altogether no active plan,. and he tL\g8 was sure that before the time control White would not even be looking for one. This lack of vigilance led to Black making a decisive mistake. Now the white g-pawn gains the opportunity to advance immediately, and the game quickly comes to an end. However, in any case White would have .retained the advantage and all the same he would have achieved the advance of his g pawn, after first transferring his king to the safer square a2. 'ii'b8 lLle7 Defending the b3 square and i ntensifying the pressure on the bl-h7 l l iagonal. 33 34 35 36 lLle2 Cj)fJ 9'b2 In the three previous moves, and also White's subsequent moves, one should 11ot seek any deep point. 111ese are waiting tactics - when short of time for I hought. And in general I had to display caution. since towards the end of the fifth hour of play the probability of mis lakes by players of the older generation increases. 36 37 lLlcJ 9a5 lDc8 lDe7 ti)c8 My opponent obviously thought that � The time has come to transfer the white knight to an active position. 31 32 33 lLlh4 �f2 40 41 g4 g5 l:f)e7 This move was tlllderestimated by Black. He was thinking only about the safety of his f5 pawn� but forgot about those threats arising from the insecure position of his king. 41 42 lhe2 /:j)fg8 h5 After 42 . . . hxg5 43 fxg5 White would have gained the f4 square for his knight. Now Black does not fear 43 ti'h8 in 323 view of 43 . . . Wxa3, attacking the bishop at d3 . But after analysis during the dinner break, it all ends quickly. The closed nature of the position cannot save Black. 43 44 45 46 @fl ltJgJ .i.e2 49 1rg7+ Black resigns (49.. . :�e6 50 l2Jf8 mate). Game 3 58 .i.c8 .lld 7 1i'a4 ..ie8 Donner-Botvinnik Palma de lt.,fal/orca 1967 Griinfeld Defence 'iit>f2 The sacrifice on g6 can no longer be avoided (White is threatening 47 lbxg6 lbxg6 48 lbxh5, and if 47 . . . �xg6 48 'i'h8 rj;i/ 49 lLJxh.5)� the position of the bishop at e8 changes little. 1 2 3 4 5 d4 c4 g3 .i.g2 cxd5 lhf6 g6 c6 d5 This seemingly unpretentious ex change nevertheless requires Black to defend accurately. Then he obtains an equal game. 5 6 7 t£ifJ cxd5 .i.g7 • 0-0 0-0 For 7 lLJc3 0-0 8 tLle5. cf. Game 254. 7 8 • . l2:\c3 Or 8 �e5 ltlg4, and Black achieves full equality. 47 ltlxg6 47 48 ... l2:\xh5 It was noticeable that this move came as a surprise to my opponent. He has little choice: 47 . . . �xg6 48 'i'h8, 47 . . . lLixg6 48 liJxf5 , or . . . lt\xd5 If 48 cxd.5 �xg6, then Black would have been threatening to exchange queens, and also in some cases the f4 pawn would have been en prise. But now the g7 square cannot be defended. 48 • • • ti)de7 8 9 ... ifb3 l2:\c6 ltJe4 10 ltdt ltla5 As was pointed out in Game 254, at one time I was afraid of reaching this position with Black. However, my fears regarding the fate of the d5 pawn were unfounded. It turns out that White cannot take it with either his queen ( 10 'i'xd5 lbxc3), or his knight ( 10 lbxd5 ..i.e6). Therefore 9 'iib3 loses its point and is an opening mistake. The normal continuation for White is 9 tbe5 e6 10 lDxc6 bxc6 1 1 ..tf4 ·�r l I ltJa4. 324 Again Black need not be concemed about the capture of the pawn - 1 1 'fixd5,. in view of l l . . . llJxc3 . 11 14 15 16 .. 18 19 lDel lixd2 1fxd2 lDxcJ 'ifxc3 Of course, the queen is badly placed on the c-:file, but after 12 bxc3 the black knight would have become firmly established at c4. 12 13 14 · Threatening 17 . i.:d3 - 1 8 .i.xf3 j,,xd4 . 1fb4 17 i.e3 Since White has managed to defend against the immediate threats, and the position is symmetric and the open c file is a convenient springboard for the exchange of the heavy pieces, Black himself prefers to provoke an endgame, on the grounds that he is "Qetter mobilised and that White's queenside is slightly weakened. It is also worth remembering that I could not be too aggressively inclined, since before this I had won three games in a row. 1lb4 Curiously, my opponent had already reached this position in games played both with White, and with Black. Against Barcza (1 958) he played 1 1 . . . b6. and was unable to exploit White 's mistake (9 'i'b3). Therefore, not long before our game, he himself repeated it with White against Filip, where it all turned out well for White� since with l I . . .t2Jc6 Black immediately demon strated his consent to a draw. On this occasion, however, White does not get away with his opening inaccuracy. 11 12 Vb6 lDc6 .*.e4· bJ 'i' d2 .if4 'ifel .i.f5 llc8 20 21 Only here will the white queen feel fairly secure. e5 19 Black hurries to open up the position, while the rook at al is not yet in play. dxe5 l:.cl .ixe5 l£ib4 Black consolidates his hold on the c-file, but, unavoidably, he is forced to concede the d-file. 325 22 l:txc8 23 · .i.xe4 24 l:td7 l:lxc8 dxe4 anything specific. Even so, after gaining the better of the opening;': I . have merely managed to retain the more pleasant position, which, however, must be considered drawn. Well, - such things happen. Since the c 1 and c2 squares are defended, White can pennit himself this slight activity. Of course, he could not take the pawn (24 i.xa7 :c l). 24 25 26 �xc7 a3 :c7 iJ.. xc7 • • • lhc2 l£ic6 i.e5 Restricting the white knight. Black ' s advantage i s that after . . . f7-f5 his king will occupy a favourable position in the centre of the board. 28 f3 � i.d2 a4 �g2 37 lDb4? e3+ White resigns If in this game I did not exploit my chances as well as possible, it should be borne in mind that this round was held on the island of Menorca, where the enthusiastic chess fans forgot that for the players the most important thing is quietness during the game . f5 The e4 pawn must be retained: it restricts the activity of White ' s king and knight. 29 30 31 32 fxe4 g4 hJ <t>f2 Throughout the game my opponent has skilfully defended, but here, affected by shortage of time for thought, he commits a vexing oversight and loses a half point. And now if 26 i.xa7 there would have followed 26 . . . b6. 26 27 fxe4 ..te5 /,£je7 /,£jd5 33 34 35 36 �7 @e6 .i.d4+ a6 Game 3 59 Botvinnik-Larsen Palma de Mallorca 1967 Reti Opening 1 2 3 4 5 6 Even now� when I analyse this game, I find it hard to reproach myself for c4 tfjtJ g3 .ig2 0-0 b3 /,£jf6 e6 d5 i..e7 0-0 c5 This is Black ' s most active plan. 6 . . . a5 has also been successfully played. Long ago . (Levenfish-Bogol jubow, 1 924) White even ended up with 326 the worse game after 7 .tb2 a4 8 d3 c6 4!bd2 CDa6 10 d4 a3 . Possibly, accord ing to analysis by Euwe, the most effec1 i vc is 7 .i.b2 a4 8 bxa4 dxc4 9 CDe5 . Regarding 6 . . . c6, see game 300, and ror 6 . . . d4 - Grune 230. lJ 7 i.b2 lbc6 It would be appropriate to mention t hat this position occurred .long ago in t he game Capablanca-Marshall (Mos cow 1 925), where in reply to 7 . . . d4 t here followed 8 d3 CDc6 9 e4, although t he modem view is that 8 e3 is correct. 8 eJ b6 Black can also deploy his queen's bishop differently: 8 ... .td7. In this case for the moment his position does not hold any promise of counterplay, but it is sufficiently sound. 1 8 .i.e5) Black would have maintained a defensible position. 9 . . • .i.b7 9 . . ..i.a6 10 'iie2 llc8 has been played against me on more than one occasion. Against Bakulin ( 1967) I correctly con tinued 1 1 thb5, whereas after 1 1 llfd l , as in my game with Van Scheltinga ( 1 966), by l l . . . d4 12 exd4 cxd4 1 3 tLlb5 43 Black could have gained suf ficient counterplay. 10 d3 10 11 12 13 :ct ife2 l:tfdl If immediately 10 'iie 2 there could have followed 10 . . . d4 1 1 exd4 cxd4 12 lbb5 d3 , and Black has a perfectly good game. I also did not want to hurry with IO cxd5 and 1 1 d4. since with 10 . . . CDxd5 followed by . .· . .tf6 Black would have equalised. lic8 J:lc7 ll.d7 lie8 The immediate 1 3 ... 'i'a8 and then 14 . . JHd8 was better. 14 15 9 cxd5 lbxd5 ti:)xd5 l:xd5 lbc3 It is doubtful whether the immediate manoeuvre 9 ii'e2 .ii.b7 10 l:d l is preferable. For example, in the game Vaganian-Karpov ( 1 97 1 ) after 10 . . . 'iic7 1 1 tLlc3 ltad8 12 cxd5 ltlxd5 1 3 tLlxd5 )�xd5 1 4 d4 cxd4 1 5 tLlxd4 tLlxd4 16 i.xd4, with 1 6 . . . :d7 (instead of 1 6 . . J id6, with the sequel 1 7 l:tdc l ! ii'd7 327 Here the rook is badly placed, which White is immediately able to exploit. 1 5 . . . exd5 was sounder. 16 although there was also a combinative solution: 2 1 l£if6+! gxf6 i2 1Wg4+ �h7 23 :d7 ! During the game I only had time to consider 23 .i.e4+ f5 24 i.xf5+ exf5 25 ifxf5+ @g8, after which I would have had to give perpetual check, since if 26 'ii'f6 there follows 26 . . . l£ie5 27 i.xe5 1he5 . But after 23 :d7 ! :e7 (or 23 . . . i.e7 24 .i.e4+ 'it>h8 25 :xe7) 24 i.e4+ f5 25 .i.xf5+ exf5 26 'iixf5+ �g8 27 'i'f6 Black is no longer saved by 27 . . . tl:Je5 on account of 28 i.xe5 laxe5 29 Vxt7+. d4! In the given situation the opening of the position is in White's favour - his bishops are more active than the opponent 's. 16 1i a8 Now such strategy is fraught with unpleasant tactical dangers. As in the Vaganian-Karpov afore-mentioned game, the rook should have immediately been retreated to the seventh rank. llxdl + 17 dxc5 • . • 18 19 ktxdl ti)g5! 19 20 l£1e4 i.xc5 The possibility of this move, with the threat of 20 'i'h5, and hence the need for Black to immediately drive back the knight, thereby weakening his king' s position, demonstrates the faultiness of my opponent' s strategy. h6 .tf8 21 22 23 21 ltd7 Despite the fact that I was already beginning to run short of time, I was able to find quite a strong move, . • . f5 Larsen wants to parry the threat of 22 GDf6+, but it turns out that all the same this could have been played: after 22 . . . gxf6 23 Wh5 �e7 24 'ifg6+ Black is not saved by either 24 ... .%lg7 25 :xg7+ 3 4 iixg7 26 ..ixc6 ..ixc6 27 .i.xf6 'if:IB 28 i.xg7 'iixg7 29 1Wxe6+ and 30 1Wxc6, or 24 . . . .i.g7 25 ..i.xc6 :xd7 26 .i.xd7 i.hl 27 .i.xe6+ �h8 28 f3 .i.xf3 (28 . . . 'i'xf3 29 ile8+) 29 .txf6. I did study another way to win - 22 1ih5 .:e7 23 :xe7 l:tJxe7 24 l:tJd6 ! i.xg2 25 Win+ 'it>h7 (25 . . . 'it>h8 26 tl:Je8), but due to shortage of time I failed to win the winning continuation 26 i.xg7! i.xg7 27 ttJe8 Vxe8 28 ifxe8, when Black loses one of his pieces. l£1d6 lixd6 .txd6 This is the decision of a practical player, preferring in time trouble to aim for a positional win, rather then combin ative obscurity. After briefly examining 23 ];Ixg7+ 'it>fS, I did not like the fact that after 24 1i'h5 l:e7 25 l::tg6 l:tJd8 the black king would be able to escape to 328 30 31 32 33 t he queenside, and I failed to find the quiet intermediate move 24 l;Ih7 ! �xd8 'i'c2 11c7+ rM7 11fxd8 1fd5 Here White should have prevented the switching of the queen onto the a6fl diagonal by 33 a4. However, in time trouble the temptation to give checks was too great. 33 34 23 24 25 :xd4 l:td7 l£id4 .i.xg2 Ji.bl 26 f3 l:id8 27 l:txg7+ <ifte8 1fb8+ Even after the better 25 . . . .i.h l 26 f4 White should also win. 26 . . . e5 was Wlsatisfactorv " because of 27 'i'c4+ 'iti>h7 28 'fic7 Ilg8 29 .i.xe5. Studying the variation 27 'i'd l :xd7 28 •xd7 'i'f8, I rightly judged that 29 ffxe6+ 'iif7 was insufficient for a win, hut I missed the fact tliat by playing 29 't\Yxa7, then a2-a4 and b3-b4, White, despite the loss of his bishop, would then be able to exchange queens, and his a-pawn would be unstoppable. <i>t'8 27 28 29 30 34 Another winning possibility was offered by the cahn 30 .tc3 followed by 3 1 !lxh6, not fearing the invasion 30 . . . 'i'hl in view· Of 3 1 nh8+ rt;e7 32 .i.b4+ �d7 33 'i'd3+ • • 'iti?d7? . . . 'i'd5 ii'dl llh7 @fl :hs+ • Larsen still had some 20 minutes left on his clock, but because of his im patient character he hurried his move and . . . missed drawing chances which after 34 1id8 could have arisen in the ending, thanks to the opposite-colour bishops. 35 36 37 'i'xa7+ 1fa6+ !fc4+ �c8 �c7 Now White himself goes into the endgame, but during this time he has won another pawn. . 329 37 38 bxc4 39 . i.d4 11xc4 'it?c6 h5 40 a4 Black can only move his king. and its forced retreat allows White to create a passed p awn. 40 41 42 43 44 c5 .ixc5 .1lb4 '3;c7 bxc5 <lt>c6 <lti>b6 g4! �g3 Monte Carlo 1968 Queen's Indian Defence 1 2 3 4 5 6 This breakthrough was immediately pointed out by Smyslov during a brief analysis before the resumption. 44 45 Game 360 Botvinnik-Padevsky hxg4 e5 Now� of course, Wlrite wins very easily. But Black was powerless against the threat of 46 fxg4 fxg4 47 e4 followed by e4-e5. Then the king goes to f6 and the bishop to g3 . as a result of which the e6 pawn is won. 46 . e4 fxe4 47 fxg4 Black resigns It remains to add that some of the combinations given in the notes� which remained 'off-stage', were pointed out by Gligoric, Kotov and Flohr. 330 tt)fJ c4 g3 �g2 0-0 �f6 e6 d5 !JJ..e7 0-0 d4 With 6 b3 it was still possible · to continue playing the Reti Opening, as, for example, occurred in Game 359. However, the Catalan Opening, a position from which is reached after 6 d4. was an opening that I more rarely employed, and this is also a factor of no small importance against an inexper ienced opponent. However, with this the opening metamorphoses are not exhaus ted, and soon the play moves onto lines typical of the Queen's Indian Defence. 6 7 b3 7 8 .i.b2 l£ibd7 7 'S'c2 is more often played. b6 Ab7 9 cxd5 At just the right time. Since after 1> �xd5 10 tt:lc3 or 9 . . . l2Jxd5 (in a si milar situation this is what Donner played against me - Game 320) 10 l.i)bd2 Black to some extent loses control over the centre, he considers it obligatory to take on d5 with the pawn. I 'ut in this case the activity of his queen' s bishop is reduced. · . . . . 9 10 • • • lDcJ exd5 l:te8 It should be mentioned that the i mmediate 1 0 . . . c5 would have allowed. White after 1 1 dxc5 bxc5 1 2 tDh4 liJb6 ( Konstantinopolsky-Belavenets, 1 937) Io gain a solid advantage by 13 lD:f5 . l lut after the systematic preparation of t his advance - 10 . . . a6 1 1 l:k l .i.d6 12 d l:e8 1 3 lDe2 lbe4 1 4 'i'c2 c5 < Andersson-Matanovic, 1 976) Black's chances are not worse. 1 1 li)e5 If White declines this opportunity, he will not have any good prospects at all. rhe point of the manoeuvre is that after I 1 . lLlxe5 1 2 dxe5 t:De4 1 3 llixe4 dxe4 1 4 'i'c2 (as Euwe played in an analogous situation against Capablanca, AVRO Tournament 1938) Black gets into difficulties. · . . 11 ... f4 13 ttlxe4 13 14 15 e3 In view of the threatened exchange 1 3 . . . ttJxc3 1 4 .i.xc3, after which White has a backward e2 pawn and a weak point at e4, this decision is obligatory. dxe4 ttlf6 a3 An essential move. White is aiming to restrict the activity of Black's king's bishop and in some cases to take control of the c6 square by b3-b4-b5 . 15 16 17 18 1fe2 �xd4 b4 c5 cxd4 1fe7 a5 Black must clarify the situation on the queenside. By forcing b4-b5, he sub sequently secures the c5 square for his knight. .i.d6 If 1 1 . . . .i.f'8 White could have continued either 12 f4, or 12 l:k l lDxe5 1 3 dxe5 :xe5 1 4 lbb5 :!e7 1 5 .i.xf6 gxf6, and here in the 1 1th game of my match with Bronstein ( 1 95 1 ) I could have developed my initiative with 16 b4 followed by liJd4 and b4-b5 . 12 12 The sharper 1 2 . . . c5 occurred in the game Larsen-Gulko ( 1 976). 19 b5 -*.xe5. Since 19 .. �xa3 20 .i.xb6 favours White, this natural exchange enables Black to defend his "b6 pawn and simultaneously to rid himself of the . · 33 1 annoying knight at e5, which at a propitious moment could have occupied the key c6 square. 20 .*.xe5 Why does White avoid the natural reply 20 fxe5 ? Because after, for example, 20 . . . ltJd7 2 1 ifb.5 g6 22 'ifh6 ttJxe5 23 l:tf4 (23 .i.xe5? 'ifxe5 24 :xn? 'S'xal+) 23 . . .fS it is not apparent how the attack can be continued. Now, however, White retains some advantage, although his hopes of a swift attack are reduced to the minimwn. 20 ltJd7 tbc5 21 il.d4 ltld7 22 f5 A superficial move, which leads to a rapid defeat. It was essential to play 22 . . . f6, when Black could have put up a prolonged resistance. True, here too White would have retained the advan tage, since in the endgame his chances are better. In particular, the position of the knight at c5 is insufficiently secure. after 23 . . .liJxf6 24 l:txf6 gxf6 25 'ifg4+ the picture does not change. 23 1fe6 24 'ifh5 White, of course, does not huny with the capture of the g7 pawn, but intensifies the threats. Now if 24 . . . gxf6 there follows 25 �h3, while if 24 . . . lhxf6 25 l:Ixf6. 24 lhe5 25 l:lf5 The rook is transferred to · an attack ing position, and the knight is driven from its key square, in order to expand the scope of White's centralised bishop. 25 thg6 25 . . . g6 cannot be played because of 26 'ii'h6. and 25 ... ttJf3+ is also bad in view of 26 .i.xf3 exf3 27 Wg5. 26 fxg7 Tirreatening mate in four moves. Black's reply is of course unfortunate, but one cannot help remembering Tarrasch, who wrote that in a bad position all moves are bad! 26 llad8 23 f6! After 23 ... gxf6 · White's advantage is obvious: 24 'ifg4+. <&t>h8 (24 � 25 ifh4) 25 l:lxf6 lDxf6 26 'iig5, while · . . . 332 • 27 • • 'ihh7'+ Bla:ck: resigns Gmne 3 6 1 Botvinnik-Larsen Monte Carlo 1968 King's Indian Defence 1 2 c4 lllf3 ·lt)f6 . g6 In contrast to- our game in Mallorca ( No. 3 59) Larsen decides to fianchetto his king's bishop. 3 4 5 6 g3 �g2 0-0 �CJ 8 llg7 0-0 d6 c6 Black usually plays 6 . . . tbbd7 followed by . . . e7-e5, or 6 . ..ttJc6, aiming a fter d2-d4-d5 to play the laright to a5 . Here and on the following move my opponent is clearly aiming for a less common and not so well-analysed continuation. 7 d4 1!fa5 At one time 7 . . . a6 8 e4 b5 was fashionable, but after 9 e5 liJe8 10 'ife2 White has a serious spatial advantage. 8 White gained an advantage by 8 d 5 'ib4 9 ttJd2 .i.d7 IO e4. I also did not know that White usually prevented the bishop move to g4 by playing 8 ·hJ. '. But even if I had known, I would probably not have changed my decision. Can Black really solve his opening problems in such a simple way (7 . . . 'ii'a 5)? Then his task would always have been straight forward! • • • -*.g4 It can be shown that 8 . . . 'i'h5 would also have led to a difficult game for Black after 9 e5 ! : 9 . . . tbe8 I O .i.g5, 9 . . . tbfd7 10 exd6 exd6 1 1 .i.f4, or 9 . . . dxe5 10 lt:Jxe5 (10 . . . ltJg4 1 1 �). 9 l:el e5 10 b3 i.g4 1 1 d5 and then 'i'd3 is also good for White. 9 10 hJ .i.xf3 11 J..eJ .t.xf3 lllfd7 Black intends, by playing 1 1 . . . c5, either to gain control of d4 ( 12 dxc5 dxc5), or to win a pawn ( 12 d5 .i.xc3 ). However, this plan is impracticable due to the weak:Q.ess of the b7 pawn - the absence of the light-square bishop tells. e4 I have to admit that I had not seen the game Ivkov-Larsen ( 1 965), in which 333 11 • . • c5 During the game I thought that 1 1 . . . 'iib4 was stronger, since after 1 2 'i'd3 ( 1 2 'lb.3 ifxb3 1 3 axb3 tLla6) 1 2 . . . 'i'xb2 13 %tabl 'ii'a 3 14 .:xb7 tbb6 the white rook might end up in a dangerous position. Just in case, I even prepared the variation 12 .i.e2 W'xb2 1 3 tba4 'ili'a3 1 4 i.c l 'i'b4 1 5 .i.d2 with a repetition of moves . . . In fact after 1 2 'i'd3 li'xb2 1 3 l!ab l 1i'a3 14 .r:.xb7 tbb6 the white rook turns out to be a troublesome prisoner: 1 5 .i.e2 ! (but not 15 l::txe7 .i.f6 1 6 :b7 'i'a6 17 e5 'i'xb7 1 8 exf6 ti:J8d7) 15 . . .'i'a6 ( 16 ti:Jd5 was threatened) 1 6 Ilc7 ! , and there i s no way of attacking the rook. 12 dxc5 13 e5 Of course, I repeated moves merely to gain time, since · I already had in mind the following decisiv� continuation. 18 b4! 1 8 e6 fxe6 19 .ltg4 (19 tLlxe6 :xn) 1 9 . . . l:if6 would have been weaker. Now� however, the position is opened up, and White - who has mobilised his forces earlier - dominates the board. 18 l2Jxe5 dxc5 I should mention that in the game Lengyel-Honfi ( 1 968) the other capture 12 . . . tt:Jxc5 also led to a difficult position for Black: 1 3 e5 lDc6 14 exd6 l:tfd8 15 ti:Jd5. 13 • . . 'ilc7 After 1 3 . . . 'i'a6 14 ltJd5 .i.xe5 1 5 lbxe7+ @h8 1 6 'ifd 5 Black cannot avoid loss of material. 14 15 liJd5 l2Jf4 ifd8 Again attacking the b7 pawn and simultaneously threatening the un pleasant e5;..e6 . Black has only one defence. 15 16 17 liJd5 &Df4 ifb6 ifd8 irb6 19 .i.xc5 19 20 21 ifxf3 liJd5 A mistake, caused by an hallucin ation. After 1 9 tbd5 'i'd8 20 iLxc5 tLlbc6 2 1 b5 ifa5 (or 2 1 . . .b6 22 i..a3 tt:Jxc4 23 bxc6 liJxa3 24 :cl 'i'd6 25 c7 with the threats of 26 tbxe7+ and 26 �c6) 22 iLxe7 ti:Jxf3+ 23 'i'x:f3 tbxe7 24 &Dxe7+ �h8 25 l:.acl White remains a pawn up. But I decided to play an even stronger move ! The b7 pawn can be defended only by the queen, and this creates the pre conditions for the further development of White's initiative. lDxfJ+ 1fc6 l:te8 Here I noticed to my surprise that the previously planned 22 Ji.xe7 i.. xal 23 lhal did not achieve anything in view of 23 . . . tbd7, when the f6 square is defended. I also had · to reject 22 'i'e3 334 Jlxal 23 l:xal ttJd7 24 C/Jxe7+ .tixe7 25 .fl.xe7, when Black plays 25 . . . l!i)b6, if there is nothing better. 22 li:Jxe7+ %he7 23 'irxc6 ttJxc6 is bad for Wlrite. Meanwhile, I wrongly r�jected the quiet intensification of the · pressure by 22 Hae l e6 23 'i'e2 tiJd7. I did not like 24 4'Je7+ J:.xe7 25 ii.xe7 l2Je5, · but then 26 f4 was possible, when · Black has insufficient compensation for the lost exchange. Therefore I decided to go into a better ending. 22 23 lladl lDf6+ 9l.f6 11xd7 25 26 27 28 il.d4 ..txe5 if xe5 29 30 ltdl c5 1tc8 'i!fc6! This is what is most important for Black: control simultaneously of d7 and e8. e6 The whole point is that after 23 t:De7+ Ihe7 Black can answer 24 :ct8+ with 24 . . . l:e8 (25 'i'xc6 li:Jxc6) . 23 24 25 In this position, where White's queen is better placed, he has an ex1ra pawn on the queenside, and it is his tum to move, he cannot achieve either a material advantage, or the so desirable invasion of his rook on the seventh rank. For example, 29 'ifd5 'i'c8 30 l:e l 1lc7. 31 32 lld6 ifd5 1fb5 1ixb4 33 34 l:ld7 @g2 Vet+ Again the only move. After 32 . . . lte8 33 :ct7 .li:te l + 34 rt>h2 White gives mat�. first! ..i.xf6 lDd7 Alas, this is only an exchanging operation - the last lightning flash of the departing thunderstorm . . . 1'xd7 e5 !bes 1Wxh3 34 'ife6! Forcing the exchange of queens. as after 35 'ii'xb7 .:es 36 c6 'iife4+ 37 @h2 l:e5 White would already have had to think about how not to lose. 35 36 37 335 'ilfxe6 lhb7 lba7 fxe6 l:tc8 By playing 37 l:tb5 White could have hoped to win if his king were able to reach: e5, which in the given case is ruled out. 37 38 • • • White first moves his knight from c3 to . a central position. 8 · .0-0 When I played 7 . . . f5 I had thought to delay my kingside castling until White determined that his king would remain on its own wing, but now I decided not to fear queenside castling by the opponent, since waiting tactics had their drawbacks. It soon transpired that White had quite different intentions. • llxc5 <it>b3 . In order to answer 3 8 . . . hS with 39 <it>h4, not allowing 39 . . . g5 . :rs 38 39 f4 :d5 e5 40 . a4 41 a5 Draw agreed Game 362 Benko-Botvinnik c4 gJ .i.g2 g6 i.g7 e5 It now turns out that White is playing the Closed Variation of the Sicilian Defence with an extra tempo. 4 5 ttJc3 e4 l'iJe7 d6 If now or on the ne,.,,_1 move Black had played . . . c7-c5, this would have led to an ahnost symmetric position with a minimal advantage to White. 6 7 8 If 8 i.e3 replied 8 . . . lDge2 d3 ttJd5 . 9 .i.eJ i.e6 10 11 12 'W'd2 0-0 l:lael 9d7 IUi Another plan was chosen by Csom against Gulko ( 1 976): 12 l:tacl l:taf8 1 3 f3 <it>h8 14 b 3 l2Jg8 1 5 exf5 gxf5 ( 1 5 . . . i.xfS 16 d4) 16 f4. English Opening 1 2 3 • Strictly speaking, Black already has a lead in development over his opponent, who has spent time on c2-c4 and liJd5. In the game D .Byrne-Benko ( 1 966/67) he did not exploit this factor, and after 9 . . . h6 1 0 'i'd2 g5 1 1 f:[jxe7+ l2Jxe7 1 2 f4 he conceded the initiative. For example, 4 1 . . . exf4 42 gxf4, and the black king hurries to the queenside. Of course, there are also other ways to draw. Monte Carlo 1968 • ttJbc6 f5 Black would probably have �d4 . (9 lDxd4 exd4 ), so 3 36 12 13 l:taf8 f4 In this position I outlined the following plan, which is not without its positional novelty and psychological elements. First Black exchanges pawns on e4 (White is forced to recaprure with the pawn, so as not to let the enemy knight in at f5), and then the light-square bishops, in order to weaken the e4 pawn and the position of the white king, and finally he exchanges pawns on f4, opening the e-:file for a frontal attack on the e4 pawn (White will be forced to recapture on f4 with the pawn, so as to control the e5 square). When the rook attacks the e4 pawn from e8, White will probably play liJe2g3 , after which the black h-pawn · will advance, accentuating the weakness of the e4 pawn and the open position of the white king . . . 13 • • • fxe4 The correctness of Black's plan was indirectly confirmed in the game Schmidt-Ribli CD<i4 14 fxe5 dxe5 15 i.g5, and White has an appre ciable advantage. (1973): 13 . .. 14 dxe4 lhc8! 15 Ji.bl By defending the c7 pawn with his . rook. Black gains the opportunity to exchange the light-square bishops. 16 17 18 19 c5 b4 �xg2 gxf4 lDg3 . 19 • • h5 • Now what is White to do? 20 h4 looks the most logical (securing the position of the laright at g3 ), but then the h4 pawn would need defending. If 20 f5, then 20 h4 2 1 fxg6 (2 1 f6 i.118) 2 1 . . . hfl 22 :xfl hxg3 23 l:tf7 'ifg4 24 ltxg7+ 9;xg7 25 .th6+ �g6, and Black must win. But White also has an intermediate move at his disposal, driving away the knight from the central squares d4 and e5. . . . 20 b5 tfl6e7 Perhaps 20 llJd8 (defending f7) would have been more circumspect. Now great complications arise. ... -*.xg2 exf4 :eS 21 The psychological part of the plan also justifies itself: in defending the pawn, White aims simultaneously to cover his king's pOSition, but 1 9 l£lec3 was probably strom.:ger. However, · . · already at this point White had obvious ly foreseen the resulting complications, and he thought that the weakening of the black king's pawn cover would be bound to tell. 22 23 f5! fxg6 h4 :xn hxgJ .:.xn 24 :n If 24 .i.h6 Black has a good reply in 24 . . . We6, whereas now-:..24 . 1i'e6 would be met by 25 .:xg7+ r/;xgJ 26 .i.d4+ 3 37 . .. �g8 27 ltJf6+. If 24 . . .Vg4, then the same _sacrifice is decisive: 25 lhg7+ �xg7 26 •ct4+ <ifiixg6 27 'iff6+ @h7 28 iff7+ 'ilg7 29 liJf6+ and mate in two moves. 24 • • • .i.e5! By temporarily retaining his bishop, Black beats off the attack. 25 .i.d4 31 lU6+ 32 33 34 ci>f2 �h7 Now the checks collie to an end - 32 :f7+ @gs 33 :g1+ ct>fs. .i.xg3 cxd6 1l'd3+ · · •xb5 9'xe8 White resigns 1fg4 Game 363 Botvinnik-Portisch Monte Carlo 1968 English Opening 26 :f4 Of course, stronger was 26 .i.xe5 gxh2+ (26 . . . 1i'xe4+? 27 <ifilh3) 27 .i.g3 I (27 �xh2 'l'h5+ and 28 . . . 'iixg6+), but �en so after 27 . .. 1ixe4+ 28 �2 'i'xg6 (if 28 . . . WfxdS or 28 . . . ltlxd5 there . follows 29 'ilh6) 29 liJxe7+ l:Ixe7 ! 30 l:xe7 tfJx.e7 3 1 cxd6 ltlf5 32 ii'd5+ �f8 White is a pawn down, and he has a difficult task to draw. 26 • . • 1.ib5 Now, if White parries the threat of 27 . . . 'i'xh2+, the reply 27 . . . l£ixd5 will put an end to his attack, and Black is after all a piece up! 27 28 29 30 J.xe5 'it>f3 tLlf6+ ��+ In our day it is almost impossible to win in the style of the old masters. For this you need your opponent to be in the wrong frame of mind and psychologic ally not ready for a battle. Usually with Lajos Portisch that is in fact what happens - once or twice over the course of a tournament. In particular, this told duritlg the following old-fashioned game. 1 2 3 4 5 c4 thcJ gJ cxd5 .i.g2 5 6 thfJ e5 �6 d5 tt:lxd5 In this variation of the Englisll Opening White' s plan includes pressure on the h l -a8 diagonal and along the file . . . But which file - this depends on Black: either the c-fi.le (if· the knigh1 simply leaves d5), or the b-file (if the knight on d5 will be exchanged for the! knight at c3). 'lfxh2+ 'ifxd2 �g7 �xg6 7 8 338 0-0 dJ J.e6 tt:lc6 tLlb6 opening the c-file and in some cases of occupying c5 with a minor piece. Black, possibly again aiming for a draw, exchanges lo:lights, ·1osing a tempo in development. Meanwhile, he had available the energetic reply 1 1 . . . ttJdS: This, for example, is what Mariotti later played against Gheorghiu ( 1976), and the variation 12 .i.c 5 b6 1 3 i.xe7 tlJdxe7. 14 b4 axb4 1 5 axb4 1!i'd6 enabled him almost to equalise. If White is aiming for an immediate fight, he can play 8 a4 or 8 d4 exd4 9 . 4.lb5 . .. 8 9 �e7 a5 a3 11 12 13 14 In several games my opponents as Black carried out the plan of castling kingside and advancing . . . f7-f5, without preventing b2-b4 (Nos. 2 7 1 and 302). In Ulis case White gained a clear advan tage. 9 . as was once played against me (before this game) by Flohr, and our game quickly ended in a draw. Portisch, however, was wrong to imagine that I was in a peaceful mood. When I played Flohr, I was mainly concerned that my opponent did not lose interest in the . . tournament and would play as well as possible in his subsequent games. But on this occasion I was ready to try and exploit the defects of Black's last move, which weakens somewhat his queenside pawn formation. 10 11 .i..e3 ll:\a4 0-0 � White carries ou a typical man oeuvre in such positions, with the aim of 11xa4 llfcl llc2 ltlxa4 .i..d 5 lte8 · .tf8 Black again displays a certain com placency. It was also not possible to oppose White's aim of doubling rooks by the excessively active plan with a pawn sacrifice: 14 . b5 15 W'xb5 :bs 16 'i'a4 ( 1 6 .i.b3 1 7 'l'xc6), since it is not apparent how the white queen' s seem ingly dangerous position can be ex ploited. However, as was pointed out by Smyslov immediately after the game, Black could have successfuHy defended with 1 4 . . . .i.d6, and if 1 5 'i'bs f:De7. . . . . . . 15 :act lhb8 Black's desire to play 16 . . . c6, com pletely suppressing White's activity on the c-file, is understandable, but for the moment he sacrifices the c7 pawn and t11e consequences of this tactical oper ation were not calculated by him with sufficient accuracy. In the event of 1 5 . e4 (to which Black's preceding move seemed to have been directed) 16 dxe4 .i.xe4 17 l:Id2 1i'f6 18 .i.f4 :acs the immediate threats would have been parried, although White would have retained the initiative. 3 39 . . 16 lhc7 .t:.c6 be accepted: 1 8 <it>xfi 19 11fc4+ <ai>g6 20 9g4+ W 2 1 liJg5+, and Black has to part with his queen, otherwise he is mated. h6 18 Defending the g5 square, but weak ening the light squares still further. tics 19 :b7 . <ifi>h8 20 1Fc4+ Also after 20 . 1We6 2 1 tDxe5 Black would have been unable to resist for long. ... • • • .. Black continues his manoeuvre, ex pecting that White will have to give up rook for bishop ( 17 :7xc6 CD.xc6)� after which Black's slight material advantage will be compensated by White's positional tnunps. Alas, disillusionment awaits Portisch. . 17 lltxc6! Of course, Black also took this sacrifice into account, but he incorrectly thought that the rook at c7 would remain trapped and would also have to be given up. But the bishop at c6 is no longer there, and the b7 square remains undefended! However, even this is not the main thing. The rook sacrifice has eliminated the enemy bishop, which was covering the light squares in Black's position. 17 bxc6 Black retains his b7 pawn, but this is of no significance, since White is already aiming at the other flank. 18 1lxf7! Rarely almost in the opening is one able to give up both rooks, one after the other. True� the second sacrifice cannot • . . 21 tDb4! Once again White can ignore the threat to his rook - he is playing for mate. 21 11xb7 <t>h7 22 lDg6+ 23 .i.e4 With the threat of 24 /1Je7+ and 25 1t'g8 mate. 23 j\d6 24 li)xe5+ g6 Or 24.. . �h8 25 ltJt7+ <i>g8 26 liJxd6+. 25 .t�g6+ <i>g7 26 .*.xb6+! Black . resigns 340 . • • If 26 . . . <iifxh6 there would have followed 271 1ih4+ i-J;g7 28 'i'h7+ �6 (28 .. . <tif8 29 'i'xb7) 29 ltJg4+ @e6 (29 . . . �g5 30 1ffb5 mate) 30 'ii'xb7. As I have already mentioned, the Hungarian grandmaster \ised to suffer lapses, when he would lose even to weak masters. On this occasion the ' lucky recipient' was the author of these lines. Even so, I hope that the kind reader will not regard me as a weak player. 8 cxd5 exd5 9 b4 c 5 10 dxc5 bxc5 1 1 bxc5 .txc5 12 0-0 'i'e7 13 .tb2. 8 i.xc4 c5 Game 3 64 Botvinnik-Kholmov Now the exchange on c5 suggests itself, leading also to the exchange of queens, when for the coming endgame White is better mobilised. However, I had to decide: should I take on c5 immediately, or first castle? In the first case White would gain time, since his plan includes placing his king on e2 and his king's rook on d l . Fearing some counterplay by Black, associated with the fact that the white king remains in the centre, I preferred to castle, And I was wrong! Aloscow Team Championship, 1969 Nimzo-Indian Defence 1 2 l 4 d4 el li)f6 e6 ilb4 0-0 5 6 ..i.d3 al d5 JJ.e7 c4 lt)cJ 4 . . . c5 would have restricted White's possibilities somewhat. I would remind you that Tal played 6 . . . dxc4 7 .txc4 ..i.d6 against me (Games 285 and 286)� this was also played by Balashov (Game 346). 7 lLifJ Also possible is another favourable plan, which was employed by Petrosian in a game with Antoshin (1957) : 7 cxd5 exd5 8 b4 b6 9 ltJge2 c5 10 b5 a6 1 1 0-0 axb5 1 2 .txb5. 7 . • • 9 10 0-0 dxc5 li)c6 White could also have avoided the exchange of queens, by 10 i.d3 . The game Lipnitsky-Polyak ( 1 949) showed that he retains the advantage after 10 . . . cxd4 1 1 exd4 b6 1 2 b4 ..ib7 1 3 i.b2. But why take risks in a team event? dxc4 10 7 . . . b6 also comes into consideration. However, in the game Mecking-Donner ( 1 97 1) White reUtjnect · ·some advantage: 341 . • . 1ixd1 Black is obviously aiming for a draw. 11 12 %1xd1 b4 .i.xc5 ll.e7 13 .tb2 1 8 . . tt1e4 only seems more active, but after 19 .td3 the rooks are exchanged, and the knight has to abandon its position in the centre. .td7 . 19 .i.b5 White provokes the exchange of rooks and preserves his dark-square bishop from exchange (after 1 9 . . . .ilf6). 19 20 · i.xcl !ixcl+ lLlb6 White controls more space and has a lead in development, but can this be exploited? Up to a point, Kholmov dem onstrates that Black has quite adequate resources for maintaining the balance. 14 ltle4 Jlfd8 Black has no reason to fear 1 5 tiJxf6+ .i.xf6 16 .i.x:f6 gxf6, since all his difficulties would be behind him. It would have been simpler for him to play 14 . . . ltJxe4 1 5 l:.xd7 ctJd6! followed by 16 . . . :fd8, as my opponent pointed out after the game. 15 At last the king can be brought into play, and within ten moves it will occupy a dominating position in the centre of the board. 21 22 23 ltld6 . Only this continuation can promise White some advantage. 15 • . • a.c8 Of course, Kholmov prefers to retain his dark-square bishop, which will counter the activity of the bishop at b2. 16 17 'Dxc8 l1xd8+! 17 18 l:lcl ttJxd8 ltld7 a6 ltlc6 g6 Black has no reason to fear the exchange on c6, and the h7 pawn has to be covered. so that the king can move towards the centre. 24 25 26 .:axc8 White's plan is clear: to exchange all the rooks. after which it will be possible to activate his king. <it>fl .i.dJ .i.e4 �e2 �d3 .i.b2 � �e8 ltld5 The whole point is that Black could not continue his king march: if 26 . . . <&t>d7 there would have followed 27 .i.xc6+ and 28 t2Je5+, winning a pawn. Now 342 White has to exchange bishop for knight, then also the central pawns, but on the other hand a way is opened for his king to advance. .lxd5 e4 'it>xe4 'it>d5 27 28 29 30 exd5 dxe4+ �d7 For Black it would be most advan tageous to exchange the knights, and for White - the bishops. As for the ex change of the white bishop for the black knight, it too can be considered to favour White, but even so this is what Black should have decided on (32 . . . tLle7+). Strangely enoug� Black ex changes bishops, after which the proba bility of zugzwang is greatly increased. 33 .tc5 ttle7+ Now this check is possible (bad is 34 !JLxe7 �xe7, when the f2 pawn is en prise), but this is no longer of signifi cance, since the exchange of the white bishop for one of the enemy pieces is assured. 34 h5 30 Up to this point Black has defended well, but here he plays imprudently. He should have acted wit11out delay by playing 30 .i.d8 ! , whea without ex changing knights (3 1 ltJe5+), White would be unable to maintain his king at d5 (or e5) in view of the inevitable 3 l . . .ltJe7+. But in the bishop ending the chance of putting the opponent in zug zwang is minimal, and White would probably have been unable to convert his positional advantage. . .. 31 34 il.g7! il.f8 i.xc5 The reader already knows that it would have been better to agree to the exchange of the knight; 34 . . i.c7 should have been played. Now the e7 square is no longer ac cessible to the black knight: if 3 1 . .. .i.d6 there would follow 32 i.f6, and the white king remains on its central square. 31 32 'it>c4 After 34 �e5 i.xc5 3 5 bxc5 the situation would be more complicated (due to the weakness of the c5 pawn), and so White chooses a different way. .i.d8 .fi.. b6 343 . 35 'it>xc5 �c7 Defending the b6 square by 3 5 . . . tDc8 also had its drawbacks - the black knight would become too passive. 36 itlg5 36 37 38 itlh7 h4 Forcing the advance, and hence the weakening, of Black's kingside pawns. f6 f5 Zugzwang is approaching. In the event of 3 8 . . . b6+ 3 9 ct>d4 <it>d6 40 tDf8 etJc6+ 4 1 <;t>e3 tDe5 42 'it>f4 the g6 pawn cannot be defended. Kholmov seeks salvation in tactical complications. 38 39 . li_)f8 • . f4 b6+ The best practical chance was probably 39 . . . D 40 g3 tills 4 1 tDxg6 lDxg3. Although after 42 'it>d4 White also comes out a pawn up. the technical difficulties would be more significant. My opponent underestimated White's 42nd move. 40 <ifi>d4 ttJf5+ 41 <i>e4 tLlxh4 40 . f3 41 g3 ctJf5+ is now too late, if only because of 42 �d3 ! .. · 4 1 . f3 again does not help in view of 42 <it>xf3 l'Dxh4+ 43 lt>g3 li:Jf5+ ( 43 . . . g5 44 ltJe6+) 44 �f4 It would appear that Black has nevertheless succeeded in creating some counterplay, since it seems impossible to prevent him from creating a passed pawn on the queenside. White, however, has prepared the necessary antidote. 46 . lDe6+ <it>c6 Black's last hope is to attack White's queenside pawns. 43 ttJ:d'4 g3 tLlxg6 • • • ltlc4 'il?a4 <it>bJ 48 49 50 51 52 ltlxb6 'it>xa3 lLld5 <it>b3 f4 �c4 ltlc7 <ii>xb4 ltlxa6+ Black resigns <it>b5 When he played 39 . . . b6+ Black over looked that if 43 . . . g5 White replies 44 g3 ! gxf4 45 gxh4 with a won pawn ending. Therefore my opponent is forced to readjust. 44 45 46 47 The situation is not changed by 47 . . . b5 48 lDe5 �xa3 49 tDc6! or 47 . . . ct>bS 48 ttJb2. . . 42 li)e5! A manoeuvre, after which everything becomes clear. tLlf5 itlh6 344 Game 3 65 Ostojic-Botvinnik Beverwijk 1969 French Defence 1 2 e4 d4 e6 d5 3 lhd2 4 e5 5 lDf6 lDfd7 f4 At present this variation is not as popular as 5 i.d3, but even so it also sets Black difficult problems. 5 6 7 cJ lDdfJ c5 �c6 in a number of other games), 8 . . . h5 9 gxh5 :xh5 10 t'De2 tbb6 1 1 ltig3 lth8, did not appeal to me, since I thought that in certain circumstances Black should have the possibility of initiating immediate counterplay on the queenside by . . . b5-b4. 9 10 it)e2 .i.h3 lt:'lb6 First 10 fi)g3 should have been played, to prevent Black' s undennining advance . . . h7-h5. This is what hap pened later in the game Belyavsky Bagirov ( 1 974): 10 ltJg3 a5 1 1 Ag2 b4 12 0-0 a4 13 a3 bxc3 14 bxc3 lDa5 1 5 lta2 ctJb3 16 f5 h6 17 �hl. 10 11 12 gxh5 .i.g4 h5 !lxb5 llh8 Now there begins a battle for (and against) f4-f5 . 7 • • • Petrosian's idea. which he employed against Cooper in the 16th Olympiad ( 1 964). The point is that White's king's bishop will now be unable to reach the b l -h7 diagonal at an early stage of the game. However, the blocking of the queenside allows White to gain a men acing initiative on the kingside. However, does Black have anything better? If 7 . . . 'i'a5 the simplest is 8 �f2. 8 �g3 1£1e7 ffc2 g6 h4 After 1 5 <it>f2 lLlf5 White would have lost the opportunity of advancing his pawn to h4 without exchanging the knight at f5, but after the exchange (as occurred in the game) the pawn march h2-h4-h5 is no longer dangerous for Black. . • . b5 The continuati�n .that occurred in the afore-mentioned game of Petrosian (and 15 16 .i.xf5 16 ... 1£1f5 1 6 'i'h2 was no better on account of 16 . . . ltJxh4 17 t:Dxh4 i..e 7. g4! The most energetic and logical reply. lf 8 b3 Black can reply 8 . . . b5, and the variation 9 a4 ttJas 1 O ltid2 b4 is quite safe for him. 8 13 14 15 c4 gxf5! Now Black's position on the kingside is secure. He has sufficient space for manoeuvring, while the white h-pawn is not only blocking the . enemy forces, but is also subsequently a target. The 345 picture would have been quite different after 16 . . . exf5 17 h5, when White's advantage is obvious. 17 25 l:tagl l:tgh8 Parrying the threat of 26 l2Je4. 26 27 28 29 l1h3 l:tghl l::t l h2 �g2 �c7 J:l8h7 'ilh8 ltlc8 tl'lg5 The knight 's position may look menacing, but all its 'shots' are blanks. 17 18 19 20 exf6 'ife2 h5 f6 1f xf6 .i.d6 llh6! The knight .embarks on a lengthy journey; its ultimate destination is e4. The h-pawn must be blockaded, since it is restricting three of its own pieces: queen, rook, and knight at g3 . 21 tl'lf3 White redirects his knight to another square: where. however, it will be just as harmlessly placed. Perhaps he should have tried to maintain the status quo. 21 22 ... ll'le5 30 31 32 ..i.e3 � ltle7 .i.d7 White has returned his laright to its former position, where it at least prevents the opponent from tripling on the h-file. .i.d7 0-0-0 Here the king feels very much at home. This is one of the subtle points of Petrosian' s generally unsuccessful idea. So, in this game too I castled late, but I cannot especially boast about this, as my opponent did not castle at all. 23 24 i.d2 ltlfJ tl'lg5 :lg8 il.e8 346 32 33 34 35 lDfl <ifiih l l:th4 l:tg7 lDgS l:Ie7 ll'lf6 From this point until the time control White does not do anything, since he has no such opportunity, and Black waits for the end of time trouble. 36 lt2h3 37 lDgJ 38 . �g2 <it>c8 1ff8 .i.c6 3 9 ltlxe6 1s not dangerous for Black in view of 39 . . . 'i'g8. hope of further strengthening his position on the kingside. His reckoning proves to be correct, since White is condemned to passivity� with each move Black's advantage increases. 46 lt)fl il..e7 'i'g8 47 :!4h3 48 lig3 a5 39 �1 It is not a bad thing also to 'frighten' the opponent with the prospect of a breakthrough on the queenside. 1f.e8 40 l:[b2 lhe4 41 1fe1 The invasion of the knight in the centre has at last taken place. It is not easy for White to tolerate the enemy piece here, although had he done so he would probably have been able to prolong the resistance. But after the following exchange the active white knight is eliminated, while Black acquires a passed pawn and his queen's bishop gains in scope. 42 l£15xe4 After 42 t2J3xe4 the h5 pawn would have been in danger, but now White's cover on the g-file disappears. It should also be remembered that Black always has active play on the queenside in reseive. dxe4 42 43 ife2 .i.d5 lib7 44 a3 i:'t g7 45 <st>f2 Black decides to defer . . . b5-b4 in the White allows the invasion of the black bishop at h4, after which the out come is decided. But, on the other hand, what could he do? After all, the un pleasant invasion of the rook at g4 was threatened. 48 :xg3 49 ltlxg3 1fd8 To defend the h4 square a second time, White would have had to make two moves in a row: 'i'e2-fl-h3 . . . so· 1ifl .th4 51 'i'b3 lixh5 The further resistance does credit only to the stubbornness of my opponent. 52 �e2 ilh8 53 lLlxf5 A desperation saCrifice. 347 • • • 53 . 54 • • • .i.el was carried out two years earlier by Basman in a game against me in Hastings, and folloWing his example this method of play for Black became firmly established in tournament play. exf5 The immediate 54 °iixf5+ 'ifd7 was even wors-e. 54 55 56 57 1lxf5+ l:lxh8 .i.h4 il.f6 <J;c7 .i.xh8 1fd7 7 The simplest way to win. 58 59 1if8 <i;d2 ltld5 Since 7 d3 .i.xc3 8 bxc3 e4 leads to a position from my game with Basman, in which I did not gain any tangible advantage, and my opponent, naturally, was hoping for this continuation, I decided to play differently. 1ig4+ e3+ As will be apparent later, this is a useful pawn sacrifice. 11xh4 60 <i>xeJ 61 'i'c5+ White resigns, without waiting for the obvious reply 61 . . . Jl.c6. Then if 62 d5 there would have followed 62 . . . 'i'g3+, 63 . . . 'i'd3+ and 64. . .'i'xd5 (this is why the enemy king had to be enticed to e3 ), while after 62 ifa7+ <t>c8 63 'i'a6+ �d7 64 'i'a7+ �e6 there are no more useful checks. 7 Game 366 Botvinnik-Ree Beverwijk 1969 English Opening 1 2 3 c4 ttJcJ g3 e5 ttlf6 .i.b4 4 5 6 .i.g2 ttJfJ 0-0 0-0 ite8 ttJc6 This variation, a Vladimir Simagin, is strongest one for Black. favourite of probably the . . • .i.c5 Inconsistent. As will be seen, Black had decided to exchange on d5, and he should have done this immediately (7 . . . ti:Jxd5 8 cxd5 �d4), when White would not have had a convenient retreat at d2 for his knight from f3 . Another interesting reply was 7 . . . .i.f8, which was employed much earlier (Portisch-Filip, 1 962). After 8 d3 h6 Black had a satisfactory position. 8 d3 Now Black goes in for the exchange in a less favourable situation. 8 9 The most advisable here is 6 . . . e4 7 �d4 �c6 or first 7 ... .i.xc3 . This idea 348 cxd5 t2Jxd5 tiJd4 10 lhd2 White's pieces are temporarily com pressed, like a spring, but soon it will uncoil ! 10 11 12 e3 lhc4 13 i.d2 • • • 17 18 19 'lrc2 ltcl 20 d4 f5 l:e7 1.b6 d6 lhf5 �d7 Perhaps Benko played more subtly against Gheorghiu ( 1 972): 12 . . .tDe7, although after 1 3 .i.d2 c6 · 14 b4 .i.b6 1 5 dxc6 bxc6 16 b 5 he failed t o equalise. A move which I deeided on only Now the threat of 14 b4 i.b6 15 a4 is unpleasant. 13 14 15 • . • b4 · a6 ll.a7 .fl.c8 lDa5 A difficult decision, but Black was afraid of playing his rook to b8, rightly assuming that this might hinder the defence of his backward c7 pawn. 16 l:lcl 17 .:cJ . i0h6 Black is hoping to create counterplay on the kingside. He has no other active plan. This direct pres�ure on the c7 pawn suggests itself. · after lengthy hesitation. 20 . . . exd4 2 1 exd4 i s unfavourable for Black (2 1 . . . .i.xd4 22 :xc7), since the opening of the game is to the advantage of the better mobilised side. Of course, I also considered 20. . . e4, shutting out of play the bishop at g2. In this case, however, Black would have been deprived of the chances associated with . . . f5-f4, and hence his bishops too would have remained passive. In addition, . after 20 . . . e4 White could have immediately intensified the pressure on the c7 pawn by 2 1 a4 ! , with the threat of 22 l0c4 i.a7 23 a5. 20 • • • �g4 This is the most hannless reply, since White exchanges his doubled pawn for the good pawn at e5, without which the f-pawn cannot advance, and he will be free to intensify the pressure on the weak c7 pawn. 349 21 dxe5 �xe5 In the event of 2 1 . .. dxe5 Black would have lost control of the c5 square, and also the laright at e5 is needed to pre vent the manoeuvre mentioned earlier: a2-a4 and ltlc4. 22 �c4 It is useful for White to exchange knights: then the other side's lack of development will become even more marked. 22 23 · • • • %bc4 �xc4 .i.d7 At last this bishop has gained the opportllllity to remind one of its existence� but how will Black be able to defend the c7 pawn? 24 25 a4 bxa5 a5 .t.c5 After 25 J..xa5 26 .i.xa5 l:txa5 27 :xc7 .?lxa4 (27 . . . .lxa4 28 %1c8) 28 :xb7 White is a pawn up with the better position. this the dark-square bishop becomes very strong, Black's position proves to be indefensible. 26 27 • • • tfxc5 dxc5 c6 Of course. not 27 . . . i.xa4 because of . 28 'i'c4 with the tltreats of 2 9 1!fxa4 and 29 d6+. 28 29 30 31 l:tbl .lb4 l:tdl d6 .i.e8 ltti l:lc8 Now all the black pieces, apart from the queen, are condemned to passivity, and on its own the queen cannot do anything. 31 32 33 34 . . . hJ <ifi>h2 l:ld2 .i.d7 h6 'ffg5 lle8 White spends a further three moves destroying Black· s last hopes of counterplay. 35 36 37 f4 .i.d .i.e5 Wf6 9e6 9b3 Now it appears that the worst for Black is over, but . . . 26 %lxc5! White's extra pawn allows him to sacrifice the exchange, eliminating the opponent's only active piece. Since after a 3 50 Now White is again able to exchange devalued doubled pawn for an since after then playing . . . e6-e5 Black can develop his queen' s bishop at g4. 9 b4! An idea which, in reply t.o .8 . . .ltlc6, I carried out in the 3rd game of the return match with Tai (No.286). White uses the tempo· saved by not castling, .in order to activate his queen's bishop as quickly as possible. Other continuations: 9 e4 e5 (Gligoric-Pachman, 1961) and 9 1Wc2 e5 (Van Scheltinga-Euwe, 1948), would seem to give Black an equal game. e5 9 10 .i.b� important enemy pawn, obtaining in the process an outsi9e passed pawn on the a-file. 38 l:tb2 1fxa4 39 lbb7 :as ·11c2 40 Ji.cl The time control was reached, and there was no doubt that White would seal the obvious 4 1 ifc4 with the threat of 42 llxd7. Then the only 'defence' would be 4 1 ...l:td8 (4 1 . . .� 42 .i.xg7+), allowing the a-pawn to advance. Therefore . . . Black resigns. Game 367 Botvinnik-Van Scheltin ga Beverwijk 1 969 Nimzo-Indian Defence 1 d4 tLlf6 e6 2 c4 3 lDcJ .i.b4 0--0 4 e3 Tastes differ, it is true, but I, as a rule, preferred to castle later with Black in the Nimzo-Indian Defence. d5 5 .i.dJ 6 a3 dxc4 7 .*.xc4 7 i.xh7+ is another possibility, but, although it has been employed several times, in this case Black can maintain the balance. 7 .i.d6 • 8 • • tDf3 • • • • . 8 f4 does not promise White any advantage (cf. Game 346). 8 t0bd7 The usual reply} · · ·lhc6 is stronger, • • 10 e4? After this Black begins to have difficulties. When White has not yet castled kingside, this advance must be approached with great caution. A simi lar situation arises in the French Defence when White prematurely plays e4-e5. However, in this situation 10 . . exd4 1 1 Wxd4 is also favourable for White - this is where the absence of the knight from c6 tells. 10 . . a5 1 1 b5 exd4 12 ti)xd4 l'Db6 1 3 .i.e2 also cannot be recommended for Bl�ck. Therefore he 351 . should have chosen a cautious move (10 ... a6, or 10 ... c6, or even 10. . . 1ie7), retaining a defensible position. 11 12 thd2 1'.e2 · Also bad is 14 . . . ti:Jxc3 15 gxf6 li)xdl 16 fxe7 ll'lxb2 17 exf81i+, when White has a material advantage sufficient for a win. thb6 iie7 15 fxe3 Of course, not 15 gxf6 'ifxf6 16 In this way Black appears to secure his central pawn. But since White has not yet castle� in the battle against the e4 pawn he can make use of an unexpected resource. �dxe4 ( 1 6 fxe3 9h4+ 1 7 <ifi>fl .i.h3+ 18 'ifi>gl 1!i'g5+) 16 ... 1i'h4 with some complications. Now, however, White is simply a piece up. 15 • . • li)d5 Since 1 5 . . . tiJe8 16 ctJdxe4 leaves Black with no hopes of saving the game, he tries to pour fuel on the fire. 16 17 li)xd5 Q:)xe4 1i'xg5 In order to win one game, one extra piece is sufficient. 13 g4! In view of the threat of 14 g5 (and if 13 ...h6 14 h4 and 15 gS), Black cannot defend his central outpost. It may be recalled that in a slightly different situation (to eliminate the defence of the h7 pawn) I canied out such a pawn storm back in 1934 in a game with Alatortsev (No.59). At that time this move was virtually a revelation, where as now it has almost become a standard procedure . . 13 ... 14 g5 17 18 19 20 t[}xd6 0-0 20 21 22 23 :n 1fd3 e4 .i.fJ 1!fxd5 �h8 1fxd6 It is only now that White needs to castle. thbd5 My opponent makes a desperate attempt to confuse the play, but it is not hard to find a way through this 'confusion' . . thxeJ 352 Jl.h3 l:tae8 .b6 lle6 This is the most obvious way of making it hard for White to develop his kingside play. Now the advance of the d-pawn is impossible, and that of the g pawn is more difficult. This method was well known earlier, but usually Black first developed his king's knight at f6, then (after castling) moved it, and only after this played . . . f7-f5. By making this advance immediately, he gains two tempi. · The early advance of the f-pawn is an idea that I also carried out in earlier times, both with White (in the English Opening), and with Black in the Closed Variation of the Sicilian Defence (for example, in the 1 3th game of my 19 54 match with Smyslov, No.223). Black has a pawn for the piece, and the moment there is no immediate danger, and . so he considers himself obliged to play on. for 24 25 26 27 28 d5 <ilbt 1id4 :gt <ifi>xg1 ii g6+ 1fh4 f6 l:.xg1+ l:le8 29 30 31 32 �ht lle2 .i.xg4 b5 .i.g4 'i'xg4 Even if Black were allowed to switch his rook directly via e5 to g5, it would be an ' attack' without any threats. 'tle3 Finally White creates his first threat: n i.xf6 (33 . . . gxf6 34 'i'h6+ 'it>gs 3 5 ng2)� therefore, most probably, Black resigned. Grune 368 Medina-Botvinnik Beverwijk 1969 Sicilian Defence 1 2 3 e4 lBc3 gJ c5 lBc6 Theory considers that only by the fianchetto of his king's bishop can White hope for an opening advantage. 3 4 5 6 i.g2 d3 f4 g6 i.g7 d6 6 .i.e3 is also played, but without particular success. 6 • • . 7 tl)f3 7 8 9 0-0 'i>ht 7 exf5 .ltxf5 can hardly be dangerous for Black, and, as shown , by the game Tarve-Pohla (Parnu 1971), 7 . . .gxf5 is also possible: 8 'i'h.5+ <ifi>f8 9 'i'dl ttJf6 10 liJf3 <:J;f7 1 1 t2Jg5+ 'it>g6 12 h4 h5, with a promising position for Black. f5 353 ltlf6 0-0 'ii?b8 If White has vacated the g 1 square for his queen's bishop, why should Black not follow such a good example? In the second edition of the Ency clopaedia (in contrast to the first), this plan is also recommended. As for 9 . . . J..d7 10 Ji.e3 ztb8 1 1 'i'e2 b5 12 J..g l b4 1 3 ti:Jdl ti:Je8 14 c3, which occurred in the game Smyslov-Larsen ( 1958), then White's chances would seem to be preferable. 10 il.eJ .i.e6 1 1 1i'd2 For some reason the Encyclopaedia ignores this natural continuation and recommends playing 1 1 J.gl 'iid7 12 exf5, reckoning that after 12 ... .i.xf5 13 d4 White has the advantage. Tiris may be so. But if Black replies 12 . . . gxf5, then 13 d4 no longer leads to an advan tage for White, as stated in Informator No.25 (Rajcevic-Martinovic), since with 13 . . . .il.c4 1 4 l:te 1 cxd4 15 lDxd4 :ae8 Black achieves a good game. ifd7 11 :ae8 12 .i.gl b6 13 ltael This advance is motivated by the need to defend the c5 pawn in the event of a possible e4-e5 . 14 b3 But this similar reply was not induced by anything. and is at best a waste of time. 14 .. �g8 15 tLlh4 White is aiming for an attack, which, however, is not easy to build up. The direct 15 e5 dxe5 1 6 lDxe5 ltJxe5 1 7 f.xe5 would have been met by 17 . . . lDg4 (18 'i'f4 .i.h6). • ltlg4 15 A good position for the knight, since h2-h3 would weaken the white king' s position. Now B1ack gains the oppor tunity to begin active play, by advan cing his central pawns. 16 tLld5 In order not to remain out of play, the knight aims to exchange itself for its active opponent. 16 e6 17 ltleJ tl)xe3 d5 18 .i.xe3 19 e5 Now the play becomes closed, and since on the kingside Black has a solid position, White subsequently cannot do anything active. And at the same time the bishops at g7 and g8 only seem passive, but nothing will be able to prevent them from switching to good positions. Therefore White would have done better to maintain the tension. in the centre. 19 d4 In this way the weakness of the c3 square is fixed. 20 Af2 3 54 . . . 28 b5 White is relying on the closed nature of the position, but in vain. It is now that Black gains complete freedom for manoeuvring. Therefore 28 bxc5 really would have been better. 28 29 a4 lhe7 a6 Since the h 1 -a8 diagonal can always be blocked by the laright at d5, this last move by Black, emphasising the weak ness of the white a- and b-pawns, is perfectly possible. 20 30 31 32 g5! For the present this is only a prophyl uct ic measure against possible activity by White on the kingside; by advancing his pawn to g4, Black will guarantee hi n iself against any unpleasantness. However, this seizure of space also n l lows Black at the necessary moment lo himself begin energetic play on this pm1 of the board. 21 c4 lhxc3 axb5 dxc3 axb5 ltld5 Of course, Black declines the gift (32 . . . 'i'xd3) in view of 33 'i'b3 with the threats of 34 :t'dl and 34 'i'xe6. 33 34 lhxd5 ltbt exd5 : as lhf3 After 2 1 fxg5 ttJxe5 the centre of the hoa rd passes into the opponent's possession. 21 22 lhgl g4 ltc8 The switching of the forces to the q11censide begins. 23 24 25 lhe2 Ilcl ifdl ltfd8 J.h6 The threat of 25 . . . ltJxe5 has to be parried. 25 26 27 a3 b4 .i.f7 .i.e8 The only active possibility. 27 • • • J..f8 What can be said about this position? The passed e5 pawn is of no signifi cance, since it will be blocked by the light-square bishop. And at the same time the black c- and d-pawns are shut ting both of the white bishops out of the game. 355 35 d4 Now Black obtains a passed pawn, but White is hoping to bring his queen' s bishop into play vi a b4. 35 36 • • • 11e2 47 ifxb4 48 49 50 l:lcl 11el �cbl Now Black need no longer fear h2-h4 - the g3 pawn is en prise. c4 :sa4 :aS It was more logical to play 3 6 .i.el immediately. 36 37 • • • .i.el .tf7 .fla4 Exploiting the opportunity offered, for the moment Black prevents further simplification. 38 39 40 41 .icJ l:lfcl lk2 �gl .i.e6 l:t.da8 .i.e7 �g7 Both players realise that the action is bound to spread to the h-file, and so it is 50 useful for the kings to leave it in good time. 42 43 43 • • • 1i'e7 Here the queen is most active. h4 1id2 .i.fl h5 � 51 h4 White's position is becoming hope gxh3 44 �xh3 would have less. His pieces are tied down, and are inevitably led to an attack by Black on unable simultaneously to defend both the kingside. White continues to set his the queenside and the kingside. hopes on avoiding the opening of the position. . 43 44 • • • � g6 .:aJ to exchange the • in view 55 hxg3 of the threat of !!h2+ 56 �gl second rank. 54 55 56 dark-square inunediately makes itself felt. • llh8 I:txg3+� but in so doing he concedes the c4 pawn arl.d the third rank, which • ... l:tcJ 54 . . . hxg3+ bishops, but in so doing he unblocks the 46 53 54 necessary White straight away takes the oppor tunity :as White defends this rank, which is h4 (45 . . .gxh3 46 .i.xh5+). Itcb2 .ib4 11cl :c2 Trying to block the third rank. �f7 ll.e2 Creating the positional threat of 45 44 45 46 52 53 .i.xb4 hxg3 l:tb2 hxg3+ l:a2 �h2+ Here the game was adjourned for the second time. but it was not resumed. If 3 56 usually endeavours as · soon as possible to initiate an exchang� of fire with 7 . . l'Llg4. "i7 cJi?e3 the .s.imple 57 . . . 'i'a7 is possible, nnd soon BI�ck wins the b5 pawn. White resigns. . Game 369 Botvinnik-Lombardy Beverwijk 1969 Sicilian Defence 1 2 3 c4 lbf3 e4 c5 g6 d4 lbxd4 .i.eJ .ie2 d6 . :i; 9 0-0 .i.d7 Here 9 . . . t'Dg4 no longer works because of 10 .i.xg4 i.xd4 ( 10 . . . .ixg4 1 1 tLlxc6) 1 1 .txc8. This move signifies that White is rejecting the English Opening in favour o f the Sicilian Defence. It is tempting 1 hat this leads to the Maroczy Variation ( with c2-c4), which, in my view, sets Black a number of difficult problems. 3 4 5 6 7 8 Here mention can be made of a dubious experiment by Bronstein in a game with Aronin (195 1): 8 . . .lLle8 9 'i'd2 f5 10 exf5 gxf5 1 1 f4 tiJc7 12 0-0-0 d6 1 3 .i.f3 with a clear advantage to White. 10 'ild2 In recent times preference has more often been given to 10 lt.c 1 . 10 • • • lhxd4 Experience has shown that it is hard for Black to manage without this exchange, although he has often first played . . . a7-a6 and . . . l:c8. At the given moment the exchange allows the queen to be brought out to a5, since White's queen is unprotected and he does not have the reply ttxl5. Jlg7 cxd4 lbc6 lDf6 lbc3 11 .ixd4 1la5 Another method of play consists in playing the bishop to c6 followed by . . . a7-a5 and . . . l'Lld7. 12 7 . • • 0-0 1* It is hard to call this move a mis e, but nevertheless, in view of White's obvious dominance in.· the centre, Black - ltfdl lifc8 It is well known that if in similar positions Black succeeds in advancing . . . b7-b5 (after the preparatory . . . a7-a6), he obtains a good game. Bearing in mind my tournament position, I decided to risk a possible draw in a slightly better endgame, but on the other hand I clearly disrupted my opponent's plans. The following exchanges deprive Black of any counterplay, which was 357 certainly not to the liking of my young opponent. 20 21 22 23 Itel .i.f1 ctfl i.e8 iJ..f7 a6 lie.3! 23 . . . b5 has not yet been prepared, since there follows 24 M, and White uses a standard procedure for trans ferring his rook for an attack on the h5 pawn. 23 This loss of another tempo merely indicates that Black does not appreciate the dangers of his position. 13 14 .i.xf6 thd5 24 25 26 .txf6 ifxd2 lhxf6+ �g7! After 15 . . . exf6 16 l:txd.2 l:!c6 17 l:tadl Black would also have lost his d6 pawn, but now he has a weak pawn at h5. 16 17 lhh5+ llxd2 llac5 l18c7 White has deployed his pieces in the best way possible, and now he begins the 'pursuit' of the h5 pawn - the concluding stage of the plan, begun after the opening with 14 llXI5. Otherwise Black loses his d6 pawn. 15 a4 .i.e2 llg3+ gxh5 f6 For the moment the h5 pawn is im mune: White's bishop has to defend the c4 pawn. Therefore Black first of all parries the threat of 1 8 e5, and 1 8 :d5 is·not dangerous: he can reply 18 . . . .i.e6. 18 19 b3 f4 lies ltac8 Despite the transition into the end game. Black still has only tl1e one active plan� consisting in the preparation of . . . b7-b5 (for example. 1 9 . . . .i.e8 and 20 . . . l:.b8). Of course, this would not rid him of all his difficulties: the h5 pawn would remain :weak, and on the queen side one pawn , would have to oppose two enemy pawns. However, marking time would be even worse. <t>f8 26 After 26 . . . 'it>h6 27 l;t.b3 White would have then played f4-f5 with the threat of �i.xh5 and g2-g4. And if Black replies to the advance of the f-pawn with . . . 'it>g5, then l!.dd3 underlin�s the danger of the black king' s 'thoughtless behaviour' . 358 27 . . :lh3 Winning a second pawn (d6 or h7). :1c6 33 34 35 36 ltxd6 ltdJ :g4 .i.g8 <be7 :e5 J..ti 37 38 l::tg7 l:h7 h6 ltb6 The first and last attempt to create counterplay: Black is ready to give up t he exchange (2� b4 llxc4 29 i.xc4 J1xc4), after which his position is a I right. But White continues attacking 1 hc h5 pawn. The invasion of the seventh rank by the rook is unavoidable. 30 . . .txh5 3 1 l:xh5 b5 was bad for I Hack on account of 32 l:txh7 bxc4 3 3 Iih8+ ! �g7 34 l:tclh3, when mate is inevitable. In .the vain hope of continuing the resistance after 3 9 :xh6 :xe4. White, however. can conclude matters with an attack! 39 .i.g4 <t>f8 28 29 30 f5 lldd3 i.. xh5 l:tb6 l:tb4 . J..g8! . 40 lid8+ i.e8 41 lah8+ � Black resigned, without waiting for the obvious finish: 42 .:!dxe8 (or 42 l:thxe8) 42 . . . :xe8 43 �h5+. The plan devised after the opening proved justified: White was able to win without any risk. Game 370 Ciric-Botvinnik Beverwijk 1969 But how can White now parry the threat of 3 1 . . . b5, when Black's kingside appears to be securely defended? 31 Pirc-Ufimtsev Defence This imperceptible move solves the problem. Now if 3 1 . . .b5 there follows 3 2 l:tg3 bxc4 3 3 l:thg4 ilf7 3 4 .i.xfl �xfl 35 �g8 and 36 kt3g7 mate. There fore Black is forced to forget about counterplay' and to prepare an escape for his king. 31 32 33 1 2 3 4 5 Jih4! fxe6 J.. d l e6 i.xe6 e4 d4 ltlfJ ltlc3 ile2 g6 !i.. g7 d6 ltlf6 Regarding the development of the light-square bishop at c4, see Games 353, 374 and 378. 5 .. 0-0 . 6 0-0 .i.g4 The most natural development plan, although here other continuations have 359 also been employed: 6 . . . lDfd7; 6 . . . c5; 6 . . . a6; 6 . . . e5; 6 . . . �c6 etc. 7 · · .i.eJ 7 .i.g5 is · perhaps stronger. That is what Matulovic played against me in the 3rd round of the 'Match of the Century' (Belgrade 1 970). 7 �c6 10 11 .i.e2 11 12 dxe5 . • ltlb6 • 1 1 b3 was simpler. e5 Aiming for simplification. 12 • • dxe5 . Black has no reason to avoid the ex change of queens, since he will have the hope of establishing his knight on d4. 13 1ixd8 On this exchange White loses a further tempo, but he was afraid that after the more natural 1 3 :ad 1 there would follow 1 3 . . We7, retaining the queens. . 13 • . . l:laxd8 As will be seen from what follows, it would perhaps have been better to take on d8 with the king' s rook. 8 h3 8 9 .i.xf3 Now Black effectively gains a tempo and must be able to achieve an equal game. 8 'i'd2 is more consistent, 8 d5 is also often played, but my opponent was apparently thinking only of a draw, and so he was in a hurry to force simplification. iLxfJ tiJd7 Had Black too been aiming for a peace agreement, he could have played 9 . . . e5, and after 10 dxe5 (of course, 1 0 d 5 i s more promising) 1 0 . . . dxe5 1 1 ltJe2 the chances are equal. Black, however, prepares . . . e7-e5, so as to be able to answer d4-d5 with . . . ctJd4. 10 · 1fd2 10 �e2· circumspect. e5 11 c3 was 14 ii.cs This is the whole point. By forcing the rook to occupy e8, White hopes to continue advantageously with ..tb5, so as not to let the enemy knight in at d4. more 360 14 . 15 .i.b5 16 l!adt - ·: 1 :res lle6 'drawing' bishops is now inevitable, this is to White's advantage, and Black loses his winning chances. The manoeuvre of both bishops to 1 he fifth rank should logically have been completed by the exchanges 16 .1'..xb6 axb6 1 7 .i.xc6 �xc6. lhd4 a6 lDa4 16 17 18 tDe2 ..id3 19 20 21 .i.xd4 bJ lhg3 22 J..c4? 22 23 ... f3 :.e7 h4 lDhl ile2 l:tfel? b5 .i.b6 · Practically forcing the ·exchange of l he bishop for the knight at- d4, since if I 9 i.a3 there could have followed 1 9 . . i.f8. Now opposite-colour bishops remain on the board, but this does not prevent Black from increasing Iris advantage. . exd4 lhc5 h5! In order to defend his e4 pawn, White will be forced to play f2-f3 , which leads to a further weakening of his position. 26 27 <ltt'l 32 33 34 'it>gl lDd3 lhgJ+ c5 l:hel + 35 36 37 lbel bxc4 lhe5 c4 bxc4 :a7 axb5 :dd7 fxe4 tDxe4 In the event of 34 . . . ctJe2+ 35 �fl tLic3 White would have been satisfied with the simplest solution: 36 .D.xe7 llxe7 37 :el . Why drive away the knight which was already out of play! It would have been correct to defend the rook by 23 . . . :ld6, and then perhaps play . . .b7-b5 and . . . .:t.c6. A clever trap, which Black failed to see through in time. i.xb5 lbeJ :teel fxe4 lLlf2 The possibility of this move, forcing further exchanges, increases the likeli hood of a draw. A useful move, since in some cases Black was already threatening to ex change on d3 and then play . . J :k6. 24 25 26 28 29 30 31 32 The decision of a practical player: · that in time· trouble I should in the first .t.eJ+ f5 Here too it was essential to play 27 . . . lid6. Although the exchange of the 36 1 instance eliminate the opponent's out side passed pawn. 38 39 40 l2Jxc4 l2Je5 litdl l:.xa2 :xc2 ltJe2 3 On the last move before the time control if was hard to refrain from · giving this check, which, besides, cannot spoil anything. 41 42 43 <it>ft 'lt>gl ltlxg6 3 4 lLlgJ+ ltlf5 lLlxh4 Even without this combination a draw would have resulted, but this way the game reaches a forced conclusion. lLlxh4 44 lLlxg2 45 l:txd4 46 lle4 Draw agreed. If Black had been wanting to 'torment' his opponent in the ending with rook and knight against rook, then after 46 . . . �6 47 <it>fl <ifrf5 48 :e2 he would have had to give up such an idea. Game 3 7 1 Langeweg-Botvinnik Beverwijk 1969 English Opening 1 2 c4 g3 lLlf6 e5 c6 ltlf3 Now a well-known variatioll is reached where, instead of ltlc3 , White has played Ag2. The fact that in this situation Black easily gains equality was demonstrated many times by Keres. 4 d4 was essential, to answer 4 . . . e4 with 5 ltlc3 and then .i.g5. However, the Encyclopaedia considers that Black's position is also not worse after 4 d4 exd4 5 'ii'xd4 d5 . <it>f7 4 5 6 44 .i.g2 White wrongly avoids the immediate development of his queen's loright. lLld4 cxd5 e4 d5 6 d3 looks more natmal, although in the game Benko-Tai ( 1 962) after 6 . . . Ji.c5 7 lDb3 ii.b4+ 8 i.d2 .i.xd2+ 9 'i'xd2 dxc4 10 dxc4 'ile7 White did not achieve anything significant. 6 ltlc3 cannot be recommended, in view of the simple 6 . . . dxc4. After the continuation chosen in the game, Black obtains fine piece play. 6 7 • • • tixd5 ltlc2 7 lLlb3 is no better, for example 7 . . . 'i'h5 8 h3 ifg6 9 ltJc3 ltJbd7 IO 1fc2 e3 (Larsen-Korchnoi, 1 973 ). 7 8 • • • 1fh5 h3 The enemy bishop cannot be allowed to go to h3 ! Say, 8 tt:)c3 i.h3, and if 9 .i.xe4? lDxe4 I 0 ltJxe4 Ji.g2. 8 'ilg6 All the same, this retreat would have . • • had to be made after tbc3 . In addition, this hinders White's d2-d3 . True, even 362 in the variation 8 i.d6 9 d3 exd3 10 'tixd3 i.c7 the position is roughly equal. .. . 15 9 b3 The correct idea. By exchanging the dark-square bishops, White makes the position of his king more secure; in the variation from the previous note, from c7 the bishop could have been used in an attack. 9 t'Llc3 was also quite good, of course� in this case the play would have been more complicated. 9 �c5 d3 In this clever way White achieves his aim� but at a high price - the resulting ending is clearly unfavourable for him. 15 16 lDe5 17 1fxd2 18 lDxg6 19 ibf4 exd3 d2+ :xd2 liad8 ibb4 The forcing variation continues. 20 21 22 a3 ibxc2 0-0 lDc2+ llxc2 For the moment Black prevents his opponent from castling: 10 0-0 .i.xh3 (or 1 0 . . . 'i'xg3) 1 1 �xh3 'i'xg3+ 1 2 Ji.g2 '1Jg4. As for the exchange of bishops, he has no reason to obj ect to it. 10 11 12 13 lDbxa3 lDc4 ib2e3 13 14 1fc1 i.aJ .i.xa3 0-0 :d8 It is still dangerous for White to castle, and he prepares to eliminate enemy No. I - the .central e4 pawn, which restricts the activity of his bishop. lDa6 .*-e6 363 White had this position in mind, of course, when he played 14 We i , which does credit to the resourcefulness and foresight of my young opponent. More over, he had to foresee that in this position after 22 . . . .i.xb3 23 llfb l :c3 24 a4 White's chances are not worse. 22 • . . �dd2 This looks very powerful, but 22 . @£8, inunediately bringing up the king, came into consideration. .. 23 24 li)xe6 l:tadt fxe6 In order to answer 24 . . . .U.xe2 with 25 l:ld8+ and 26 l::tb 8. 24 � • 25 • • @e7 e3 Now it becomes clear that for the endgame Black's king is much better placed than White's. · 26 27 28 llxd2 %:tel ktxd2 �a2 llc4 28 a4 .ttb2 29 l::tc 3 would have led to an even more passive position for White. 28 29 • • • l:r.b4 l:lxa3 b5 The b-pawn is more valuable than the c-pawn. 30 31 .i.xc6 a6 .i.b7 'iiid6 It becomes clear that all White's pieces (king, rook and bishop) are badly placed. 32 g4 li)d5 Too hasty. The transition into the rook ending increases White 's chances of a successful resistance. 32 . . . h6 was simpler, when Black maintains all "the advantages of his position and the dangerous tlueat of 3 3 . . . �c5 . 33 34 ll.xd5 :r4 exd5 For the first time in the game, White obtains real counterplay. Since he is intending to play 35 b4, Black must accept the temporary pawn sacrifice. 34 35 36 llxbJ It.ti %bg7 a5 Now after the straightforward 36 . . . a4 37 :xh7 a3 3 8 l:l.a7 b4 39 g5 the g-pawn becomes menacing. In the time scramble I was able to find a clever possibility, although, of course, I would not be telling the truth if I were to assert that all the consequences had been studied. 36 d4 In this way Black eliminates any danger. 37 llxh7 My opponent no longer had the time to work out the complicated situation, and he makes a plausible, but erroneous move. Now the black d-pawn promotes. White should have accepted the challenge: 3 7 exd4 l::t.xh3 3 8 �g2 :h6 3 64 there was no reason to fear any pleasant surprises. 39 :a7, after which Black' s chances of success are highly problematic. Even so, he could have tried 39 . . . �d5 40 l:xa5 !tb6. 4 5 d5 e4 un d6 l£if6 By a transposition of moves, this position was also reached in my game with the Yugoslav grandmaster. 6 7 8 9 i.d3 h3 tf)fJ exd5 0-0 e6 exd5 9 cxd5 was unfavourable because of 9 . . . b5 ! ( 10 .ilxb5 ltJxe4 1 1 lDxe4 'ii'a 5+). 9 10 Jie3 l:e8+ .i.h6 I should remind you that Matulovic did not risk this continuation, played 10 . . . tDh5, and ended up in a passive position. But Kavalek had plenty of time for preparation, and he chose the main variation. 37 d3 c/;c7 38 llh8 White resigns. After 39 �h7+ �c6 40 lth6+ �c5 41 l:.h5+ �c4 there are no more checks, and that means that his rook is too late in reaching the d-file. Game 372 Botvinnik-Kavalek Beverwijk 1969 King' s Indian Defence 1 2 3 c4 d4 l£ic3 g6 .i. g7 c5 Here it became clear to me that Kavalek was familiar with my game against Matulovic (No. 357), had found some improvement in it for Black, and was intending to use his innovation. Since, as I mentioned in my notes to the afore-mentioned game, this variation had been thoroughly prepared by me, 11 0-0 ..i.xe3 It is interesting that up to the last move this had all occurred previously (Kluger-Forintos, 1964 ). Therefore, how ever, Black chose the less favourable exchange sacrifice: 1 I . . .:xe3 12 fxe3 365 .i.xe3+ 1 3 <it>hl tt:lli 5 14 · ife l , and White' s advantage was obvious. 12 fxe3 1i'e7 Black does not risk taking the pawn: after 12 . . .l:.xe3 13 'i'd2 ( 1 3 . . . :e7 1 4 tLlg5 tLle8 1 5 'i'f4 f6 16 .i.c2; 1 6 ltlge4 or 16 g4 is also possible) White gains time and obtains a strong attack. Black's idea is to provoke e3-e4, which restricts somewhat the activity of the white pieces, and then to gain control of e5, establishing his knight there. Alas, the half-open f-file proves to be of great importance! Another idea was tried by Polu gayevsky, in a game with Gligoric in 1 975 : 12 . . . <it>g7 followed by . . . ltle8 and . . . f7-f6, thereby neutralising the pressure on the f-file. 13 14 e4 ifd2 moment. 14. . . <it>g7 would �have met with the same reply as in the game. 15 l:lf2 During my home analysis I had prepared 1 5 .i.c2, but I decided to chose a different move, not wishing to allow the opponent counterplay after 1 5 . . . b5 . 15 16 17 18 19 lDbd7 • . . • • lbe5 thxe5 ltafl l:l xf7 1f xe5 lhd7 11'd4+ <it>hl Of course, not 1 9 l:.7f2 because of 19 . . . tiJeS 20 l:tdl .i.xh3 ! A good plan was found by Gheorghiu (against Andersson, 1 974). By playing 14 .:f2, he prevented 14 . . . liJe5 because of 1 5 �e5 ifxe5 16 'iff3 �g7 17 l:afl . 14 • A serious mistake. Black should have delayed a little the carrying out of this plan, and first defended his f7 pawn ( 1 5 . . . l:f8). Then he could have re grouped with . . . ttle8 and . . . f7-f6 (roughly as Polugayevsky played - cf. the note to Black's 1 2th move) . Then Black would have retained a solid, although rather passive position. 19 20 a6 This move is useful, to prevent White from playing · liJb5 at . an appropriate the5 11ff4 White is threatening -to conclude the game by 2 1 l:lf8+:;: Black had been 366 position of his king. However, he is not long in returning the compliment. plmming the defence 20 . . . i.fS, but only 1 1ow did he notice that after 2 1 ihb7 t/\xd3 22 'iih6 it is not possible to defend the seventh rank. He has to cast caution to the winds. 20 21 29 30 31 Ve5+ •e6+ il.e2 <i!tg8 <i!th8 l:tf8 .i.xh3 J\e2 Although the material balance is restored,. White has a decisive attack. 21 22 li.d7 1!ff6 Initially I wanted to play 22 l::tf6 with t he threats of 23 lhd6 and 23 l:tdl (the i nunediate 22 ltdl will not do, since Black gives up his queen for two rooks). Examining the reply 22 . . .c3lg7, I did not l ike the fact that if 23 l:Idl there can follow 23 . :f8 24 k!xd4 :xf6, when Black continues resisting. But in the event of 23 lixd6 l:f8 for some reason it did not occur to me that I could reply 24 Hf6 with a decisive advantage. Thus I l bought for some twenty minutes, and did not make the best move. .. 22 23 24 25 1f xf7+ 'l'xd7 'iffxd6 lt)xf7 'ii?h8 1Xf8 Simpler, as was rightly pointed out by Kavalek, was 25 :n r!xf7 26 1i'xf7, and Black himself is forced to go into the endgame by 26 . . . ifg7. 25 26 27 28 i.xfl ife5+ 'ffe6+ l:h.il+ 1ff2 <i&?g8 'it>b8 Black does not make use of a good opportunity, which was to play 28 . . . �g7 and then hide his king on the safe square h6 . . Then White would have had insufficient compensation for the open It only remained to play 3 2 d6 ! , as pointed out by 0 'Kelly, and within a few moves it would all have been over. I would have made this move, of course, if in time trouble I had managed to see that the 'terrible' threat of 32 . . J �f4 could be parried as follows: 3 3 i.g4 ! (33 . . . 'i'h4+ 34 �h3 llfl + 3 5 <i&?h2 'i'e l 36 'i'e5+ �g8 37 Ae6+ � 38 'ii'h8 mate, or 35 . . . 'i'f4+ 36 g3 'i'f2+ 37 .i.g2 'i'gl + 3 8 '&t>h3 :f2 39 ife5+ '&t>g8 40 'ii'd 5+ <3;;g7 4 1 e5 'i'c l 42 '&t>h2). But I hastened to complete the bishop manoeuvre. thinking that then there was no way that the black rook would be able to leave the eighth rank. 32 33 34 Js 1ie5+ .ltg4 i.b3 '&t>b2 'ifi>g8 1ih4+ 'lfel+ . . :n An unpleasant surprise. Now the white king is threatened with mate, 367 while the black king amazi�gly slips out of the encirclement. <t>f8 36 ii.e6+ 37 Game 373 Botvinnik-Kurajica Belgrade 1969 1fb8+ Queen's Gambit The variation 37 'i'd6+ <3Jg7 3 8 'ii'e7+ �h8 (3 8 . . . 'it>h6 3 9 �h3 'i'e3+ 35 40 g3 'i'f3 4 1 'iib4+), in which White has to give perpetual check, was pointed out by grandmaster Benko. 37 38 • • • '3;e7 38 39 40 1Wh8+ 1l'e5+ 'itf6 <it>g5 �b6 cxd5 4 5 6 ii.f4 el Is it not possible to try and run away with the king? For example: 4 1 ®h3 .D.h l + 42 @g4 'i'h4+ 43 � �fl+ 44 @e3 'i'f2+ 45 �d3. But then 4 5 . . . 'ii'xg2 ! and White is again forced to give perpetual check. 41 42 43 1fb8+ 'it>g5 @h5 "l'd8+ 1fh8+ Draw agreed lbcJ d5 e6 iJ..e7 exd5 c6 .i.d6 Previously only 6 . . . .if5 was played here, and it would appear that 6 . . . i..d6 was first employed in the game Gligoric-Portisch ( 1 967). It leads to quieter play than that in the variation 6 . . . i.f5 7 g4 (cf. No. 3 1 3). A good plan for White against 6 . . . ..td6 has not yet been found. In the present game too he gained only a minimal advantage, and yet, apart from anything else, he has an extra tempo, which Black has spent on playin_g . . . i.f8-e7-d6. 7 , d4 c4 White also has the possibility of 4 e4, leading to open piece play. 'ifxh7+ Or 38 'i'g7+ <it>e8 ! 1 2 3 4 .i.g3 The afore-mentioned game continued 7 .i.xd6 1i'xd6 8 1'.d3 l£Je7 9 ti)f3 ti)d7 10 0-0 tbf6 1 1 'ifc2 0-0 1 2 %tab 1 g6 1 3 b4, and White's position was perhaps only slightly more pleasant. 7 t;)e7 Instead of this Gligoric recommends playing 7 . . . .i.f5 immediately, although it seems to me that after 8 'S'b3 Black has some problems over the defence of his b7 pawn. 8 ttlf3 0-0 • 9 368 • • .i.dJ .i.f5 This is the clever point of Black's plan. After 10 �xf5 lDxf5 the bishop at d6 is defend� and the attack on the b7 pawn ( 1 1 'i'b3) is easily parried. 13 14 h4 li)xd6 This advance of the rook' s pawn was what White was intending when he played 1 1 lDxd3. But the entire plan requires too much time, and by straight forwardly carrying it out, he ends up in a difficult position. 14 15 16 h5 h6 lhd7 g5! Reckless and flippant - a further loss of time . . . 16 17 18 10 ll.\e5 10 11 lDxd3 11 12 13 ilf3 .i.xd6 g4 1ie7 Ve6 0-0-0 The start of White's misadventures, which continue for fully 10 moves! Of course, I did not want to waste a tempo on 10 Vic2 (in view of 10 . . . ..txd3 1 1 'ifxd3 ), but then Black would have been unable to solve his main problem - what to do with his bishop on d6? Exchang ing on g3 is dangerous, since this opens the h-file, and meanwhile White has a clear plan: queenside castling and the opening of the position by e3-e4. 18 .i.xd3 Of course, 1 1 ifxd3 was more cau tious, although after 1 1 . . . i.xe5 12 .itxe5 lDd7 13 Jlg3 White' s advantage is insignificant. lt)f5 g6 Now, when the bishop at d6 is defended by the. lmight, White has no reason to defer this exchange. Aae8 This is a mystery - what was Black hoping for? After White's queenside castling (and he had nothing better) 1 8 . . . lDe4 ! would have been especially strong, since the reply :c 1 is not pos sible. Black could also have carried out this manoeuvre on the next move, but he played it only when White was fully ready to parry it! 369 19 20 1lg2 <iflbt Now the initiative again passes to White, and it is Black who begins to experience difficult times. 2l 22 23 li)xe4 li)c5 dxc5 improved the placing of his pieces. 27 28 dxe4 ttlxc5 l:td8 The position has simplified, and to White's clear advantage. Black's king is insecure (th.is is when the h6 pawn comes in useful), and his kingside pawns are disunited. He must prevent White from becoming active on the d file� for this reason 23 . . . 1i'c4 24 l:d7 1fg3 lld3 25 26 l:txdJ l:ldl exdJ f6 1rxg4 Black assumed that the rook ending after 28 . . . Wxc3 29 bxc3 l:!d8 30 �c l would be unfavourable for him. But with the queens on the initiative remains all the more finnly with White. 29 1fxd3 · 1ih5 Black is obliged to eliminate the h6 pawn, in order to safeguard his king. Wxc5 25 rlhd l was dangerous for him. 24 11fe5+ ii'c3 30 31 32 1ld7 11xb7 'iii>c l 33 lld4! 'l'xh6 1i'g6+ 1fe4 Probably best, but the resulting passed pawn is exchanged for one of the white pawns. The black queen must be driven from its central position. The main tlring in this endgame is not material advantage, 26 1ic4 TI1e only move If 26 . . . 'i'e4 27 1i'd6 ! �g8 28 :xd3 (28 . . . 'i'xg4 29 'i'f6), or 26 . . . :ds 27 l:txd3 (27 . . . 'i'e4 28 'i'e5+ ! ), and Black stands badly. White, in � is now forced to play 27 'i'e5+. After 27' ..:d2 'i'e4 ! (with the threat of 28 . . . 1ifhTf) Black would have but the placing of the pieces and the active passed pawns. Therefore in the end White must eliminate not the a7 pawn, but the one at c6 ! 33 34 35 'iii>c 2 1le7 'i'hl+ ilf3 Here the queen is excellently placed; Black cannot pfay 3 5 . . �·Wxf2+ because of 36 l:d2 1!i'f5+ 37 e4 'i'c8 3 8 :d7. 370 35 36 ' !id2 37 �dl l:lg8 1'f5+ White aims to keep his king as close possible to the kingside, so that in the �vent of the queens being exchanged his king can stop the black h-pawn. He gets :iway with this, since with his next few moves Black gives some useless checks. 1 1 is not clear whether White could have hoped to win, if BlacJc had played n g4. as . . . 37 38 39 40 41 42 <it>el <it>e2 <it>dJ e4 <ifi>e2 Vg4+ ifgl+ 1ig4+ 1ff5+ iff4 42 43 44 45 46 47 :!d7 'it>el 'it>d2 <ifi>e2 l:td6 g4 'Iffl+ 9hl+ 111'h6+ 1tg6 Black's last hope is to open up the game with . . . g4-g3, but this leads to the loss of a pawn. 47 48 fxg3 49 'it>el 50 <it>dl 51 <it>cl g3 -.its+ W'hl+ 1ig2+ W'g l+ Now White retreats his rook and defends his c5 pawn. 52 53 54 55 lldl <ifilbl .:ct ife6 \ie3+ 1'f3 h5 Finally the c6 pawn is eliminated, and this is bound to decide the outcome. 55 56 • • • 11fxc6 :xg3 h4 Black also fails to save the game by 56 . . . 'i'd3+ 57 <it?al 'i'd4 58 1fd5 ! l1gl 59 <iit>b l . 5 7 1'd5 In this position Black had to seal his move, choosing between two options. One of these was 42 . . . iig4+. In this case after 43 'iiii>d3 'i'D+ 44 �c4 l:tb8 ! (or 44 g4 45 <it>b4 ! ! and then :d7) 45 hid8+ J:!xd8 46 'i'xd8+ rbg7 4 7 1i'd7+ �h6 48 'iixc6 ifxf2 49 'i'e6 ! \ixb2 50 'i'h3+ <it>g6 5 1 c6 White would have obtained a won queep.Jending. The other option Wid to be chosen . . . ... 371 � g5 58 to this central square. In addition, the following exchanging operation be comes possible: . . . lDxe4; "ifxe4 d6-d5. 1i'c4! The threat of advancing the c5 pawn is irresi�tible. 58 . ... ltg7 7 Or 58 . . . h3 59 c6 h2 60 c7 h i if (60 ... ltgS 6 1 'i'xg8+) 6 1 c8'i'+ �h7 62 11f7+. 59 60 c6 l:t.gl lLg4 Against Gipslis (Game 3 16) I played 7 . a5 immediately, which is probably slightly more accurate; but if Black is aiming to play . . . d6-d5, then he prob ably cannot get by without the move in the game. It has to be assumed that the afore-mentioned game had influenced the choice of opening by Matanovic, who had prepared, as it seemed to him, an improvement for White. It can also be said that other continuations (7 . 'ifc7 8 e5 dxe5 9 dxe5 llJd5 10 llel tDa6 1 1 c3 ttlc5 1 2 .i.c2, Tal-Gurgenidze, 1 972; 7 . . . d5 8 ttlbd2 b6 9 exd5 cxd5 10 l:te 1 e6 1 1 ltJe5, Gufeld Planinc, 197 1 , and 7 . . .ltJa6 8 lld 1 �c7 9 e5 l2Jfd5 10 llJbd2, Ciric-Hort, 1967) do not solve Black's opening difficulties. . . l:lc7 :g7 60. . . l:cS 6 1 W'f7 would have led to mate. 61 62 63 -*.b3 1i'hl+ lbg7 11fc1 1f xe4+ cli>at Black resigns .. Grune 3 74 Matanovic-Botvinnik Belgrade 1969 Pirc-Ufimtsev Defence 1 2 3 e4 d4 tLlf3 g6 lL g7 d6 8 9 Now 4 c4 would have led to the King' s Indian Defence, while for the Pirc-Ufimtsev Defence the most typical move is 4 tDc3 . Black has fewer diffi culties after 4 .i.e2 (as Petrosian played against Spassky in the 1 966 World Championship Match) or after the move chosen in the present game. 4 5 6 .i.c4 1fe2 0-0 lbbd2 a5 8 . . . d5 is premature in view of 9 c3 ltJbd7 10 e5 . a4 lLlf6 0-0 c6 This reply pursues two aims: it prepares . . . d6-d5, which restricts the opponent' s light-square bishop, and it defends the d5 square, so that if the knight is attacked by e4-e5 it can move My opponent, who was aiming to reach a position ftdm Game 3 16, repeats 372 tl1e error made there, whereas he had an excellent choice between the sound continuations 9 c3 and 9 a3 . 9 10 • • • d5 e5 The Encyclopaedia considers that after 10 c3 White retains the advantage. It is impossible to agree with this, if only because of the reply 10 . . e6. . 10 11 12 13 h3 lllxfJ c4 lDfd7 .i.xfJ e6 Here is the prepared improvement (Gipslis played 1 3 i.g5, under estimating the strength of the reply 1 3 . llib6). It is true that White succeeds in preventing the attack on his centre by . . . c6-c5, but at a high price: the weak ness of the b4 square will inevitably tell. 1 3 c3 was more circumspect. ii.gs .fia2 16 17 18 19 lJ..e7 .i.d6 �xa2 l:[fal 19 20 21 22 W'xc4 lLld2 .i.c5 23 gJ l:tfe8 lLlxa2 W'bJ After 1 9 l:ta3 'ii'xc4 2 0 ·,'ifxc4 dxc4 2 1 t'Dd2 Black could have • defended his pawn by 2 1 . . . b5, whereas now this will not be possible. . . 13 14 15 The time had come to maintain the balance with the exchange 1 6 cxd5. The attempt to cramp Black's position (16 c5 'ifc7 1 7 Ji.e7) would. have encoun tered a tactical refutation ( 17 . . . tbxe5 18 .i.xf8 tt:Jxf3+ 1 9 'i'xf3 :xf8 with ex cellent compensation for the sacrificed exchange). As for the game contin uation, it leads to a slightly inferior ending for White. llla6 ifb6 1f xc4 dxc4 ltlb6 kta6 Gipslis was unable to make this move, since the b2 pawn was not defended. 15 • • . lLlb4 White did White not regain his pawn? Because in the variation 23 i.xb6 �xb6 24 ll'lxc4 l:tb4 25 tLld6 (25 tt:lxa5 lta8) 25 . . . .tld8 26 l:l.dl .i.ffl Black would again have obtained an 373 advantage. If instead 23 l:.c l , then 23 . . . .i.h6. Therefore White prepares if necessary to block the c l -h6 diagonal by l2-f4. 23 24 l:.cl 25 ltJe4 . • • �d8 .ilf8 24 . . . il.xe5 25 i.xb6 l:txb6 26 liJ,xc4 .ixd4 27 ttJxb6 ..txb6 would probably have led to a draw. This seems very active, but White over-rates his chances. By playing 25 ctJf3 followed by 26 i.xb6 lhb6 2 7 :xc4 he would have almost completely equalised. 25 . . • <lJg7 So that subsequently White should not be able to move away his knight with gain of tempo (tiJf6+); my opponent had overlooked this subtlety. Black opportunely plays his rook to an active position and consolidates his material advantage. 28 29 30 liJd6 l£lxb7 l:c3 liJd7 lLlxe5 lDd3 31 32 ltJd6 �xc4 l£lxc5 lLlb3 33 34 35 36 Iixd4 l:ta3 l£le4 !tel! lLlxd4 c5 ltb6 36 37 38 ltlxc5 l:tc4! Black returns the •compliment'. The exchange of t11e c4 and c5 pawns merely increases White' s defensive resources, in view of the fact that the c6 pawn becomes a target. A more complicated and more promising position · for Black would have resulted from 30 . . . g5 ! In the double rook ending after 32 . . . l:txd6 33 :xc5 l;!d5 34 %tc2 White would have also had real drawing chances. in view of the weakness of Black's queenside pawns. In view of the threat of I:txc5 and :xa5, Black is forced to advance his r and e-pawns earlier than he would have liked. f5 llxb2 Subtly played. 3 8 . ttill 39 tt:Jxe6+ �6 40 :c6 is not dangerous for White, while if 38 . . . l:t.d2 there follows 3 9 :xd4 ! . . 26 <t>fl And this is an oversight, leading to a lost position. Both 26 tDd6 and 26 .i.xf8+ <it>xf8 2 7 tt:Jd6 would have been better. 26 27 dxc5 .i.xc5 l:id4! 3 74 38 39 40 41 l£ld3 <&t>g2 f4! 41 ... e5 libl+ @£6 Forcing a drawn ending. e4 the c-file. But even so that subtle analyst Geller found a draw for White: 5 1 �e3 ! ! 'iit?g3 5 2 @e2 f4 5 3 ®fl ! @xh3 54 <iii>f2 @g4, and since Black will be forced to advance his pawn to f4, after the loss of the a4 pawn White will succeed in reaching c l with his king! White, however, chose the other continuation, which he thought would give him a draw. 43 White is a knight up, but Black will regain it, both players having a choice between a rook ending and a pawn ending. First White has to take a decision - should he play 43 l:td6+ or 43 l:td5 ? I had to make a thorough study of both these continuations. So, let us first look at 43 :ds. With the rooks on, Black has no winning chances, e.g. 43 . . . exd3 44 ktxa5 @e6 45 l:te5+ <it>d6 46 <t>f2 d2 (if 46 . . . l:tg l , then 47 l:.e3 36) 47 �e2 l:tgl 48 �xd2 l:txg3 49 l:e3 . This means that he must go into the pawn ending: 43 . . . l:td2+ 44 � ltxd3 45 l:xd3 exd3. It may seem that Black's advantage is enough for a win: 46 @f2 g5 ! (but not 46 . . . <iti>e6 47 <it>e3 <it>d6 48 'iit?xd3 �d5 49 g4 ! fxg4 50 hxg4 h5 51 f5) 47 fxg5+ (here 4 7 <iti>e3 is now bad because of 47 . . . gxf4+ 48 gxf4 <it>e6 49 <it>xd3 <it>d5) 4 7 . . . <it>xg5 48 <aire3 h5 49 <airxd3 h4 50 gxh4+ <ittxh4. Indeed, after 5 1 �e2 <it>g3 52 h4 (or 52 <it>e3 f4+) 52 . . . <it>xh4 53 00 <it>g5 Black succeeds in �g the a4 pawn and in not allowing the enemy king onto 375 :d6+ The drawback to this move is that now the black king will easily cross the e-file, which may be highly significant. 43 44 ... l:ta6 �e7 exd3 The alternative 44 . . . lld2+ 45 ti:Jf2 e3 46 l:txa5 l:txf2+ 47 �gl did not bring Black any benefit. 45 lba5 <i!i>d6 Now the natural 46 �f2 loses quickly: 46 . . . :g I ! (47 %:ta8 <:/iJc7 48 :a7+ <iti>c6 49 l:la6+ 'it>c5 50 :as+ 'ittc4). Therefore White must urgently place his rook behind the d-pawn, and this means allowing the black king to advance. ·. 46 Jla8 �c7 Played to gain time on the clock; the immediate 46 . .. 'it>c5 was also possible. 47 48 49 :a5 .:aS ct>f2 'it?c6 'it?c5 l:tal A:r:i excellent post for the rook. From here it can control its own pawn's queening square (dl) and at the same time restrain the advance of the enemy passed, pawn. Not surprisingly� White takes urgent measures to force the opponent' s rook to change its position. Note that this could not be achieved by 50 <it>e3 because of 50 . . . l:.gl. 50 51 ltd8 'it?eJ <i>c4 Now� however, this move is possible: if 5 1 . . .llgl there follows 52 :!d4+. TI1erefore Black is forced to drive the white king to a less favourable position. 51 52 53 'it>t'2 <ifi>f3 :et+ l:te2+ l:e6 a5 l:tc8+ <ifa>cJ <ii>d 2 This . position of the rook. is also highly appropriate, since it cuts off the enemy king and also stops the a-pawn. 54 55 Black must play carefully, not cutting his king off from his . passed pawn. And he safely avoids this trap. After 55 . . . �b3 the pride of Black' s position - hi s d-pawn - would have fallen: 56 a6 ! l:xa6 (56 ... d2 57 l:.d8 �c2 58 a7) 57 'ii?e3 :Id6 58 <it>d2. 56 h4 After the game Matanovic suggested that 56 J:lc7 would have led to a draw. Indeed, the manoeuvre of the rook to al is no longer possible (56 . . . :el 57 a6 l:tal 58 a7 with a draw), but there is another way to win: 56 . . . h5 57 <it>f2 'iii>d l 58 �f3 d2 59 <it>f2 :e2+! 60 <it>fl l:e3. Now if 6 1 �fl comes the manoeuvre . . . l:e3-a3-al-cL while if 6 1 a6 l:txg3 62 a7 l::ta3 63 � h4 64 <it>fl :a4 65 @g2 �e2 66 l:.e7+ <it>d3 67 lld7+ lt>e3 etc. llel ! 56 An elegant and rapid way to win. 57 a6 l:tal What can White do? 58 l:tc6 @el 59 .:t.e6+ <"Ji>fl 60 l:td6 (60 'ii?e 3? lle l +) 60 . . . d2 6 1 l:txd2 lta3+ leads to a very rare� and possibly unique instance of a rook mate along the rank 376 In the Samisch Variation, which the play inevitably reaches, the position of the knight at d7. is not the best. with the black king to the rear of White 's. And, what's more, at fl . There is the sanie winning idea after 58 :as (58 . . . �ei' . 5 9 a7 d2 60 l:te8+ @fl 6 1 l:td8 na3+ and mate nexi move). In the game White decided to gain some compensation by winning the h-pawn, but he does not get as far as converting his pawn majority on the kingside. 58 59 l:C7 �g2 l:r.e7+ �xh7 <it>el e5 li)e7 7 8 ..i.dJ f3! 0-0 White takes advance measures aimed at retaining a pawn at e4, which will prevent the activation not only of Black's king's knight, but also of both his bishops. since he will be unable to advance his e5 pawn. l:txa6 �dl lt.a2+ <iirfl The reader already knows that 62 litf'3 �e 1 ! leads to White's defeat, due to the threat of mate. 62 63 64 i.eJ d5 Black is aiming to advance . . . f7-f5 as quickly and as comfortably as pos sible, but the move in the game also has a drawback: if White retains his central e4 pawn, the knight at e7 will be restricted. Otherwise it all ends with the same mate. 59 60 61 62 5 6 8 9 10 d2 bial lk7 :ct � White resigns thge2 'ifd2 f5 lhf6 Game 375 Botvinnik-Ostojic Belgrade 1969 King's Indian Defence 1 2 c4 e4 g6 This pawn arrangement is not Wl favourable for White, both after 2 . . . e5, and in the event of a transposition into the Sicilian Defence (2 . . . c5), and especially in the King's Indian Defence. 2 3 4 d4 thcJ ilg7 d6 thd7 10 . . c6 . Black's lot would not have been eased by 10 . fxe4 1 1 fxe4 liJg4 12 .i.g5 ltJfl 1 3 0-0 ! but perhaps he should have considered 10 a.6. . . , ... 11 377 h3 cxd5 12 cxd5 kingside. Therefore, he decided, Black's plan was clear: a pawn offensive against the white king, and the need, even if only for a time, to safeguard his queen side. f4 Black could not wait: White was already threatening g2-g4 ! with a highly favourable situation. In particular. then . . . f5-f4 would no longer have any point, whereas now Black retains the future possibility of a pawn offensive on the kingside. 13 14 .i.f2 a4 a6 Due to the threat of an attack by the opponent with . . . g6-g5, . . . h7-h5 and . . . g5-g4, for the moment White does not detennine the position of his king, but develops his initiative on the queen side. In view of the unpleasant threat of 1 5 a5. Black has no choice. 14 15 16 b4 b6 ltld7 lla2 'it>dl! 1s 19 20 <t>cl @bl 21 22 23 :ct .i.c2! Ji.b3 :n 1ic7 1i'b7 Due to lack of manoeuvring space, Black's pieces are congested, and it is hard for him to mobilise his forces. a5 In the event of 16 . . . bxa5 (here or later) both 17 :xa5 and 1 7 bxa5 ll'ic5 18 ..i.xc5 dxc5 19 l2Ja4 are equally good for White. Therefore Black prefers to wait. 16 17 18 This move was certainly not taken into account by my opponent (he was deceived by White's last move). Meanwhile, after Black had safeguarded not only 'his". but also 'my' queenside, the white king will feel completely safe there, and artificial queenside castling suggests itself. After this Black's activity on the kingside loses its point. ltlf8 .ii.d 7 ii.f6 b5 White's last move convinced my opponent that I was intending to castle The position of the bishop at b3 is more promising, since later White will try to open up the game not only on the kingside, but also in the centre. 378 23 24 25 26 27 . �:ac2 lZ)dt lZ)b2 lZ)dJ memories remain. With every move the white pieces gain more and more scope. <3itg7 tbg8 i.d8 lZ)f6 32 33 34 35 Here the lrnight is excellently placed. I f the opportunity presents itself. it can be sacrificed either for two pawns c liJxf4), in order to destroy the oppon ent's pawn centre, or for one (CDc5). creating two powerful passed pawns. · ·rxg3 �f6 ltlb5 1ixg3 1lf g2 tbg3 lZ)e8 27 35 lLlxgJ Thus the excursion of the black knight to h5 merely leads to the ex change of this important defensive piece. However, 3 5 . ttJf4 would also not have worked because of 36 CDxf4 exf4 (36 . Iixf4 37 lLlh5+) 37 liJe2, when White acquires the important d4 square for his minor pieces. For example: 37 . . . liJd7 38 liJd4 ltJe5 3 9 liJe6+ .i.xe6 4 0 dxe6 :c7 4 1 h 5 g 5 42 .i.d4 . .. 28 h4 28 29 30 ifdl 'Bgl And so, after arranging all his pieces in the best way possible, White begins carrying out his main strategic plan, which will rapidly prove successful. h6 �c8 36 37 White does not object to the ex change of rooks; in the attack he will be able to manage without them. 30 31 32 :xc2 g3 .. 'l'xgJ f4 :c7 Here it is, the long-awaited break through, to which Black no longer has any reply ! 37 38 �xc2 .i.c8 Of Black's math hope - that of blocking the kingside pawns - only 379 i.xc2 lixc2 The white bishop is now aiming at the g6 pawn. · 38 • • • exf4 39 40 <bb7 1!ie7 .i.d4+ 1i'xf4 Here 4 c4 is more energetic, taking play into the King's Iridian Defence. But the tournament Sitµation meant that Matulovic only needed a . draw, and he preferred a quiet variatiOn of the Pirc Ufimtsev Defence. 4 5 6 7 41 42 e5 .itxe5 But not 42 42 . . . 'i'xb4+. 42 43 44 ..tb2 h5 White waits for the natural . . . e7-e5, in order to begin exchanging. But if he is aiming for more complicated play, it is more advisable to choose ·7 d5 CDb8 8 h3 or 8 :e l . lDxe5 because of Game 376 Belgrade 1969 Pirc-Ufimtsev Defence lbc3 • e5 dxe5 • dxe5 . 1fxf4 lLlxf4 <i>g8 hxg6 Black resigns e4 d4 lLltJ • More dangerous was 8 . . . ltlxe5 9 tDxe5 dxe5 10 f4, when White has the initiative. Now if 9 .i.b5 (or after 9 'i'xd8 :xd8 - 10 .itb5) Black defends successfully with 9 . . lDd4. li)d7 iff8 Matulovic-Botvinnik 1 2 3· 4 7 8 dxe5 White completes the destruction of the royal citadel, even the queens no longer being needed for the conversion of his advantage. 44 45 46 ltlf6 0--0 ltlc6 i.e2 0-0 h3 g6 J..g7 d6 9 1i'xd8 The initial cause of all White' s sub sequent difficulties - primitive play for a draw rarely proves successful! After 9 .ig5 ! .ie6 10 'i'c l followed by 1 1 l:.dl White would have retained an opening advantage. 9 10 11 12 J.. g5 :rd1 .ie3 llxd8 ..te6 h6 It is hard to suggest anything better. Both 1 2 ..th4 and 1 2 ..txf6 are dubious, but now the initiative is with Black. 12 lbe8! The position is almost symmetric, but this 'almost' consists in the fact that White cannot occupy the central d5 square with his lolight, whereas . . . l'Dd4 3 80 will certainly follow. All this is due to the better position . of the king's bishop �. - ! · at g7 . 13 14 15 lDd4 l!d7 l'.Del J..fl l'.De2 The aim of this manoeuvre is obvious: after 16 i.xd4 exd4 17 tt:Jf4 to obtain opposite-colour bishops. How ever. this does not ease White's defence. 15 • • l:.ad8 • Threatening 16 .. tbD+. obliged to force events. . 16 17 18 i.xd4 ll'lf4 lDxe6 White is exd4 H.e7 klxe6 And now the e4 pawn is attacked. After 1 9 .i.d3 ll'ld6 20 f3 c5 2 1 b3 b5 the knight at e l would be out of play and the black pawns would bec01ne active. Therefore White immediately plays . . . 19 f3 t£id6 Another plan was 1 9 . . . l!b6 20 .:abl (or 20 lt1d3) 20 . . . ti:Jd6 with quite good prospects. 20 21 lhd3 llet Also tempting was 2 1 . . . lLJc4 2 2 lbb4 (22 lhc5 t'De3 23 ttJxb7 l:tb8 24 trn5 �xb2) 22 . . . ltJe3 23 ttld5 liJxd5 24 exd5 :xel 25 !txel l:!xd5 26 %:te8+ i.f8; although Black would have won a pawn, the position would have become simplified, and the presence of opposite colour bishops could have led to a draw. l:te7 h5 It is interesting that in the game Ciric-Botvimiik (No.370), where there was the same opening, but a different variation, after 23 f3 a position amaz ingly similar to this one was reached. Only the pieces were arranged slightly differently� but the character of the play was the same . . . 22 22 23 24 hxg4 b3 hxg4 g5 A difficult decision, but 24 . . . lllc4 was already threatened. 24 381 g4 This leads to new difficulties: the dark squares are weakened still further, and the placing of the pawns at e4, f3 and g4 is fixed. The only good chance was 22 g3 ! • • . ll'lb5 25 lle2 i.e5 27 28 Ad2 .i.g2 a5 l:th7 37 . . . .i.f4 38 l:td3 would not have introduced position. 38 39 anythiifg new into the - .i.fl ltixe5 <t>e7 This exchange consolidates Black's positional pluses, brit White was almost in zugzwang. Now at least his bishop is activated. Despite the apparent passivity of White's position, Black has to reckon with the possible manoeuvre of the bishop to f5 followed by the creation of a passed pawn by f3-f4. i.d6 c6 Restricting the activity of White's bishop on the long diagonal in the event of f3-f4, e4-e5 etc. 29 <ifi>f8 36 . £Dc4 37 . ·<i>gl Black is attentive. TI1e · last active possibility for White is· to double rooks on the h-file. Now he can gain the h2 square for his rook only at a very high price - by exchanging knight for bishop, i.e. by renouncing the opposite-colour bishops! tDc3 26 a4 a6 39 40 i.c4 fxe5 <li>d6 Now a possible minor piece ending will be won for Black due to the weak ness of the a5 pawn. 30 31 32 33 <ittf2 l1h1 l:.b5 l:lxh8 <lig7 f6 lab8 White hurries with this exchange� but in time it would all the same have been inevitable. 33 34 35 �1 lLlb2 citxb8 <lig7 .il.e5 41 <ii!i>g2 This move was sealed by White. In the event of 4 1 l:.t2 (in order to take the rook via fl to al for the defence of the a5 pawn) Black would have won, by first taking play into a rook ending 4 1 . . .�d l 42 lld2 ttJe3 43 c3 tt:Jxc4 44 bxc4 �c5 45 cxd4+ exd4 46 <it>f2 <i&>xc4 4 7 %:tc2+ �b4 48 �e2;and then into a 382 -- pawn ending � 48 . l::th2 + 4.9 �d3 .:xc2 50 �xc2 c5 ! . 41 5 . �.. llh6 Hindering White's main counterplay, involving playing his bishop to f5 and advancing f3 -f4. 42 43 44 ii.f7 .i.e8 .td7 5 6 �cs @b4 . • • lbb1 f4 <it>g1 .i.f5 • • d5 7 l:td3 ltlc6 liJce7 f3 If 7 11Jg3 then 7 . . . c5 is good for Black. who can also consider 7 . . . c6. Also bad is 45 l:le2 @c3 followed by . . . li:Jd2. 45 46 47 . 48 • By playing 6 . . . 11Jce7 Black avoids simplification, although in the King's Indian Defence at e7 the knight is· re stricted,. 6 . . . tiJd4 7 i.e3 would have led to a position from the game Botvinnik Alexander (No.270). Or 44 f4 gxf4 45 .i.d7 ttJxe4 46 �e2 tiJf6 47 .ilf5 :h4 48 g5 lth5 . 44 45 ttJge2 In this variation of the King's Indian Defence it is simpler and probably. better to play 5 d5. � 7 8 tiJaJ 1Dxc2 gxf4 .i.e3 f5 ltJe3 This would have been a fully appropriate moment to record the result of the game. 49 50 51 Iih5 g5 lDxf5 g6 �h.-f5 exf5 White resigns If 52 :d2 there follows 52 . . . f3 (52 .:tf8 also wins). The game was awarded a special prize by the Yugoslav newspaper ' Sport'. . . . Game 377 Botvinnik-Suttles Belgrade 1969 King's Indian Defence 1 2 3 4 d4 e4 c4 l£ic3 . . g6 · ig7 d6 · e5 8 • . . .i.b6 Since White has already spent a tempo moving his queen's bishop, Black considers that he has the right to offer with loss of tempo the exchange of the dark-square bishops. The game is simplified. and my opponent achieves a satisfactory development. But this ex change also has its drawbacks. If the position is opened up, the position of the black king after kingside castling may 383 prove not altogether secure. · It is on this that White builds his plans. 9 f5 pawn that Black obtains in an · analogous situation is advantageous to him. This is also the case .. in the present game - the f4 pawn will control the e5 · square and restrict the black pieces, while White's knight will head for the kingside via cl-b3-d4. Therefore Black should have restricted himself to 18 . . . fxe4. .i.xh6 · White himself exchanges bishops, to divert the enemy knight from its natural position at f6. 9 10 11 11d2 g3 Preventing . . . f5-f4, aft�r which it would be no� so easy to approach the black king. 11 12 . • • .*.g2 0-0 c6 On which side should White castle? The play would have been sharper, of course, after queenside castling, but for competitive reasons I wanted to aim for a win in this game without any risk. My standing in the tournament was not brilliant. and in such a situation every point was highly important. 13 14 15 16 0--0 cxd5 <ifii>h l ll.ael cxd5 'ifb6+ .i.d7 18 19 20 Preparing f3-f4, in order to open the position and activate the knights. 16 17 18 f4 lDct 21 22 since the white knight does not now reach f4, whereas Black obtains a con venient base for his pieces at f5. However, all this is misleading. It is sufficient to exchange the colours of the pieces, and one is reminded of the King's Indian Defence, where Geller and Petrosian showed that the isolated exf4 fxe4 lCif5 Black already has to parry the threat of a check on the al-h8 diagonal. Now if 2 1 'i'c3+ he replies 2 1 . . . 'i'd4. </;;g7 l:tae8 A cunning move, provoking Black into exchanging pawns on f4 and e4, gxf4 lDxe4 lDb3 .i.hJ :.cS lCigJ+ After this unnecessary exchange White will be able to occupy d4 with his knight, and his pawns are repaired. 23 24 hxg3 iDd4! ..i.xhJ ltfe8 If 24 . . . �xfl 25 tbe6+, and White wins easily, e.g. 25 . . . <it>hS 26 l'.LJxf8 i.b5 (26 . J:txf8 27 'i'c3+) 27 a4 �xa4 28 b3 37, or 25 ... <&t>g8 26 tiJf6+ <it>h8 27 384 . tlJd7. 26 l:.dJ 1ib4 Black could not delay in view of the threat of 27 b3 followed by 28 Wb2, but, instead of aiming for the exchange of queens, 26 ... ili5 suggests itself, forcing 27 tiJxf5+ gxf5 28 l:tb3 'ilc7 29 tt:Jc3 with a roughly equal position, and White would have been punished for his error on the 25th move. 1!fxd2 b3 l::tc7 :xd2 Ilde2 i.g4 If 29 . . J :ke7 there would have followed 3 0 tt:Jxd6! 30 lteJ .i.f5 This move too allows an unexpected 27 28 29 The previous variation should have suggested to me that in the diagram position it was not essential to waste time moving the rook. But I spent a long time seeking a way of developing my knight stroke. Black should defended his rook - 30 . . . i.d7. attack, and did not find anything. Alas, what told was my chronic illness - lack of combinative vision. Meanwhile. the have problem would have been solved by 25 b3 ! ! This move is especially pretty, since it is a ·quiet one (without any attack) . And if Black accepts the sacrifice - 25 . . . i.xfl (after 25 . l:txe4 26 �xe4 i.xfl White wins easily: 27 tbe6+ �g8 28 'ifb2 tbe5 29 l:tx�5 dxe5 30 Vixe5 l:tc7 38 3 1 'fixc7 'fixc7 3 2 tiJxc7 Wf7 3 3 d6 @f6 34 lt>gl .i.h3 35 @fl), then the ne:\.1 move is also a spectacular one - 26 ifb2 ! Now after 26 . . . 'ltn'8 27 ltJe6+ Z!xe6 2 8 dxe6 (28 . 'iib4 29 'i'f6 lk7 30 exf7 'i'xel 3 1 'i'd8+ <l;xf7 32 ttJg5+, and mate is inevitable) Black ' .. . . would have to resign. However, the combination was seen not by me, but by the Bulgarian grandmaster Tringov. who pointed it out immediately after the game. ltc4 25 :o 31 32 33 34 l£ixd6 l£ixf5+ lixe8 lbe8 ll\xd6 gxf5 l£ixe8 <&t>g6 Despite being a pawn up, it is not easy for White to win, since the black rook may become very active. First Black must move his king off the seventh rank, so that, when his rook also 385 leaves it, White will not · be able to attack the pawns with check. Also, given the opportwri.ty the black king would like t<rteach g4 . . 35 Itel Threatening to plac.e the rook at d2 behind the passed pawn. 35 36 37 <i>g2 :e6+ 38 ltd6 38 39 40 lth6 :m1+ There was · also a win by 42 1hb7 l:.xa2 43 �h4 l:lh2+ (43 . . . �e6 44 i:tc7) 44 <it>g5 :g2 45 'it>xf5 .l:xg3 46 l:!xa7 J:txb3 47 <br6. 42 l:icl+ l:tdl <i>f7 If 37 ... 'it>h5 White would have won by 38 hte7. This diverts the black king towards the queenside, and thereby makes it easier for White to advance his king on the kingside. <l;e7 lbd5 ci>d6 41 <i>hJ 42 'it>b4 lla5 • • b5 43 44 45 46 47 48 'it>g5 �xf5 g4 �a7 g5 JJ.a8 Jlxa2 a5 llaJ <i>c5 'it>b6 llxbJ 49 50 51 52 53 g6 <it>f'6 g7 l:tgJ b4 <i>b5 Of course, restoring material equality does not give Black any chances of saving the game; his passed pawns are hopelessly lagging behind the opponent' s. If 40 . . . <t>f'6(e6) 4 1 a4, then White wins another pawn, whereas now if 4 1 a4 Black has the reply 4 1 . . . �c6. Black aims for activity on the queen side, but White's threats on the kingside are stronger. • Trying to avoid the variation given in the previous note. g811 l:bg8 .:lxg8 a4 Black resigns Although I failed in a combinative battle, I was nevertheless able to use my customary weapon - technique in the endgame - to gain a win. Unfortunately, sometimes a master learns after a game that in the heat of the battle he missed an opportunity for a combination, and was unable to unravel its secret. Moreover, the prettier the tactical operation that he failed to carry out the more vexing his mistake. The reader, however, will not feel any loss, ·, since the aesthetic impression from the missed combination is no less than if it had been carried . out on the board. . It is only for this reason that I have given the above game. 386 Game 378 Matnlovic-Botvinnik USSR v. Rest oftf:e World Belgrade 1970 Pirc-Ufimtsev Defence 1 2 3 4 e4 d4 tl)fJ i..c4 g6 .i.g7 d6 4 c4 would have led to the King's I ndian Defence, but for the Pirc U fimtsev Defence 4 tt:Jc3 is more l ypical. After 4 Jlc4 I had already t.:mployed several times a set-up i nvolving creating a barrier on the a2-g8 diagonal ( . . . c7-c6 and . . . d6-d5). And a l though my opponent was also no novice in this variation, Black naturally had no grounds for avoiding the custom ary continuation. I have never been afraid of mirages. 4 5 6 7 fie2 ..tbJ .i.g5 tDf6 c6 0-0 to try and refute Black's plan. He had already successfully used this last move against Hubner in Athens a year earlier. White prevents . . . d6-d5, since now in some variations the e7 pawn is pinned, and it needs to be advanced to e6. 7 h6 This is the whole point ! In the afore mentioned game Hilbner played 7 . . . e5 immediately, but this central counter is best carried out after first driving back the bishop to h4, where it is not altogether well placed. 8 .i.b4 e5 Now Black has in mind counterplay involving . . . ttlli5 -f4. 9 10 dxe5 l!Jbd2 dxe5 Threatening to take the e5 pawn, which was not possible immediately because of 10 . 'iVa5+. 1fc7 10 . . 11 12 lLlc4 J.gJ �h5 li)f4 The thematic 1 2 . . . i..g4 was not pos sible here because of 1 3 i.xe5 i.xe5 14 liJcxe5 tiJf4 1 5 'i'd2 with a clear advan tage to White. Thus, after Gipslis (No.3 16) and Matanovic (No. 374), Matulovic decided 3 87 13 .i.xf4 14 0-0-0 15 . : , es exf4 ..i. g4 It was not.possible for White to win a pawn. After 15 h3 .i.xf3 16 'ifxf3 t2Jd7 17 lDd6 tLle5 18 'i'xf4 !lfd8 he would have lost a piece (after any knight move, even to f7, there follows 19 . etJd3 +, while if 19 ifd2 .i.f8). . 15 16 . . • 1fe4 . ltld7 Itad8 For Black too there is no point in going chasing a pawn - after 16 . il.x:f3 17 gxf3 tDxe5 18 li:)xe5 'ifxe5 19 Wxe5 .i.xe5 20 :d7 White has quite adequate compensation. .. 17 18 1fxf4 1ixf3 It is obvious that Black has won the strategic battle. He has already prepared a pawn storm on the queenside, whereas White essentially cannot do anything. ..txf3 b5 Black wants to regain the pawn, without exchanging knights, so as not to ease White's defence. But this does not mean that White should aim for this exchange in any situation. avoiding more ex-pedient continuations. 19 20 21 llleJ 'ifgJ a3 21 22 • • . �h7 lll g4 Aiming for the exchange of knights instead of the active continuation 22 f4 and 2 3 .f5 . Here I realised that it might be possible to carry out Black's strategic plan: the bishop at b3 cannot hinder the activity of the dark-square bishop. 22 23 24 25 lD:le5 1t'f3 libel h5 i.xe5 <i;g7 .tf6 c3 ktxd8 g3 28 29 JO 31 ..tc2 axb4 cxb4 c4 b4 axb4 ifb6 32 J:ldl 'lfxb4 Caution dictated 28 h3 . It is good when pawns on squares of the colour of the enemy bishop cramp it, but bad, as in the given case, when they can be come a target. lllxe5 a5 With the threat of 22 f4 lild7 23 'i'xg6. c5 hlxd8 26 27 28 The first of three errors made by me in time trouble is not as yet so sig nificant. Simpler was 3 1 . . . 'i'a7 3 2 �b l .:ta8 (3 2 . . . 'i'd4 33 'i'c3 'iixf2 34 'ife3 'i'xh2 39) 3 3 'i'a3 'i'xf2 (but not 3 3 . . . 'i'd4 because o f 34 �d l 'i'e5 3 5 f4) 34 'i'e3 'ii'xh2 . But this is already a serious mistake. After 32 . . Jie8 ! White would have been unable to avoid 33 ·�'l'a3 'i'xf2, since 3 3 J:td7 'ii'xb4 3 4 'i'dS Itel+ leads to mate. 388 33 l:r.xd8 34 <i>dl 35 1ld5 36 <ii?e2 If 4 1 'lt>dl .i.c3 ! , and mate is un avoidable. The game mentioned in the note to Black's 1 3th move concluded with such a mating finish. What a curious coincidence is possible after the consistent implementation of similar plans ! As I have already mentioned, grandmaster Matulovic should have known my game with Medina, but did he lmow it? 'ii xb2+ li.xd8 'i'f6 My opponent also 1nisses a good opportunity. He would have gained better saving chances by playing 36 f4, e.g. 36 . . .'i'a l+ 37 �e2 .i.f6 3 8 1fxc4 'W'gl 39 i.b3 . 36 ... i.b6 41 42 43 Black avoids a trap: 36 . . . 'i'c3? 37 0.xg6 ! 37 38 f4 9e4 1lc3 i..d4 'ifa>e2 �dl 1!fgt+ 'i'xh2+ 'ilgl+ 43 .. :i'xg3 could also have been played, of course, but it was much more pleasant to pick up the last white pawn, while maintaining the threat of mate to the enemy king. Black could have won ·a second pawn without any particular problems: 38 . . . .0.gl 3 9 f5 (after the exchange of queens by 39 'ife 5+ White loses) 39 . . . gxf5 40 '+Wxf5 'i'e3+ 4 1 <ifi>fl i.xh2. 'i'eJ+ 44 45 <i&>e2 45 46 1fxg3 11'h7+ <ii?f8 White resigns <ifi>f1 The reader already knows that if 45 �dl there would follow 45 . . . .i.c3 . Game 379 Botvinnik-Spassky Leiden 1970 Queen's Gambit However, here Matulovic misses his l ast hopes of saving the game: 39 �fl 'lia l + 40 �g2 'i'g l+ 4 1 �h3 iifl+ 42 'iig2, and he who commits the last mistake usually loses! 39 40 41 f5 1f.tf5 <it>ft .- gxf5 1ieJ+ 1 2 3 c4 liX3 d4 e6 d5 ii..e7 Although the initiator of this dev iation froin the usual move order of the Orthodox Defence is Black, it is White who gains more opportunities for choosing non-standard · ; continuations. The present game is a typfoal example. 389 4 5 6 cxd5 .i.f4 e3 exd5 c6 .i.f5 7 g4 .i.e6 Finally I gave a laugh, and cheerfully closed my pocket set. 'Have you found it?', Alberic O'Kelly asked me (while I had been working, he had tactfully kept some distance away). · I had indeed found that if now White plays 8 h4, the acceptance of the sacrifice is extremely dangerous for Black: 8 . . . i.xh4 9 'ii'b3 b6 (9 . . . .itc8 10 e4) 10 ltlf3 !l..e7 1 1 i.xb8 l:.xb8 1 2 lDe5 l:tc8 ( 1 2 . . . i.d7 1 3 e4) 1 3 1Wa4 b5 1 4 'i'xa7. White played differently, but also not badly, in the game Vladimirov-Kaminsky ( 1 974): 1 1 l2Je5 lDf6 1 2 i.e2 'i'c8 1 3 :c l 'ib7 1 4 g5 . And here, after an elapse of two and a half years, an opportunity to employ my innovation presented itself. Regarding Portisch' s move 6 . . . .i.d6, see No. 373 . 8 9 All this is theocy. Here White usually played 8 h3 (No. 3 13 ), events developed unhurriedly, and Black normally gained roughly equal chances. But on this occasion I went in for this position with other intentions, behind which there is a story. At the tournament in Palma de · Mallorca in 1 967, several of the rounds were held on the island of Menorca, and immediately after their conclusion a storm blew up. The airport was closed. It was impossible to return to Mallorca, and the competitors were idle for a full twenty-four hours. I sat in the hotel foyer · and with the storm howling outside I analysed the position shown in the diagram, which I had avoided the previous day in my game with Portisch (which went 7 i.d3 i.g6, and soon the players agreed a draw). h4 h5 liJd7 ifb6 A poor move, since Black himself hinders the development of his initiative by . . . b7-b5 . However, after other moves too, White retains the advantage. For example, after the game we examined two variations pointed out by Spassky. After the immediate 9 . . . b5 White has a wide choice of rational replies ( 1 0 i.h3 or 10 .i.e2 followed by i.f3 and l2Jge2 ), while if Black first plays 9 . . . tlJgf6 IO f3 and now 10 . . . b5. then there follows 1 1 lbge2 ! lbb6 1 2 tile 1 lbc4 13 ctJd3 . 10 11 l:lbt f3 lbgf6 b6 The unpleasant 12 g5 was already threatened. But after this weakening of his position, Black's king will be in danger on whiche_ver side he castles. 12 · il.d3 'i'a5 In a gaille ··Ftuman-Geller (1 975) Black castled imrHediately, after which 3 90 carri es out a series of exchanges, and the chances become completely equal. 17 a4 1 3 g5 would have been good, whereas 1 3 �ge2 c5 14 @fl l:fe8 1 5 �g2 ltac8 16 Viel lllli7 17 'i'f2 cxd4 1 8 exd4 i.d6 led to an unclear situation. 13 lhge2 b5 As a result Black has lost at least one tempo, and is late with this advance. Therefore White succeeds in taking control of c5 . 14 15 it)cl! �b3 1fd8 0-0 A highly flippant move, after which Black's position becomes critical. c5!: . .i.xf5 cxd4 9'b6 18 19 20 21 thct ttlf5 i.xf5 exd4 22 23 24 25 .i.d6 tLle2 .i.xf4 'i'd2 tLlxf4 l:tfe8+ 'ifi>fl thb8 Draw agreed Of White's former superiority, not a trace remains. Black's position is per fectly �afe, and it was now I who had to find a convenient moment to offer a draw. Game 3 80 Botvinnik-Larsen Leiden 1970 King's Indian Defence 16 it)e2 a5 Here I suffered from indecision. I spent a long time considering the natural continuation 17 lkl a4 1 8 �c5 lbxc5 1 9 dxc5 ttJd7, but decided that there was no point in sacrificing a pawn, since as it was my position was good . . . Meanwhile, immediately after the game I discovered that after 20 Ab I ! (20 . . . l1Jxc5 2 1 nxc5 .i.xc5 22 1i'c2) White would have obtained a winning position. 17 . tLlg3 1 2 3 4 5 But now control of the c5 square passes to Bla�k, 11� ope;ns up the centre, c4 lLlcJ d4 e4 g6 iLg7 d6 it)f6 f3 Against the move order chosen by Black I normally used to choose the Samisch Variation, since I considered this reinforcement of the centre to be the most unpleasant for my opponent. 5 ... 0-0 Black should not be in a hurry to castle. If, say, the game had continued 5 . c6 6 Ae3 a6 7 Vd2 b5, then 8 0--0-0 would be dubious for White. Black would . not have continued 8 . 'i'a5 9 e5 (9 �b l tl'ibd7 IO i.116 391 . . .. .i.xh6 1 1 1fxh6 .i.b7 1 2 'i'd2 0-0-0 1 3 d5 b4 1 4 ltice2 cxd5 1 5 cxd5 'ib6, as shown 1;>y the game Gheorghiu-Byme, 1 968, le�ds to equal play) 9 . . . dxe5 (if 9 . .. lDfd7 there follows 10 exd6 exd6 1 1 t2Je4, while if 9 . . . b4, then not 10 exf6 bxc3 1 1 fxg7 cxd2+ 1 2 i.xd.2 'i'xd2+ 1 3 hlxd2 l:tg8 with equality, but 1 0 ttJb 1 dxe5 1 1 dxe5 , and the e5 pawn is immune because of the mate at d8, while if 1 1 . . . ll\:fd7 12 f4) 10 dxe5 liJfd7 1 1 f4 0-0 1 2 ttJf3 t2Jb6 1 3 'i'f2 llJ8d7 1 4 ltid4. The correct reply is 8 . . . i.e6 ! ! , e.g. 9 i.h6 i.xh6 IO 1ixh6 'i'a5 ! ! 1 1 d5 ( 1 1 e5 b4 ! ) 1 l .. .iid7 1 2 �b l b4 1 3 lt.Jce2 cxd5 1 4 cxd5 i.a4 1 5 �c 1 li.Jbd7 1 6 l2Jd4 'ib6. or 9 e 5 dxe5 1 0 dxe5 'i'xd2+ 1 1 :xd2 ttJfd7 1 2 f4 g5. Of course, Bent could not have known all these variations, which were recorded in my notebook. 6 .i.eJ 8 i.dJ 8 9 10 0-0 In the present game Larsen ef fectively refuted this move. the result of good theoretical preparation. Therefore White's subsequent searchings involved IO ltc l, e.g. 1 0 . e5 1 1 d5 b4 12 ttJdl c5 13 g4 h5 1 4 t'Df2 with the better prospects (Petrosian-Gligoric, 1 972). But when Black deliberately avoids the plan employed by Larsen (as, for instance. in the game Spassky-Penrose, 1 970, where in the diagram position Black had not castled. but his bishop stood at b7), White can quietly continue 1 0 0-0, since . . . :b8 no longer has any point. . a6 The idea of playing . . . b7-b5 · at an early stage (instead of . . . e7-e5, which previously was virtually standard) was demonstrated in the 1 950s by Taimanov, Petrosian and Smyslov (the latter. for example, in Game 259, where, incidentally, for a long time he refrained from castling). c6 7 1!fd2 10 In view of Black's premature castling, here White could have played 8 0-0-0. For example, 8 . . . b5 9 .i.h6 'ila5 (9 . . . i.e6 IO h4) IO h4 b4 1 1 lDb l and then h4-h5 with an attack. However. on this occasion I did not want a sharp game, and I preferred the quiet continuation 8 .id3 . b5 �bd7 ttlge2 . ... :b8! It will soon be apparent that the black rook is best placed here. However, 1 0 . . . bxc4 1 1 il.xc4 lDb6 has also been played, or even 1 1 . . .d5. 3 92 11 cxb5 1 1 JLh6 is also possible. 11 12 b4 .! ·. · axb5 lhb6 Black is now threatening after 1 3 . . . i..e6 to seize control of c4 , and this obliges White to take urgent measures. 13 14 15 a4 l£ixa4 lha4 bxa4 l£ixa4 .i.d7 The fact that all the time Black is creating various annoying threats (now, for example, 16 . . . cS) shows that he has excellently solved his opening problems. ii'b6 16 :a5 17 18 �bl l£ic3 11a2 11aJ d5 21 22 e5 l£ia4 l£id7 .i.f5 23 24 .i.xf5 ita6 gxf5 Consistently played. Black eliminates a defender of the c4 square, and the doubling of his pawns also suits him: the black pawn at f5 will prevent a possible pawn offensive by White on the kingside. l:tfc8 'ifd8 The queen has courteously made way for the rook, and now it returns home, so as to ·remove from the agenda White' s pawn break d4-d5 (with gain of tempo). How should the resulting position be evaluated? On the one hand, White has rather more space and the open rook' s file, along which, however, nothing significant can be achieved. On the other hand, he constantly has to wony about his b4 pawn. In general, as it is customary to say, a position of dynamic equilibrium'. has been reached. 19 20 The pluses and minuses of this advance are obvious: the possible oc cupation of c4, but the fact that the bishop at g7 is shut out · of play for a long time. The latter is perhaps of more significance. 24 . . • l:tc7 Sharper was 24 . . .t:Db6 25 lDc5 (after 25 ttJxb6 Itxb6 White has no active possibilities) 25 . . . lDc4 26 'i'd3 e6. Larsen prefers to exchange knights on c5 , but this worsens Black's position. 25 l:tct Wes .*.e6 393 26 27 28 29 :a5 i.d2 f4 l£ic5 e6 .i.f8 <it>h8 37 Now Black can no longer play his knight to c4, since after 29 . . . lbb6 3 0 ltJa6 lbc4 3 1 1%xc4 dxc4 3 2 ltJxb8 (32 . . .'i'xb8 3 3 :as) he loses. But after the exchange of knights, the white c5 pawn will restrict the enemy pieces. 29 • • . 37 38 39 lDxc5 dxc5! It is probable that my opponent underestimated this move. Now Black's heavy pieces remain passive (the b-file is blocked), the c6 pawn is weak, and in an endgame the threat of b4-b5 with the creation of a passed c-pawn will be highly unpleasant. 30 31 32 33 11fd3 l:lcal lta8 1fd8 1i'd7 Itcb7 It is useful to exchange one pair of rooks, since this eliminates the potential threat of an exchange sacrifice on b4, after which all Black's pieces could have become active. 'il'c8 33 34 35 36 l:txb8 liaJ 1fa4 . • � . . 1ia8 1ia5 h6 l:tb8 Of course, it would have been tempt ing to begin an attack on the somewhat compromised position of the black king. But I operated 'a la Capablanca ' and, avoiding unclear complications, aimed for the exchange of queens, after which White will have a clear advantage in the endgame. 39 40 30 lia3 With this move White as though hints that he might switch his rook to the kingside. 1i'xb8 @g8 1i'c7 . . "iia6 • ifc8 i.e7 Larsen was probably afraid that as a result of adjournment analysis I would nevertheless give preference to the attack on the kingside. Therefore if 4 1 'i!fe2 he clearly wanted t o retain the option of replying 4 1 . . . l:ta8. As for the endgame. this evidently suited Black, but 40 . . . 'ii'c 7 was objectively stronger. 41 1i'xc8+ ltxc8 Paradoxically, in a conversation with Spassky, Larsen evaluated the endgame as follows: ·1 am probably slightly better, and if my king can get into the centre, I will look for winning chances' . 'iftf8 4 2 'itm h5 43 w Not wishing to allow g2-g4, Larsen makes an incautious move, which weakens his h-pawn and gives White new chances. White's positional advantage is obvious: he controls more space, which ensures him freedom to manoeuvre, the black c- and h-pawn�. need defending, 394 and the threat of b4-b5, creating a passed c-paw� is a very serious one. Then there would have followed 53 . . . :as or 53 .. d4 54 i.d2 l:a8. llc8 53 'it>fJ . 54 <i>e2 .*.d8 55 Aet f6 Black was ahnost in zugzwang. If 54 . . . f6 there could have followed. · 55 l:e7 fxe5 56 :xe6+. After the retreat of the bishop, this advance becomes pos sible, but the white h-pawn becomes very mobile. Therefore 54 ... <it>g7 was more circumspect. 44 <"Jre2! c&>g7 45 46 47 48 .\let %ta7 .tcJ 'iti>dJ <iti> g6 .id8 h4 h3 Black is aiming to activate his bishop or obtain another passed pawn, and in so doing to rid himself of his doubled pawns. But more significant is the fact that now on the sixth rank he will have another weak pawn at e6. fxe5 56 � Otherwise 45 l:h3. Now, to defend his h-pawn, both Black's king, and his bishop, must stay on the kingside. Larsen was probably afraid of White moving. his king to a4 and then playing b4-b5. In the event of this he wanted to gain some counterplay, and for this reason he sacrifices a paw� assuming that the doubled h-pawns will be of little value. Subsequent events, however, do not confirm Black's hopes. 49 50 gxh3 'ifi>e2 57 .i.h4 Now White avoids the manoeuvre of his king to the queenside, and tries to exploit his passed h-pawns. · 50 51 52 'it>fJ <it?g2 ; I fxe5 Ac7 In addition it transpires that the black king' s position has become less secure after the opening of the seventh rank. Black covers it, but with his lone rook he cannot block two ranks. .i.d8 .i.h4 l:td8 58 59 60 lla8 Ag8+ h4! Thus the h-pawn finally advances. It is immune: 60 . i.xh4? 6 1 l:lh8+. .. .ih6 60 . . . c:li>g6 61 llb8 62 .td2 was threatened. 62 b5+! Again this pawn cannot be touched (62 . . . 'ittxh5 63 i.d2), and as a result Black's position is worsened catastroph ically. 62 ®g7 • Larsen aims to complicate matters, provoking White into playing 53 :tc7. 3 95 .ig5 <i>h5 63 • . .:ta8 To deny the enemy rook any scope. 63 64 • • • lta6 73 74 75 .i.g� White sets his distant sights on the e6 pawn. 64 • • • ltd8 :xe8 c6 l:tc8 Black resigns The time for decisive action has finally arrived. 65 b5! This move leads to a win. Both sides acquire three passed pawns (a rare instance)� but White's are further advanced and therefore more dangerous. Also, Black's pieces are badly placed. 65 66 . • • lhe6 This ending was another one that I conducted by the Capablanca method, which was described in the notes to Game 74. This was the conclusion to the last tournament game that I won in my chess career. cxb5 .let Game 3 8 1 Spassky-Botvinnik . 67 l:.g6+ was threatened. 67 68 69 70 �b4 ltg6+ l:.d6 lid7+ Leiden 1970 d4 'ifrih7 .i.b2 <t>g8 Caro-Kann Defence After 70. . . �h6 71 .i.d2+ @xh5 72 l:d6 l::th8 73 c6 the passed c-pawn would soon become a queen, since the black rook is occupied in defending its king against mate. But now too the white pawns cannot be stopped. 71 72 e6 e7 1 2 3 4 e4 d4 lt\c3 ti)xe4 c6 d5 dxe4 i.f5 5 6 7 �g3 �fl h4 .i.g6 lDd7 b6 This sound line served me well in my return match against Smyslov in 1 958, and I retained a liking for it. .i.c3 lte8 3 96 8 Black successfully equalised in the game Sax-Hort ( 1 979): 1 1 . . .l£igf6 12 .lf4 'ifas+ 1 3 �d2 1Wb5 14 'ilxb5 cxb5 1 5 a4 bxa4 16 l:lxa4 e6 17 b3 i-e7 1 8 <it>e2 a6. b5 12 .i.f4 .i.d6 13 14 .i.xd6 ltle4 1fxd6 Bagirov recommends for Black 12 . . . 1fa5+ 1 3 :td2 'i'b6 14 0-0-0 .il.e7 1 5 lthhl lLlgf6 1 6 c4 4i'a6. Formerly this advance was regarded with some scepticism, but following Gligoric's example the idea has once again gained recognition. White pre pares a favourable pawn situation for the endgame. In the 3rd game of the afore mentioned return match White played the immediate 8 i..d3 , which theory now assesses as leading to an equal game. 8 9 10 .i.d3 11xd3 11 l:th4 Jl.h7 iaxd3 ifc7 • Regarding 10 . . . ti.Jgf6, see Game 345 (between the same opponents). Gligoric has persistently employed this move, although it was apparently first played in a game Keres-Bagirov ( 1 959). Chess fashion, however, is also short-lived (although sometimes the old unexpectedly returns), and the latest word in opening theory is considered to be the quiet move 1 1 .i.d2 followed by queenside castling. 11 • • • e6 14 . 1'b4+ An unfortunate check. 14 . . . 'i'e7 was, of course. more circwnspect. After 1 5 0-0-0 �gf6 1 6 lDxf6+ lLlxf6 Keres gained an advantage in the afore mentioned game: 17 �e5 0-0-0 18 ilg3, but that same year, in the Candidates Tournament in Yugoslavia, Petrosian improved the defence against Gligoric with 16 . . . gxf6. Therefore, when in our game from the fourth cycle (tl1is game was played in the second) I played 14 . . . 'i'e7 against Spassky, this quickly led to a draw: 1 5 0-0-0 ltlgf6 16 t'.Ded2 0--0-0 17 ltJc4 �b8 18 g3 lDb6 19 'i'e2 ltlbd5 . 3 97 • Further searches for White proceeded in the - direction of 1 5 1i'a3 (apparently suggested by Tolush), but after 1 5 . ." . ifxa3 16 bxa3 '3ie7 17 :bl ::tb8 1 8 t'Dcs ltlxc5 19 dxc5 as 20 l:a4 ctJf6 2 1 lh.a5 ltld7 Black has sufficient compensation for his slight material deficit. 15 11c3 After 1 5 c3 Black would have had to return his queen to e7, since 15 . . . "i'xb2 is dangerous in view of 16 CDd6+ and 1 7 l:b l . 15 • . . 1!fxc3+ Even now it was not too late to play 1 5 . . . We7. 16 17 bxcJ �c5! rlJe7 lhgf6 18 19 l0xb7 lheS %lhc8 too White's position would have rem ained preferable. . 22 ltlxd7! 23 d5! 23 24 l:txd5 lllxd7 After 22 . . J:!xd7 White would cahnly retain the advantage with 23 t2Jb3 . Now White's material advantage is reduced to the minimum, since the presence of the c5 pawn . devalues the doubled c-pawns, but the same c5 pawn blocks the black pieces, and the initiative again returns to the World Champion. In view of the Wlpleasant threat of 24 d6+, Black's reply is forced. exd5 Or 17 . . . b6 1 8 ltJxd7 �xd7 1 9 !Lle5+, winning a pawn, while in the event of 17 . . .!LlxcS 18 dxc5 Black again loses one of his pawns. Therefore he has to go in for a sacrifice, so as to complete the mobilisation of his forces. 1 9 lDc5 also came into consideration, blocking the c6 pawn, which would restrict Black' s rooks. 19 • • • c5 It would appear that Black has gained some initiative. 20 lha5 l:tc7 If 20 . . . cxd4, White would have advantageously replied 2 1 tDec6+. 21 0-0-0 l!ac8 With a double threat (both 22 . . . cxd4 and 22 . . . c4 ), but Spassky skilfully demonstrates the ineffectiveness of this idea. After 2 1 . . . lhiS the play would have been more complicated, although then 24 f1Jb6 Trying to shut the knight at a5 out of the game� but perhaps this should have been done bv 24 . . . c4. when after 25 . .l:e4+ � 26 :xc4 n�c4 27 liJxc4 liJr6 28 :d4 !Llxh5 Black obtains a satis factory ending. 25 l:.e5+ � During the game it was hard to decide on 25 . . . <;t>f6 26 l:.e3 c4, since the position of the black king b�omes insecure. But as a result of this cautious 398 retreat, White succeeds in bringing his unfortunate lrnight into play. 26 . lDl>3! lhd4 :he4 l:tf5 llc5 ltJd7 Black Wlderestimated this reply when he played 26 . c4. White is able to parry the action of the black rooks along the fifth rank and gain a clear advantage. · tl)xf5 g4 31 g6 f4 36 f5 36 37 ... �g7 �e6 A serious unpleasantness! For the moment the knight is immune (36. .. �xh6 37 .tlxg6 mate). lbf5 l:tc5 3 1 ctJd6 tDf6 32 .:t.xc4 :xc4 33 l2Jxc4 ltlxh5 would have let · Black off too cheaply. Now the very strong 32 tl'ld6 ttlf6 3 3 :d4 is threatened. 34 35 li)b7 The point of Black's plan! If now White plays 36 g5, then 36 . . . liJxg5 37 fxg5 l:hg5 and the knight at h6 is lost, and Black has every chance of drawing. . . 29 30 31 fxg6 li)f6 hxg6 liJxh6 Of course, not 3 3 ... 'iti>g7 . because of 34 l:e7+. c4 A difficult decision to take, since now the white knight will occupy an ideal post at d4. But Black needs to free the fifth rank, so as to activate his rooks. 27 28 29 32 33 ltJf8 lie8 Now it is hard for Black to find a useful continuation (of course, 37 . ttlli.7 is not possible because of 3 8 :gs+ <t>xll6 ? 39 :xg6 mate). In the time scramble he decides on a move that offers White a choice a good practical device. 37 l:a5 .. - The last tactical chance. White's knight has successfully escaped from the unfortWlate square as on the queen side, but perhaps it will be possible to exploit its poor position at h6? It is for this that Black sacrifices a pawn. 38 ci>d2 Unexpectedly, the 'device' works. This move would seem to throw away 399 the wi� since in the rook ending White is left with a weak pawn at fS . During the game I thought that White would win most easily by 3 8 'ittb2, and if 3 8 . . . l:tb5+ 3 9 @a3 lt.a5+ (otherwise Black ends up in zugzwang), then 40 'lt>b4 l:txa2 4 1 g5 with a material and a positional advantage. Also dangerous for Black was 3 8 f6+ �xf6 3 9 l:xf8+ @g7 40 l:c8 �6 41 l:xc4 l:xa2 42 l:te4 ! with a highly favourable rook ending. In Spassky's opinio� 3 8 fxg6 ltlxg6 39 li)f5+ �6 40 :e4 was also quite good for White. 38 • • • <iii>xh6 Now, when the a2 pawn will be won, Black himself goes into the rook ending. � g7 39 1'xf8 40 41 42 l:tc8 gxf5 <t>e3 gxf5 :xa2 After 42 l:txc4 l:ta5 43 ltf4 �f6 Black has every chance of drawing the game. 42 43 44 45 �e4 ltc6+ ltxc4 @16 ltxc2 52 .. 53 54 55 56 57 58 59 60 a5 46 47 48 l:tc7+ l:lc6+ <iii>d 4 <it>f6 49 50 51 52 <i>e5 <i>d5 llc7+ lta7 l:le2+ l:lf2 <i>f6 @f7 :n The main thing for the weaker side is the activity of his rook. c4 c5 c6 @cs a4 a3 l:ld2+ :c2+ It is good to drive the king to c7, and only then advance the pawn to the second rank. <ilf7 Black easily gains a draw: he blocks the f�pawn with his king, and his passed a-pawn counterbalances the enemy c pawn. :.a2 It is simplest for Black to retain his own passed paw� and not give White any practical chances involving 52 . l:xf5+ 53 \t?d6 and then 54 c4 . lld2+ a2 l:lc2 lld2+ <it>d6 <it>c7 lta5 <i>d7 It is not clear what White was hoping for. After all, it is impossible to use the rook simultaneously to defend the king against checks, and to control the a l square. 400 61 62 63 64 65 66 llc2 <it?c8 lld2 c7 l:tb2+ <iii>b 7 l:r.c2+ <i>c6 IZ.d2+ <i>d7 lle2+ <it?e8 Draw agreed . Training Games Usually I asked Ragozin to stick to the latest word of theory. Of course, I never · showed him before a game the variation I had prepared, in order to test the effect of surprise. However, my friend possessed a very obstinate chess character, and sometimes he would avoid the theoretical recommendations, which threw me into despair . . . In the present game, thank God, it all turned out well. Ragozin-Botvin�ik Zacherenye Sanatorium 1936 French Defence 1 2 3 4 5 e4 d4 tLlcJ a3 bxc3 e6 d5 .i.b4 .i.xc3+ dxe4 It was not by accident that this open ing variation was tested in a training game. In late 1 935, by employing this variation, Alekhine with White won against Euwe in the 3rd game of their World Championship Match. It was natural to exlJect that this continuation might occur in tournament play. At that time I was already systematically playing the French Defence, and so Ragozin and I decided to test this position. Understandably, it was I who defended Black's �honour'. 6 7 8 1ig4 1ixg7 1!fh6 tLlf6 %!g8 c5 9 This position was subsequently reached in many of my games (cf. , for example, Nos.228 and 229). 9 ttle2 This is what Alekhine played against Euwe; this move, accordjng to the state of theory in 193 6 � was considered the strongest. • . • tLlc6 As was shown by my 1 954 match with Smyslov, 9 . . . l:tg6 is preferable, for- . cing White to determine in advance the position of his queen. When we played this training game, this was not yet known. 10 .i.b2 The correct reply to 9 . .. tt:Jc6 - 10 dxc5, was found by Smyslov 18 years 40 1 later, during which ti.me I e,.nployed this variation several times in tournaments and matches. It is hard, therefore, to reproach Ragozin for making a weaker move, which has the aim of defending the central pawn. 1 O ..tg5 leads to an equal game. 10 . . • .i.d7 Black prepares queenside castling. It only remains for him to play 1 1 . . . llg6, to free his queen from having to defend the knight at f6, and his queen will be able to leave the eighth rank. 11 :dl This came as a complete surprise to me. 1 1 h3 was probably better. 11 12 ... 1fe3 %1.g6 1ia5 Preventing 1 3 ttJf4 in view of the reply 1 3 . . . cxd4. White is forced to take the c5 pawn - a possibility that he rejected on the 10th move. 13 dxc5 0-0-0 first sight seems illogical. However, the c5 pawn is cramping'. �e bl3;ck queen; in · addition, White's f2: n.ow comes under attack, and the play becomes forcing. 15 16 'ifxe4 ltd2 1fxc5 9b6 17 18 .i.cl ifc4 f5 'ifbl There is no way of defending the bishop at b2, and White has to make a non-aesthetic move. During the game this seemed an energetic move to me, since White can defend his bishop only with his knight or king. In reality, there is nothing for the queen to do at b l . 1 8 . . . e5 or 18 . . . ct>b8 should have been played, similar to that which later occurred in the game. 19 lhe2 11fb6 It is one thing to make a mistake, but another to persist with it (White was now intending to play 20 'iib3, and he might have been able to save himself in the endgame). 20 21 lhf4 .i.dJ llgg8 If 2 1 �e6 there would, of course, have followed 2 1 . . . l:lge8. 21 22 <i&>b8 ifb3 White is forced to move his queen from c4, since after 22 0--0 ltJge5 and 23 . ttJf3+ he would have lost the exchange. .. 22 23 . :e2 • . 1fc7 e5 Black is a pawn down, but he has a considerable lead in development. In view of the unavoidable . . . e5-e4, White's position becomes hopeless. Black exchanges his central e-pawn for one of the tripled c-pawns, which at 24 /l)d5 is no better on account of 24 . . . fi'd6 with the.t:lueat of25 . . . ..te6. 14 lhg3 lhg4 402 24 lhe6 Jl.xe6 e4 lllce5 24 25 ·1i'xe6 26 : ·ii.b5 27 c4 27 . ifxc3+ was threatened, but this .. 27 • • • tbxh2 . 28 lhh2 White resigned in view of 28 . . . lim+ 29 ®fl (29 gxf3 :g 1 mate) 29 . . . l:dl+ 30 :e l l:xel mate. Ragozin-Botvinnik Pushkin 1 941 French Defence 1 2 3 e4 d4 �d2�: �· lilc6 exd5 6 ... cxd4 7 0-0 8 ltlbJ Strictly speaking, the idea of this move belongs to Capablanca, but in the afore-mentioned game he first gave a queen check at e7, which led to difficulties for Black. Now, however, in the event of 7 lDxd4 Ve7+ (Averbakh Botvinnik, Moscow Championship 1943/44) it is not advantageous for White to retreat 8 .i.e2, as played by Keres. Black achieves development. a Ji.d6 comfortable 8 �xd4 can be met by 8 .i.xh2+ 9 �xh2 Vb.4+ 10 ®gl 'i'xd4 with equal chances (Geller-Ivkov, 1 969). . . . This insidious move leads to mate (28 . . �ef3+ is threatened)� there is no longer any defence. · lilgf3 exd5 .i.b5 In the late 1 93 0 s I became interested in the Tarrasch Variation (3 liJd2), especially after the well-lmown game Keres-Capablanca ( 1 938), which was won by White. In this context I decided to check a new idea. A familiar picture. All Black's pieces are coordinated, whereas White' s are scattered, and although at first sight the white king in the centre feels safe, its fate is sealed. too does not save White. 4 5 6 e6 d5 c5 403 8 . • • lllge7 9 liJbxd4 The Encyclopaedia considers this to be the strongest continuation, but to me it seems that 9 1Lxc6+ comes into consideration. Then 9 . . . l2Jxc6 10 :e 1 + li..e7 1 1 t2Jfxd4 0-0 12 .i.f4 l'bxd4 13 'i'xd4 leads t o a position favourable for White (Averbakh-Botvinnik, 1 9th USSR Championship, 1 95 1 ), while 9 . . . bxc6 10 'i'xd4 0-0 1 1 .i.f4 lhf5 1 2 'i'd2 'i'b6 1 3 l2Jrd4 l:td8, in the opinion of Matanovic, leads to an unclear position. 9 . • • h6 It is useful to prevent .i.g5-h4-g3 . 10 11 12 .i.e3 'ifd2 i.e2 0-0 .i.g4 Black ignored the bishop sortie to b5, and this has born fruit. The bishop has abandoned this post, and White has therefore simply lost a tempo. It was more logical to delay this manoeuvre a little and to first play 12 l:fe l , when the bishop could have retreated to fl . 12 :cS 13 14 15 J:adl h3 lhxc6 .i.b8 .i.h5 White falls into the positional trap, prepared by Black's 1 3th move. Before the exchange on c6, . . . 1li'd6 was not a threat, since White could have answered it with ltJb5; now, however, this threat becomes a real one. Ragozin apparently assumed that it was not possible to play 1 5 . . . bxc6 in view of the weakening of the a6 square, but this proved to be a tactical oversight. 15 • . . bxc6 16 g4 A decisive weakening of the castled position. For the moment there was nothing tenible threatening White, and he could have calmly continued, for example, 1 6 �c5 . 16 17 .ia6 .i.g6 'ifd6 18 19 lUel 11e2 i.e4 l£ig6 20 <it>fl %ke8 21 22 23 24 lDgl gxf5 irg4 f5 lhh4 lhxf5 . • • A simple refutation. White does not have time to take the rook in view of 1 8 . . . .te4. And now 20 i.xc8 is not possible because of 20 . . . ltJh4. So, White has not gained any material advantage, and his king cannot be defended. f3 White resigns There could have followed 24 . . . .i.xf3 25 t'Dxf3 ltJxe3+, if there is nothing better. 404 9 Botvinnik-Ragozin 10 11 12 Sosny ;$anqrorium 1945 Slav �fence 1 2 3 4 5 6 7 d4 c4 lLJtJ cxd5 ll:\c3 .tf4 eJ 'i'bJ gxf3 l:lcl l2Ja4 i.xf3 e6 l:tc8 Since his kingside castled position is already compromised, White decides to leave his king in the centre, in order to gain time for the development of his initiative. d;; c6 ll:\f6 cxd5 l2Jc6 i.f5 'ifb6 12 • • • 1fa5+ The retreat of the queen to d8 looks too passive. 13 14 <ili?e2 .i.d3 .*.b4 With his king at e2, White cannot, of course, part with this bishop. And in addition, now it Wlexpectedly transpires that it is hard for Black to defend his b7 pawn. 14 • . • 0-0 Black overlooks one tactical nuance. 14 .. . ttJh5 was better. 15 16 a3 11xb7 17 il.b5 fj,,e7 11xa4 Ragozin usually used to sense inI was intending to employ a new move, and he always a i med to forestall this! Here in reply to 7 . . . e6 I was planning to test the variation i nvolving 8 .i.b5. as occurred 16 years l ater in my games with Tai (No.290) and then with Pomar (No.340). Even so, the planned move is also possible in the resulting situation. 1 uitively when 8 .ib5 .i.g4 A simple way of preventing the attack on c6 by 9 ll:\e5, but this leads to the loss of a tempo and it gives White two good bishops. 'Il>:e Encyclopaedia considers the lesser evil to be 8 ... e6, but it is hard to agree with this. This is the whole point If 17 . . . 1ib3 there follows 18 i.xc6, and the b2 pawn is defended. Black prefers to give up the 405 exchange, hoping that the open position of the white king will give him some · counterplay. 17 18 19 20 exd4 lbc8 J:r.hct li)xd4+ 1fxd4 40 1'xf4 The entire variation was forced for both sides. Unfortunately for Black, he is obliged to waste a tempo moving his bishop. 20 • • • .ld8 This leads to an endgame, in which White wins easily. However, after 20 . . . .i.d6 2 1 l:txf8+ li.xf8 22 l:tc8 the threat of 23 1i'b8 would have quickly decided the outcome. 21 22 23 24 1fb8 'it'xf4 'ifi>f1 b4 White first invades with his rook into the enemy position, and then plays his bishop to c6; after a different move order the activity of his rook would have been restricted. 27 28 29 30 31 a4 llxf8+ rl;e7 ii.c7 lDxb4 <i>d6 Forcing Black to block the c6 square, where his knight might have been advantageously placed. 31 32 33 34 35 lLlb5 t2Jxf4+ .i.b6 1ha7 :xn l:txg7 'ifi>e2 <i>xc6 <i>b5 'ifi>xa5 lDd5 Since Black has been unable to make use of his passed pawn, it means that the endgame is hopeless for him. White's main trump is his extra pawn on the queenside. 24 25 26 a5 Ji.c6 l:xc7+ lid7+ 35 36 37 38 39 40 d4 lt)<15 'ifi>xf8 l£if4+ <li>b4 <ifi>d2 llxh7 '1t>c4 '1t>d5 b4 h5 cJ.ie5 llh8 Black resigns Botvinnik-Ragozin Bolshevo Sanatorium 1947 Queen's Gambit 1 2 3 4 5 6 27 l:tc8+ d4 ltlfJ c4 ti)c3 cxd5 e3 · d5 ltlf6 e6 c5 ltlxd5 cxd4 Before the Chigorin Memorial Tournament we decided to check once 406 again the position that occurred in the game Botvinnik-Alekhine (No.94) after White· s 1 0th move .. Although instead of this last mQ�e the World Champion played differently (6 . . . l'Dc6 ), this merely leads to a µansposition of moves. 7 8 9 exd4 i..c4 0-0 10 �et and Black's bishop at c8), where White played i.b5, but Black gained sufficient counterplay. ltlc6 �e7 0--0 In Grune 1 33 Black played 9 . . . a6. 10 �e3 (No. 1 32) is weaker. Against Szabo (Groningen 1946) I carried out the same plan as in the present game, but without 10 J:te 1 . Here Alekhine played 1 0 . . .b6 and after 1 1 ltJxd5 he ended up in a difficult position. All the commentators thought that Black was obliged to exchange first on c3 � it is this advice that Ragozin follows. 10 11 12 bxc3 Ad3 12 13 ... 1!'c2 ltlxc3 b6 There is no longer anything for the bishop to do at c4. 15 16 i.b7 A manoeuvre that was employed back in the game Chekhover-Levenfish ( 1 934 ). If 1 3 . . . h6 White advantageously plays 1 4 1ie2, threatening 1We4. 13 14 . • . J.h6 • • 1i'd2 • • • l:tc8 .hlabl ! A necessary link in White· s plan. Now a possible attack on g2 ( .. ."i'd5) can always be parried by l:b5 . 16 ... 17 h4 .i.f6 This proves to be a loss of time; 16 . . . i.f8 was simpler. g6 The pawn sacrifice 14 h4 .txh4 leads to unclear consequences (Larsen Unzicker, 1 968). 14 lle8 The same position was reached in the · afore-mentioned game with Szabo (only, there Whit�' s rook was still at fl , • 15 An important link in White 's plan. It is useful to move the queen off the c file, and also to take control of the g5 and h6 squares, preparing the advance of the h-pawn. For the moment Black leaves his queen at d8, hoping to prevent h2-h4. It transpires that after l 7 . i.xh4 18 tLlxM irxh4 1 9 i.g5 'i'g4 (19 . . 'i'bs 20 .ie2) 20 lle4 'i'f5 21 l:te5 1i'g4 22 .ie2 Black has to give up his queen for rook and knight. 407 . . . · 17 18 1fd6 .*.f4 Preventing the opening· up of the position by 18 . e5, when there would follow 1 9 .i.g3, and it is hard for Black to escape from the pin. . . 18 1ia3 18 . 'i':f8 was more cautious; the .. counterattack against c3 comes too late. 19 20 h5 .i.e5 lDa5 1ie7 The queen has to return, since after 20 . . . .i.xe5 2 1 tDxe5 'i'xc3 22 iff4 it would be hard for Black to defend. 21 22 23 � .i.xf6 tLlg4 9xf6 l:ted8 'Wg7 24 25 26 hxg6 ..g5 l:b5 hxg6 <ii>f8 lDc6 There is nothing better. If 23 . . . ilh4 24 'iff4, threatening to win the queen. 29 30 31 ttlxg8 1fe7 llbe5 �xg8 i.c6 l1d7 32 33 34 1ih4 '8'f4 �gJ ifh8 1lg7 34 35 %ieg5 White' s queen sortie would not appear to have achieved anything, since now it is forced to leave the enemy position. However, this is not so: Black has been forced temporarily to block the c-file. The storm clouds are again gathering over the position of the black king. To be fair, I should mention that it was not so difficult for White to find the correct plan here: all he needed to do was copy Lasker's play in his game with Capa blanca (Moscow 1 935). ..td5 There is no defence against the threat of 36 ..i.xg6. 35 l:Xc3 A vain attempt to parry the bishop sacrifice. 36 .i.xg6 37 .i.xf7+! l:lxg3 In a difficult position Ragozin de fends coolly; the knight manoeuvre to g8 saves Black from the immediate threats. 27 28 lDf6 :eJ Intending 29 l:lh3 . 28 lDe7 lhg8 408 .; The subtle point is that after 37 :xf7 3� 'i')fg3 (but not 38 .&txg7+ l:.gxg7) Biack does not have suffi cient compensation for the queen. . . . 37 38 39 [ � li.xg7 l%.xg7 i.xe6+ Black resigns Ragozin-Botvinnik Nikolina gora 1951 efl_e_ nce���� ch_D Fr_ _ en_ � ��� __ 1 2 3 4 5 6 e4 d4 tLlcJ e5 aJ 'i'g4 e6 d5 .i.b4 c5 ii.as 6 b4 is a popular continuation, as, for example, in the 9th game of the Smyslov-Botvinnik World Champion ship Match (1954). 6 i.d2 is also quite good for White. 6 . • • lj_)e7 were especially necessary - after all, for three years I had not taken part in any competitions. Again I had to turn to the French Defence, since during that time new ideas had appeared in practice. In my game with Reshevsky from the World Championship Match-Tourna ment (No. 1 81) after 5 ... il.xc3+ 6 bxc3 'ilc7 7 1i'g4 f5 8 1i'g3 cxd4 9 cxd4 I ended up in a difficult position. There fore I decided to test a continuation from an earlier game with the same opponent (No . 1 66). 7 1fxg7 It is well known that 7 dxc5, as Reshevsky played against me, is stronger, arter which White retains an opening advantage without any compli cations. But on this occasion too Ragozin went his own way. 7 l:tg8 8 9 'iixb7 b4 cxd4 dxc3 Black could have considered 9 . . . i.c7 10 liJb5 a6! (but not 10 . i.xe5 1 1 ttJf3 :i.h8 12 'i'd3 .i.f6 1 3 i.f4 ttJg6 14 l'iJc7+ <t>f8 1 5 i..g3 ttJc6 16 l'iJxa8 e5 17 'i'b5, which does not give him an equal game) 1 1 ltJxd4 �xe5 12 ti)gf3 1'c7, when for the sacrificed pawn he has some initiative. .. 10 bxa5 11 tl)fJ /i)d7 10 . . .liJbc6 is worse on account of 1 1 tt:Jf3 (with the threat of 1 2 tiJg5) l l . . .Wfc7 12 .llb5. In 195 1, befqft�.t We match for the World Champio�r,.jp, ;training games Another plan is to make rapid use of the passed h-pawn: 1 1 f4 'i'xa5 12 tlJf3 liJf8 1 3 'i'd3 .i.d7 14 tl'lg5 :cs 15 I!b 1 b6 16 h4 (Gligoric-Dµckstein, 1 955). 409 11 • • . · .1fc7 The exchange of the light-square bishops is a · positional mistake - the black rooks gain freedom of movement. In the 9th game of the World Championship Match against Smyslov ( 1 954) I played 1 1 . . . tt:illJ , which is far weaker, since instead of making a further attack on the central e5 pawn Black removes the only threat to it. I have to admit that three years later (during the game with Smyslov) I quite simply forgot what I had played against Ragozin - after all, I was in my 43rd year, and at such an age a chess player' s memory i s no longer faultless. The move played has the advantage of forcing White to take urgent measures to defend his e5 pawn; more over, 1 2 i.g5 is not possible on account of 1 2 . . . l:txg5 1 3 tLlxg5 'i'xe5+. 12 13 14 15 16 17 13 . • .i.b5+ g3 tt'lfg6 hJ l:lc4 lt:\gxe5 tt'lxe5 0-0-0 .i.eJ ltJc6 It transpires that, as a result of the unjustified exchange of bishops, White is unable to defend his central pawn. .i.f4 • .i.d7 'tl'xd7 :g4! 15 . . . J::txg2 was dubious in view of 16 .i.g3 . Now, however, White is forced to weaken his position by g2-g3 ; 16 Jlg3 is met by the unpleasant 16 . . . l:e4+, while after 16 i..e 3 the reply 16 . . . :xg2 is now possible. 18 19 20 21 22 Subsequently Euwe recommended this way for White to retain the initiative: 12 i.b5 a6 (if 12 ... J%xg2, then 1 3 @fl and 14 :lgl) 1 3 .i.xd7+ i.xd7 14 0-0 d4 (14 . . . 'ii'c4 1 5 ttlg5) 15 ..i.g5 .i.c6 16 JJ..xe7 @xe7 17 'i'h.4+ <it>e8 1 8 tLlg5 11fxe5 1 9 f4. 12 i.. xd7+ 1id3 0-0 lt:\xe5 1i'h7 .i.xa7 tt'lf8 White has regained his pawn, but at a high price: his bishop is shut out of the game for a long time, and the initiative completely passes to Black. White's main chance is to create a far-advanced passed pawn. 410 22 23 24 25 lDtJ+ �g2 llfel lDd2 d4 f3 Black was threatening .2 5 . . . fid5+ or 25 . . . 'ifc6+. · 25 26 27 28 29 1fxd7+ l:te5 l:tael l:tb5 f5 �xd7 <be7 :c6 After 29 .:I.xf5 d3 30 cxd3 c2 White woulq have lost immediately, because his rook would be unable to leave the e file (3 1 J:c5 tDb3 32 �xc6 bxc6 33 i.e3 llxd3 , or 3 1 i.c5+ 'it>d7 32 llf7+ 'it>c8). 29 30 . . d3 • l:tcl White would also not have been saved by 30 l::t xb7+ �6 3 1 cxd3 (or 3 1 J:!c l ttJc4, transposing into a position from the game) 3 1 . . .c2 32 .:I.c l l:txd3 33 :b2 ctJxf3 . 30 31 32 33 ltxb7+ cxd3 .?!bJ ltJc4 �6 lixd3 If 3 3 .if2 there would have followed 33 . . . c2, while if 33 :Ixc3 :xc3 34 i.d4+ e5 . 3 5 i.xc3 l£le3+ and 36 . . J:hc3. 33 34 . J.f2 • lld2+ c2 . A mistake, which could have led to a draw. Black should have played 34 . . . ctJb2 with nwnerous threats (. . . ltJd l , . . . lD<l 3 and . . J�xf2+). 35 a6 35 36 37 llc3 f4 If 35 l:.c3 , then 35 . . . tDxas. � llxa6 lDxaJ Adl 38 :txc2 41 In time trouble White fails to find 38 i..e3 , after which Black's pieces are tied to defending one another, and no way of strengthening the position is apparent. But with the advantage of the exchange Black has some winning chances, despite the fact that all the pawns are on one wing. 38 39 40 41 �xc2 l:la4 llc4 lld3 Now Black doubles rooks on the third rank, and the enemy pieces will be condemned to passivity. :ccJ 42 <ifi>f2 43 44 45 lle2 'it>f3 �e7 ci>d6 g4 It would have been better to refrain from this move, which weakens the f4 pawn. Now the black king must immediately return to the kingside, in order to succeed in reaching g6 (in the event of the further advance of the white pawns). 45 411 llxc2 :b2 .i.eJ • . • "1e7 46 . 47 48 49 · · ltel :e2 Itel l:te2 <i;f7 Botvinnik-Bronstein :bJ Voronovo 1952 Queen' s Gambit lidc3 lic4 1 2 3 d4 c4 cxd5 d5 e6 A premature exchange, after which Black quickly develops his queenside pieces and gains equality. Black has arranged his pieces in the most advantageous way and is threat ening by 50 . . . e5 5 1 fxe5 f4 to win a piece. White decides to parry the threat by exchanging pawns� but after this he no longer has any chance of saving the game. 50 51 52 53 54 gxf5 lta2 lle2 Itel h4 exf5 lte4 'iti>g6 �h5 Preventing the further advance of the black king (54 . . . <ittxh4 55 l:thl mate), but now the h-pawn becomes a target. 54 55 56 57 l:lgl l:tel Ahl 57 �. . l:a4 .:aa3 lta2 Black was threatening ... l:a2-h2xh4. Zugzwang. ll'lc3 .i.f4 exd5 c6 il.f5 6 e3 ltld7 This is the whole point! If this bishop reaches f5, it is obvious that White has played this variation of the Queen's Gambit inaccurately, having failed to develop his king's bishop at d3 in time. In view of what follows in the game, it would evidently have been better to play 6 . . . CDf6 or even 6 . . . .i.d6. 7 g4 Only in this way is it possible to complicate the play. Since 7 . .. .i.g6 8 h4 does not look very attractive for Black, his bishop has to retreat to e6, and the move .i.d3 becomes possible. As has already been mentioned, I carried out a similar idea (g2-g4) back in 1934 in a game with Alatortsev (No.59), then 29 years later in several games from my match with Petrosian (for example, No. 3 1 3), and, finally, against Spassky in 1 970 in Leiden (No.379). 7 .i.e6 • 8 llc2 58 ....:ltgt l:h2 . White resigns 3 4 5 • • .i.d3 g5 Riskily played. Black obviously thought that White would be tied to the defence of the g4 pawn. However, this 412 pawn is sacrificed for the sake of seizing the initiative. 8 . . . ile7 was preferable, when ;:iµ comparison with Game 3 79 White has made the not very useful move 8 i.d3 , but perhaps 8 . . . t'Db6 followed by . . . i..d6 was stronger. . 9 i.gJ 1lfb6 White also has the initiative after 9 . . . ctJgf6 1 O h4. Correct was 9 . . h5 1 O gxh5 i!Llgf6 with the intention of taking the h5 pawn with the knight. 10 'ifc2 .i.xg4 If 10 . . . ttlgf6 there would have followed the same move as in the game. . 11 12 13 h4 llxh4 ltlge2 gxh4 thgf6 A somewhat unexpected move. Of course, White must first of all prepare queenside castling, in order to connect his rooks. Therefore 1 3 f3 suggested itself, but I did not want to weaken my · pawn chain. 13 14 15 16 0--0-0 �M l:lh2 h5 .i.h6 i.g5 .i.f3 So, Black i s a pawn up, but his king's position is insecure, and, as will be seen from what follows, queenside castling does not ease his task. In this con nection, White's natural plan is to open up the position with e3-e4. Now this would lead to simplification, so he does not hurry, but first plays his rook to the e-file. 17 !tel 1fa5 17 . . . 0--0--0 seemed dangerous, for example, because of 18 ttlgl ( 1 8 tiJb5 l£ie8) 1 8 . . . i.g4 1 9 f3 .ie6 20 i!tJh3 i.h6 2 1 e4. For the moment Black pins the knight at c3 , and thereby forestalls e3-e4. 18 19 .i.e4 In this way White defends his rook at e 1 and forces through the advance of his e-pawn. However, given correct play by his opponent he would have had to abandon the idea of a coordinated advance of his pawns in the centre. Ther�fore 1 9 f3 was more prudent. 19 20 21 413 ltlgl f4 ifxd3 e4 .lxd3 il.h6 34 0-0--0 21 The losing move. After 2 1 . . .dxe4 22 l'Dxe4 �f8! Black would have . main tained a defensible position. Now, however, the e- and f-pawns advance very strongly. . 22 23 24 • . e5 f5 e6 lDe8 .i.g5 Again opening the diagonal of the dark-square bishop. It only remains to include the g 1 knight in the attack. 24 25 26 27 iDfl lDe5 J..h4 .i.xd8 iDxf7 exf7 35 llgl 1i'h3 li:)xe4 • • 1!f d7 35 :gS JJ..e7 36 37 f6 '9b.3 i.f8 37 38 39 40 41 e3 <i&>c2 �c7 11 xd7+ <;i?xd7 llhh8 e2 <it>d2 Black resigns 3 5 . . . 'i'xd4 was not possible on ac count of the prosaic reply 36 f8'i', when all the critical squares (b2, d l and gl) are defended, and after 36 . . . 'i'd3+ 37 l:lc2 Wfl + 3 8 %:tc 1 the counterattack comes to an end. The simplest. is �fe4 l:xd8 iDxf7 White is already the exchange up, but this is not the end of his gains. 30 31 32 33 • A last attempt to create at least some kind of counterplay. lDdf6 .*.h6 ti)d6 Loss of material for Black inevitable, e.g. 27 . . . J..g7 28 .:tg l . 27 28 29 30 1fxh5 J.g7 11fc7 Ji.f6 . dxe4 Kan-Botvinnik Nikolina gora 1952 Slav Defence 1 2 4 14 d4 c4 d5 c6 3 ltlf3 4 lLlcJ 5 eJ 6 · · 11c2 lhf6 e6 lLlbd7 6 .*.d6 In 1 943 Makogonov played 6 tLle5 against me {No. 1 27). • • • In those years I often played the Slav Defence, and therefore I had to test the continuation that occurred in the game Najdorf-Kotov (1 950). There White played 7 i.e2 followed by b2-b3, and the resulting variation was judged to be in his favour. I, however, wanted to test a new plan for Black. 7 b3 I should remind the reader that 7 .itd2 can be found in No.2 16. where it was also pointed out that after 7 e4 dxe4 8 ltixe4 thxe4 9 'ifxe4 e5 the game is equal. 7 8 .. il.e2 . 0-0 1fe7 Playing against Reshevsky in 1 95 1, Euwe tried to simplify the game inunediately by 8 . . e5, but after 9 cxd5 iDxd5 1 O .ib2 White retained some advantage. . 9 0-0 dxc4 At that time I liked this variation, but later I came to the conclusion that it is more useful for Black to play 9 . . .b6. From the routine positional view point this exchange seems dubious: the central d5 pawn is exchanged for the flank b3 pawn. But with the help of analysis and its testing in a training game, a better judgement of the plan can be made. 10 11 bxc4 .i.b2 e5 :eS 12 :aet Soon it became lrnown that 12 l:tfe l is preferable, vacating the fl square for the (subsequent) manoeuvre M-d2-fl . 12 • . • .i.c7! It is this that constitutes Black's new idea. In the Najdorf-Kotov game· there followed 12 . . . e4 1 3 ltJci2 ltJf8 1 4 f3 exf3 1 5 .txf3 with the better game for White. But after the text move there is the possible threat of . . . 'iid6. Another advantage of 12 . . . .i.c7 is that the advance e3-e4-e5 will not involve an attack on Black's king's bishop. White's next move takes control of the d6 square. 13 14 15 16 c5 tLld2 f3 i.xf3 e4 tt)f8 exf3 tLlg4 17 18 .i.xg4 e4 .i.xg4 llad8 An essential move, giving Black the advantage of the two bishops. If at the time White had played 1 2 I;lfel (instead of 1 2 ltae l ), he would now have been able to reply 1 7 lDfl . 415 Bringing the queen's rook into play with gain of tempo� the advantage of 1 2 . . . il.c7 becomes obvious. 19 lLlb3 f6 20 21 h3 'iff2 ..t.h5 b6 pieces are unable to come to the aid of their king. Preventing e4-e5 and tbe4-d6. 30 ttlct Also after 30 'i'd2 Black would have won by 30 . . . 'i'g6 (3 1 'it?h2 tbd3 ). This attack on White s pawn chain puts him in a critical position. With a pawn sacrifice he tries to transpose into a position with opposite-colour bishops, but the play remains complicated, and apart from his extra pawn Black also gains an attack. 22 23 24 25 26 e5 ll:\e4 ll:\d6 cxd6 l:txe5 30 31 11'e3 "'xfl �gl • • 'ii'g6 1ib3+ A typical ' spite' check. 31 32 33 fxe5 ll:\e6 ii.xd6 11'xd6 �h8 g4 .i.xg4 lLlxh3+ 1!f g3 White resigns Botvinnik-Averbakh Nikolina gora 1956 Nimzo-Indian Defence 26 dxe5 would have inunediately given Black the d-file, but now he seizes the f-file. 26 27 28 29 • l:lf8 l:xfl+ l:tf8+ lLlf4 1 2 3 4 5 d4 c4 lDc3 eJ .i.d3 ttlf6 e6 -*.b4 0-0 c5 6 7 8 a3 bxc3 e4 .i.xc3+ ttlc6 For 5 . . . d5 see Game 346 It transpires that White's kingside cannot be defended: his rook and minor 4 16 The usual variation. . in which, lDxd6 17 exd6 'ii'c6 1 8 tt)f3 would have led to unclear complications) 1 3 .ilb4 d6 14 Axh7+ <it?xh7 1 5 ilxd4 a5 16 .itc3 f6 17 1fh4+ @g8 and Black won quickly. Of course, 1 6 .i.xd6 lhxd6 17 exd6 would have been better for White, but Black has positional compensation for his material deficit. White's plan in the present game consists in the rapid mobilisation of his forces and an attack on Black's king side, not a premature attac� but one that is well prepared. according to theory, White can aspire to retain an opening advantage, is 8 'De2. However, I had long been interested in this pawn sacrifice (8 e4 ), and I was intending to try it in my forthcoming match with Smyslov. However, in the Alekhine Memorial Tournament ( 1956) I was forestalled by Szabo; he employed this idea and in fact against Smyslov. 8 9 • • • cxd4 cxd4 tLlxd4 There is little justification in declin ing the pawn sacrifice. In the game Aronson-Aizenstadt ( 1961) after 9 . .. d5 l 0 cxd5 exd5 1 1 e5 ttJe4 White could have consolidated his advantage by playing 1 2 ttJe2. 10 e5 'ifa5+ 11 <Ml ltle8 In the Encyclopaedia I pointed out that 1 0 . . . ttJe8 came into con.sideration. Black has no other reply� after 1 1 . . . 'ifxe5 12 .tb2 he loses a piece in view of the double threat - 1 3 1Lxd4 'i'xd4 14 .i.xh7+ or 1 3 ctJf3 . 12 13 .i.b2 tLlfJ ltlc6 f5 In this way it would appear that Black has radically forestalled the opponent' s attack, but in reality it trans pires that it is easier for White to open the g-file. Black must evidently seek other ways of defending. 14 1fc2 In the first instance it is necessary to keep control of the central square e5 . 14 15 16 liel lhxe5 16 17 18 19 h3 g4 lhxc6 d6 dxe5 The 'work quota' has even been over-fulfilled: not only has the attack on the e5 square been parried, but White's queen's bishop has also been activated, which increases the importance of g2-g4 (the g7 square!). The Szabo-Smyslov game continued 1 2 .td2 ii'd8 ( 1 2 . . . 'i'c7 1 3 .i.b4 d6 1 4 .txh7+ �xh7 15 'i'xd4 a5 16 i.xd6 · ·:. lhf6 1ic5 tLle4 1fxc6 White can now regain the pawn, but the weakness of the g7 square is so obvious that another, more effective solution suggests itself. 4 17 4 5 6 7 ltlxd4 lhcJ .i.g5 1id2 tLlf6 d6 e6 a6 8 0-0-0 h6 Here I also employed 7 . ..h6 on many occasions, the last time in Game 347. 20 11.gl! This entire plan was not hard for me to find - it was developed 25 years earlier and carried out in a similar situation in the game Botvimrik Myasoedov (No. 3 1 ). 20 21 %1.eJ I prepared this variation for my match with Smyslov, expecting the · reply 9 .i.f4. Unfortunately, my prepar ations proved to be in vain, since in the 2nd and 4th games of the match Smyslov retreated his bishop to e3 - a move which I had not even considered. 9 .i.f4 10 Ji.g3 .i.d7 lati The decisive doubling of rooks on the g-file is prepared. The reader will no doubt already have noticed how well placed the white king is at fl , even though deprived of the right to castle. 21 22 gxf5 23 .i.xe4 24 'ifcJ Black resigns. If 1ic5 exf5 fxe4 24 . . . 'i!le7 24 . . 'iff8 there follows 25 :eg3. . Averbakh-Botvinnik Nikolina gora 1957 Sicilian Defence 1 2 3 e4 lLlfJ d4 c5 ltlc6 cxd4 or Later it became known that 10 tbxc6 .i.xc6 1 1 f3 'ifb6 1 2 .i.c4 secures White an advantage. After the move in the game Black obtains time for the completion of his development. 10 i..e7 At present the Encyclopaedia recommends first playing 10 . J�c8. . 11 12 .i.e2 f4 0-0 ltlxd4 White' s pawn offensive on the kingside can only be countered by a 418 natural plan of counterplay, involving pressure on the central e4 pawn. For this it is necessary to exchange knights. 13 14 1ixd4 .i.f3 Jl.c6 'ii a5 After this move White gains the opPortunity to move his bishop from g3 with gain of tempo and clear the way for the advance of his g-pawn. But Black had no choice, since with the bishop at g3 it would have been dangerous to develop the queen at c7. 15 16 -*.et h4 1fc7 b5 · 19 ltJgJ d5 20 21 e5 .i.xe4 ltle4 21 22 ... 'ifeJ dxe4 22 23 24 .i.f2 l:Xdl Forcing the following exchange of pieces on e4 and the opening of the d file, after which the poor ·position of White' s queen and the fact that his rooks are disconnected will tell. White's next move is forced, since if 20 g5 Black can reply 20 . . . dxe4 (2 1 gxf6 .i.xf6), while if 20 exd5 .i.xd5, and the position in the centre is opened. The other way of exchanging (2 1 ltJxe4 dxe4) would have involved a loss of tempo on the retreat of the bishop. The opening of the d-file is important, in that the exchange of the rooks becomes inevitable, and with it White's hopes of an attack disappear. It is essential to prevent . . . e4-e3 . ltfd8 :xdt + Were it not for the advance of his pawn to b4, Black would merely have been able to watch while the opponent developed his attack on the kingside. While the white knight is making its way from c3 to g3, Black succeeds in intensifying the pressure on the a8-h 1 diagonal, and in particular in securing control of d5 . 17 18 g4 lhe2 b4. If 1 8 'i'xb4.. Black would have advantageously continued 18 . . . d5 . 18 ... 'i'b7 419 After 24 �di .:cs (with the threat of 25 . . . .i.c5 26 ifxc5 e3) it is doubtful whether Black's attack could be parried. Here it became clear to· · me that the game was decideCt: si,:nce liQt long before : it Averbakh and l cMfI:.- ·in · fact been discussing the importance. of widely separated passed pawns in :endings with opposite-colour bishops. White ho:P,es to �aintain the balance in the endga:Qle; but his 'bad' king and I the weakn.�s- of. h.i �·far-advanced pawns on the king$J.Qe· inaf<e this an impossible task. 24 • • • .i.xh4 Forcing a favourable ending witl1 opposite-colour bishops. 25 26 27 ti)xe4 JJ.. xb4 l:td8+ i.xe4! :cs 27 28 . . l::txd8 1fd5 27 l:td2 can be met by the simple 27 . . . .i.f3 (28 g5 h5), when White has no chances of a successful defence. • il.xd8 . JJ... e7 1ff2 J.xb4 'ifrd2 <it>b7 34 35 36 37 38 39 1fc4 JJ..c5 .1ld4 fle2 �e3 �f2 .i.g6 'tibl 1'b7 ifb4+ 1'b1 .txc2 40 1'xa6 ifdl The loss of a second pawn is inevitable. The game was not resumed. 'i'c4 The natural continuation, although it leads to the loss of a pawn. 30 c3 was weaker because of 30 . . . bxc3 3 1 ·bxc3 .i.d5 etc. 30 31 32 1fc5 In view of the open position of the white king, there is no point in Black going in for the exchange of queens. Threatening not only 29 . . . .i.xc2 (30 <&tixc2 'i'xd8), but also 29 . . i..f3 , winning the g4 pawn. 29 30 33 Of course, not 3 3 ... .i.xc2 in view of 34 'ires+ �h7 35 Wfxc2+. Botvinnik-Furman Nikolina gora 1960 1f xa2 •at+ 1fxb2 Nimzo-lndian Defence 1 2 3 4 d4 c4 tf)cJ a3 ti)f6 e6 .i.b4 Then this variation was fashionable, and I decided to test it once again. Now, however, not without reason theory disapproves of 4 a3 . 4 5 • • • bxcJ .ixc3+ c5 The reader �s already · been able to make the acqiiaintance -of 5 . . . t2Je4 (No. 278) and S.t.'.b6 .(No. 28 .: 6 o· · tt)c6·= '7') 420 6 . . . d5 is the conventional reply. After the move in the game White can create a strong pawn centre; nevertheless the main defect of his position - the · weak c4 pawn - gives Black counterplay. 7 8 e4 d6 .i.e3 Practically forcing the reply 8 . . . b6� however, all the same this move is necessary for Black to mount an attack on the c4 pawn. 8 9 10 i..dJ tlle2 b6 0-0 10 A well-known manoeuvre (intro duced by Capablanca and Reshevsky), preventing i..g5 . However, here Lombardy successfully played 1 0 . . . .ia6, and then . . . llc8, . . .ltJa5 and . . . 'ifd7-a4. 11 12 0-0 tllg3 tlla5 f5 Things have essentially reduced to a position which occurred in the game Botvinnik-Reshevsky (Moscow 1 948), the only differsnce being that the e pawn has advanced to e4 not in two moves, but inunediately, and thus White 42 1 has gained a tempo. It can also be mentioned that Black is not in a huny to develop his bishop at a6 (as Reshevsky played). In this case it would be harder for him to make the blocking advance . . . f7-f5. 13 exf5 13 14 tllb5 In order to weaken the pressure on the c4 pawn, White aims to open up the game. exf5 It is necessary to prevent the advance of the f5 pawn. Given the opportunity, it may be possible to play the knight to d5� in addition� subsequently Black will also have to keep an eye on the e6 square. The knight feels quite secure at h5, since . . . ctJf6 can always be answered by .ig5 . 14 15 16 1fe2 g4! tllc7 i.a6 Had it not been for this move, White· s position would have become difficult, since there is no other way of countering Black's plan ( . . . 'l'd7, . .. :aes and . . . Va4). 16 . . . fxg4 17 fxg4 would have led to the opening of lines on the kingside, and therefore Black is tied down by having to look after his f5 pawn.· 16 • • • 'iie8 Exploiting the fact that 1 7 .i.xf5 is not possible due to 17 . . . l:xf5 . 17 · :aet 18 d5 'llf7 At the last moment covering the c4 pawn. 18 19 • • . l£lf4 l:tae8 lie5 After 19 . . . g6 20 gxf5 gxf5 2 1 �hl Black's kingside would have been in danger. The exchange sacrifice con ceived by Furman should have led to an equal game. 20 21 22 23 gxf5 .i.xc4 'flxe3 ltJe6 ltJxc4 %be3 .i.xc4 ltJxd5 The decisive error. Black should have played 23 . . . !te8, and after 24 'i'e4 i.xd5 25 'i'd3 .txe6 26 f.xe6 l:.xe6, with two pawns for the. exchange, he stands not worse. 24 25 11fe4 /l\xf8 .i.xfl .i.hl It was this position that Black was counting on, when he made his 23rd move. Indeed� 26 . . . Xlxf5 is threatened, after which he would have quite good chances. 26 'i'e8 'iixe8 27 l:he8 <3;f7 28 !:ta8 .i.xf5 is also not dangerous for him, since the white knight has no retreat. Now after 26 lDxh7 Xlxf5 27 ltJg5 �xe4 28 ttJxf7 <l;xf7 29 fxe4 tbxc3 White would have had to think about how to make · a draw. However, an un pleasant surprise awaited Black. 26 l£lg6! Alas! Now Black cannot play 26: . .hxg6 because of 27 fxg6, when he loses his queen (27 . . . 'i'f8 28 iixd5+ leads to mate). And if 26 ... ..txfS� then 27 'ifxd5 'i'xd5 28 lDe7+ and 29 tDxd5, and White is a rook up. The doomed f5 pawn plays the leading role in the concluding attack. 26 27 • • • ltJc7 <i>f8 Nothing is changed by 27 . . . <l;h8. ltJe7+ 28 1ih4 Black resigns (28 . . . i.xf5 29 lDxf5 'i'xf5 30 'i'd8+). Botvin'nik-Furman Nikolina gora 1961 Nimzo-Indian Defence 1 2 3 4 5 d4 c4 ltJcJ el li.e2 .: �f6 e6 i.b4 0-0 This move has its points: in some variations of the Ninizo-Indian Defence 422 the bishop is better placed at e2 than at d3 . 5 6 7 8 lhf3 a3 b4 d5 b6 ..i.d6 out this same idea in a somewhat different position, where it proved unsuccessful. It is doubtful whether White has any other way of fighting for the initiative. Now Black must make a choice between counterplay involving . . . c7-c5, and the continuation in the game. 8 9 • . • dxc4 i.xc4 Now the position has become fully theoreticai since the distinction between 5 .ie2 and 5 .i.d3 has been erased. 9 10 11 0-0 i.b2 The initial cause of Black' s sub sequent difficulties: it is not easy to achieve equal chances in such a simple way. The game Spassky-Tal ( 1 975) took an interesting course: 1 1 . . .a5 1 2 b5 e5 1 3 a4 'ile7 14 h3 (14 l::te l i.b4 ! ?) 14 . . . l:ad8 15 'i'e2, and here, in Tal's opinion, he should have eliminated White's opening advantage by 1 5 . . . e4 1 6 tLlh2 %Ue8. 12 12 13 14 15 .lb7 l'.l)bd7 e5 ti)xd6 lDd2 b5 e4 cxd6 'ife7 Ensuring the possible activation of the bishop on the a3-f8 diagonal. l'.l)b5 The simplest plan: by exchanging the d6 bishop� White secures his kingside and awaits a propitious moment to open up the position and exploit the power of his long-range bishops. The positional threat of 1 3 tDxd6 cxd6 14 dxe5 cannot be parried by defending the e5 pawn, and therefore Black is forced to advance it further. In the first game of the return match with Tai (No.285) I tried to carry 423 15 16 ... 11a4 :ac8 Tying down the opponent's pieces to the defence of the a7 pawn. Subse quently White exchanges rooks in order to deprive Black of play on the c-file, and continues aiming for the activation of his minor pieces. 16 17 ... l:tfcl llc7 Of course, not 17 'i'xa7 because of 1 7 . . . l:.a8. 17 18 . i.fl . • ilfc8 Finally threatening 1 9 'ifxa7. 18 19 20 21 l:txc7 .:tel lhc7 .i.d5 lhc7 li)f8 1fxc7 26 ct>f8 26 . . . li:Jc7 27 a:4 . a6 was dubious because of 28 a5 tfJd7 29 bxa6. So, the first part of the plan · has been carried out - the rooks have been exchanged. 22 23 �c4 1fb4 �e6 In view of the weakness of the d6 pawn, sooner or later Black will also have to part with his second bishop. 27 28 . . · . {i;Je7 Finally the second bishop also comes into play, as was envisaged 1 3 moves earlier. 28 29 i.xc4 23 a4 .taJ+ �d7 f3 Since it is wtfavourable for Black to exchange on f3 (in this case White can carry out a pawn offensive in the centre), the· f6 knight will subsequently be tied to the defence of the e4 pawn. In addition, White is preparing a march by his king. 29 30 24 ifxc4 The only way! After 24 .i.xc4 d5 25 i.e2 'i'c2 or 25 .i.b3 etJg4 Black would have gained counterplay. whereas in the endgame he is condemned to passive waiting. 24 25 26 .txc4 .i.a2 .i.f8 �c7 g6 30 . . . tiJe6 was more prudent. The weakening of the h6 square is to White's advantage, although it was hard to imagine that within 1 0 moves the enemy king would end up on this square. 31 <ifi>f2 �e6 if xc4 d5 In such a situation it is useful to keep the bishops as far away as possible from the black knights. TI1e structure of the resulting position resembles the 6th game of the Flohr-Botvinnik match ( 1 93 3 ), which was excellently con ducted by the Czech (at that time) grandmaster. In the present game tl1e advantage of the two bishops is even more obvious. 424 32 33 <li>g3 ilh6 �d7 The bishop is switched to the h2-b8 diagonal for a.ii attack on Black's main weakness - his a7 pawn. He cannot get rid of it, since if 3 3 . . . a6 there follows 34 ilf4. For the moment 34 fxe4 is threatened. 33 34 35 36 .i.f4 fxe4 'it?h4 f5 tDe8 fxe4 • • • 41 ltlg3 42 <&txb7 'i!tf5 Or 42 . .. tLlfl 43 <it>xg6 ltlxe3 44 g4 ! etc. The further advance of the king cannot be prevented. 36 37 If 4 1 <it>xh7 there would have followed 4 1 . . . tDf4, whereas now · this will be impossible because of 42 exf4 e3 43 .i.dl . ltld6 fryi 37 38 . �g5 39 h3 g5 ... 44 g4 45 hxg4+ <&t>xg4 46 .ie6+ Black resigns. The d-pawn cannot be prevented from queening. <i>xd6 'ifi>e6 Furman-Botvinnik Gaining control of the g4 square in advance, so that the black knight cannot go there. 39 40 .i.xd5 <:iJg7 Pa ng the threat of 44 l'fill , on which there would now have followed 45 g4 mate. .ixd6 Now White no longer needs the two bishops. On the contrary� in the resul ting position the fewer the pieces, the fewer the opportunities for counterplay. 43 44 Nikolina gora 1961 Sicilian Defence thf6 1 2 3 4 5 6 �h6 e4 tho d4 lDxd4 c4 iLeJ c5 lLlc6 cxd4 g6 .*.g7 ltlf6 This opening variation occurred in the game Smyslov-Botvinnik oscow 1 956). I wanted to test a system of defence which was frequently employed by . Larsen and Petrosian. 7 iLlcJ lt)g4 (M 8 9 40 41 .i.bJ . thb5 1f xg4 'ifdl tDxd4 lt)e6 This is where I deviated from the afore-mentioned gam:e, in which 9 . . . e5 was played. 425 10 llcl d6 11 il.d3 .i.d7 Another plan involves 10 . . .b6 . A more accurate continuation was demonstrated 1 3 years later by Petrosian with Black against Portisch: 1 1 . . .0--0 12 0--0 a5 1 3 .i.b 1 .i.d7 1 4 f4 .i.c6 1 5 Vie2 b6 16 f5 lllc 5. 12 0-0 13 b4 .i.c6 The most energetic plan, which puts Black in a difficult position. True, energetic play sometimes has its draw backs: in some case it allows the opponent opportunities for counterplay. . • . 0-0 .. .fl.bl 15 16 ttJa4 17 18 .i.b6 1i'd� . � . i.d7 l:tc8 11e8 Until this last move Furman had played well and gained an obvious advantage. 1 8 ..i.xa5 was possible, when 1 8 . . . i.xb5 does not work because of 1 9 lllb6 42, o r 1 8 . . . l:.a8 19 i.c3 , and Black remains a pawn down. But White decided that there was no reason to hurry, and he apparently wanted to gain a tempo: 18 . . . tLlf4 19 'i'd2 (19 . . . .ih6 20 .i.e3). Here the natural continuation 1 3 . b6 led to an advantage for White in the game Polugayevsky-Sakharov ( 1 972). 14 b5! Black is in trouble. Now White could have played 17 lllb 6 %k7 1 8 ttJd5, forcing his opponent to give up the exchange by 18 . . . :cs, or to play the totally unattractive 1 8 . . . %lc8 1 9 i.b6 1i'e8 20 .i.xa5 . But the move in the game also leads to the win of a pawn. 12 . . . a5 (preventing b2-b4) was more prudent, as Larsen played against Portisch ( 1 968). 13 15 Of course, White avoids the opening of the a-file and takes the opportunity to cramp the opponent' s pieces. a5 18 • . • .i.h6 Now the worst for Black is over. If 1 9 J::tc 3 there follows 1 9 . . . tLlf4 with the threat of 20 . . . tLle2+. 19 20 l:tc2 'ifxb7 20 21 22 23 24 lDb2 •a6 1i'b7 'ifa6 lLlf4 The queen should have been retreated to d l . A familiar idea (cf. No. 82): as soon as the white rook is unable to occupy the a-file, this advance ( . . . a7-a5) gains in strength. But here it involves a great risk. �e6 l:xb8 l1a8 llb8 White is now agreeable to draw' but this does not satisfy Black. 426 pieces have been diverted to the other side of the board. After 25 �g2 'i'd7 26 @h 1 l:a8 Black regains his piece, maintaining a strong attack. 25 26 27 1i'xa5 l:ldl l:ld3 .i.h3 fld7 lt:\e3! 28 29 fxe3 1'g4+ 30 31 f4 e5 f3 e4 White resigns Blocking the path of the rook which was hurrying to the aid of its king. 24 • • • lDxg2! Tit for tat! Black's queenside has been destroyed, but he embarks on the destruction of his opponent's kingside, exploiting the fact that five(! ) white 427 f5 <itf2 The advance of this pawn (with the threat of opening the f-file) prevents the white king from fleeing. Postscript Titis book concludes the publication of my games, played in the period from 1 925 to 1 970. Many of them have been annotated specially for the present edition, while the remainder have been substantially revised. In my analysis of the opening, the middlegame and the endgame, I have aimed to convey to the reader the most important thing in chess - po�itional understanding. Sometimes it is founded on general principles, but at other times it is based on accurate, concrete calculation. I hope that these games will not only afford the reader aesthetic pleasure, but will also provide instructional material. 428 Tournament and Match Cross-Tables ( 1 957� 1 970) 1957 Match for the World Championship with V. Smyslov, Moscow Result 0:1 1/2 1/2 0:1 1:0 1 :0 1/2 1:0 1/2 1/2 1/2 1:0 1:0 1/2 1/2 1/2 0:1 1/2 1/2 1:0 1/2 1/2 429 1.0:0.0 1.5:0.5 2.0:1.0 2.0:2.0 2.0:3.0 3.0:3.0 3.5:3.5 4.5:3.5 5.0:4.0 5.5:4.5 6.0:5.0 7.0:5.0 7.0:6.0 7.5:6.5 8.0:7.0 8.5: 7.5 9.5:7.5 10.0:8.0 10.5:8.5 1 1.5:8.5 12.0:9.0 12.5:9.5 1 958 Return Match for the World Championship with V.• Smj&lov, Moscow Result 0:1 1 :0 0:1 1/2 1 :0 1:0 1/2 1/2 1/2 1/2 1:0 1:0 1/2 1:0 1:0 1/2 1/2 1:0 1:0 1/2 1/2 0: 1 1/2 430 1.0:0.0 2.0:0.0 3.0:0.0 3.5:0.5 3.5: 1.5 4.5: 1.5 5.0:2.0 5.5:2.5 6.0:3.0 6.5:3.5 6.5:4.5 7.5 :4.5 8.0:5.0 9.0:5.0 9.0:6.0 9.5:6.5 10.0:7.0 1 1.0:7.0 1 1.0:8.0 1 1.5:8.5 12.0:9.0 12.0: 10.0 12.5: 10.5 1 958 1 3th Olympiad, Munich Preliminary Group =ftn@t :llt@fil 1itHtI lUJlM j�$l¥ l� !jj§fUJ j�•�tmt �•i@m l�IJjj�jjjjf 2.5 4.o 4.o 3.s 4.o !lttlt.t1n �j�t!'!!!'! !'Wn!'!'!'!':·:·�:ss'!".'!'!:.:·:::.'!';'!'!:Tlt�Jj��j]j�!'!'!'!'@t!'!'!'!'t@�mn jjf.tB.1r·=·=·=:=·:;· ·==:=r=�i1m o. 1.5 4.o 3.5 3.5 3 .5 llit�stmr 2.0 2.5 4.o 3 .5 4.o l#l�IHjj@j �ji�t&UU.61Hl@H1fi� 1.5 3 .o 4.o jj#.J!.SjJlHl .o 3 .o jlifNiliifliilijI� 1.s 1. .5 4.o Uitllmt lj#.�@Uiii.fililtll 1.0 3 .5 at�J1lt 2. iilil llMl: o.o ljlflilt o.o M @ilit.jljM jlmf:=:-�=:=:·:··:]t1MfMl jl$.Uliiffl fi. ii l llll: o.5 Final Group 3.5 3.5 2.5 2.5 jjJi Jtl.§B.1 l@ttltf j�J�
[email protected]@litlt l l�i.jti.MitiiHHlMll 2.0 �ll�YfillfllMfiMll 2.0 3 .5 2.5 3.0 2.5 2.0 1. 5 3.5 2.5 2.0 2.0 2.5 Hl!JIE 1.5 @11.�il 2.5 dli�ll mt&fi�jiji 2.0 2.0 43 1 I)�):)!IiEIJ:I�t�:�rnm:I�l)t�:t:i:tfiitf:I@:):tet:a�a.1%1&tlJ.itf:JI11It�i:t:t:J:t:MtEit:�:1Eit:J:i:[:irn�:rnm 1st Round 2nd Round 3 rd Round · 4th Round 5th ound 6th Round 7th Round 8tb Round Botvinnik - Donner Diickstein - Botvinnik Botvinnik bye Botvinnik bye Botvinnik - Raizmann Walsh - Botvinnik Botvinnik bye Cintron - Botvinnik The Netherlands Austria Bul2aria Italy Francia Ireland Danemark Puerto-Rico 1/2 1:0 1-0 0-1 0-1 I[M)�ffI1j)j\j)lM)])�)��)�)1{1)1�M)j@)jij)1�)j\jtmmmmmmjfifdi.i)1)1fi.ijf.lm1m)j)j�j)jj1)1Il1)j)j)]�1j)jtllj)j)j�)tEjl])[j )1�lIM)I[j[{�m 1st Round 2nd Round 3 rd Round 4th Round 5th Round 6th Round 7th Round 8th Round 9th Round 10th Round 1 1 th Round Botvinnik - Lombardy Botvinnik - Unzicker Botvinnik bye Botvinnik bye Pachman - Botvinnik Botvinnik - Pomar Botvinnik bye Botvinnik - Alexandrov Uhlmann - Bot\innik Botvinnik - Diickstein Ne.ikirch - Bot\'innik USA West Germany Bul2aria Yu2oslavia Czechoslovakia Spain Ar2entina En2Iand East Germany Austria Switzerland 1 958 International Tournament, Wageningen 432 1/2 1/2 1/2 1:0 1:0 0-1 1-0 0-1 1 959 2°d USSR Spartakiad, Moscow Preliminary Group Final Group 1tf1tftt\1HtnJtllff11t:i.lin.ltiifiWmtJF::lii)lGfi.uliitt JftE11lJ@E1KtFf.\JiiiWlJI 01. Geller - Botvinnik 02. Botvinnik - Gur2enidze 03. Kholmov - Botvinnik 04. Botvinnik - Nezhmetdinov 05. Botvinnik - Alejandrovich 06. Spassky - Boninnik 07. NurmamedO\' - Botvinnik 1/2 1/2 1/2 1/2 1/2 1/2 0-1 433 1960 Match for the World Championship with M. Tai, Moscow ,. Result 1:0 1/2 1/2 1/2 1/2 0:1 1:0 1:0 0:1 1/2 1:0 1/2 1/2 1/2 1/2 1/2 1:0 1/2 1:0 1/2 1/2 434 1.0:0.0 1.5:0.5 2.0: 1.0 2.5: 1.5 3.0:2.0 4.0:2.0 5.0:2.0 5.0:3.0 5.0:4.0 5.5:4.5 6.5:4.5 7.0:5.0 7.5:5.5 8.0:6.0 8.5:6.5 9.0:7.0 10.0: 7.0 10.5:7.5 1 1.5:7.5 12.0:8.0 12.5:8.5 1960 14th Olympiad, Leipzig Preliminary Group 2.0 ��lm!llUJ · �\lllUlM �ltlllM��� Final Group 43 5 ![l[JrntI!IlEI!il!I11IIIII]IItiittrnJtmlll.Mii.n{liiuit:�:IJ!lfl:ff�!ItIIJ!Ititil[J\IIII1Et 1st Round 2nd Round 3rd Rouild' ·,·�· 4th Round""f·"·t!' · --�·, . 5th ound 6th Round 7th Round t 8 h Round 9th Round Botvinnik - Kostjeorin Aaron - Botvinnik Botvinnik - Donner Botvinnik bye Botvinnik bye Tamburini - Botvinnik Botvinnik - Eliskases Botvinnik bye Botvinnik - Durao Monaco India Netherlands Phillippines Austria Italy Ar2entina Poland Portu2al . · '· · ' 1-0 0-1 1/2 0-1 1-0 1-0 t\i!t:IM!mJm!irnti!mini�mmmmmm�miiitmm�I:1111um1u.11tIIi:1:1ttt�:�11jiIIIt:ti t:mm111mmi11tr 1•t Round 2nd Round 3rd Round 4th Round 5th Round 6th Round 7th Round 8th Round 9t11 Round 10th Round 11 th Round Ne.ikirch - Botvinnik Botvinnik - Ivkov Botvinnik bye Portisch - Botvinnik Botvinnik - Lombardy Botvinnik bye Botvinnik - Schmid Botvinnik - Pachman Botvinnik bye »rimer - Botvinnik Botvinnik bye Bul2aria Yu2oslavia Netherlands Hun2ar-v USA Ari?entina FRG CSSR GDR Romania En2land 0-1 1/2 1/2 1/2 1-0 1-0 1/2 1 960 Match Moscow vs. Leningrad Moscow r11111mi111111rtmmmrnr1rrruwrwrnrnt1111m:lmnr1nrtn1rn1rm@mrnw1r:n:rnrnirdmMnummm1 1-0 Kortschnoi - Botvinnik 1/2 Botvinnik - Kortschnoi 436 1 96 1 Return Match for the World Championship with M. Tai, Moscow Result 1 :0 1 :0 1:0 1/2 1/2 1/2 1 :0 1 :0 1:0 0:1 1:0 1 :0 1 :0 1/2 1:0 1/2 0:1 0:1 0:1 1/2 1 :0 437 1.0:0.0 1.0: 1.0 2.0:1.0 2.5:1.5 3.0:2.0 3.5:2.5 4.5:2.5 4.5:3.5 5.5:3.5 6.5:3.5 7.5:3.5 7.5:4.5 8.5:4.5 9 .0:5.0 10.0:5.0 10.5:5.5 10.5:6.5 1 1.5:6.5 1 1.5: 7.5 12.0:8.0 13.0:8.0 1 961 2°d European Team Championship, Oberhausen 01. Botvinnik - Perez Perez 02. Perez Perez - Botvinnik 0 3. Botvinnik - Unzicker 04. Unzicker - Botvinnik 05. Pachman - Botvinnik 06. Botvinnik - Pachman 0 7. Szabo - Boninnik 08. Botvinnik - Szabo 0 9 . Botvinnik - Gli2oric 1-0 1/2 1-0 1-0 1/2 l-0 0-1 1/2 1/2 1 96 1-62 International Tournament Hastings * * 1 0 0 0 0 * * * 0 0 0 1 0 * * 0 0 0 1 0 43 8 * 1 1 1 * 0 0 1 1 1 * 1 * 0 1 962 International Tournament, Stockholm 15th 1 9 62 Olympiad, Varna PRELIMINARY GROUP A 439 Final Group 3.5 . 3.0 : 2.0 2. 5 2. 5 2. 0 2. 0 2.5 3.5 : 2.5 3.0 2.5 1.5 2.0 3.5 1.0 IlitffIJJit:t:tt:l:t:1i�:1:tlIIttfa�l.lfllil.Wiiil: @fiGIJJlltiffIIt:11::i:Jm1:1: 11:it:rnt:l 1st Round 2°d Round 3m Round 4th Round 5th ound 6 th Round 7th Round 8th Round Uhlmann - Botvinnik Botvinnik bye Lundin - Botvinnik Botvinnik - O'Kelly Botvinnik - Pomar Botvinnik bye Botvinnik bye Kokkoris - Botvinnik 1st Round 2°d Round 3 rd Round 4th Round 5th Round 6th Round 7th Round 8th Round 9th Round 10th Round nth Round . Botvinnik bye Botvinnik bye Botvinnik - Fischer Botvinnik bye Botvinnik - Padevsky Unzicker - Botvinnik Botvinnik - Uhlmann Botvinnik bye Filip - Botvinnik Botvinnik - Donner Botvinnik - Robatsch East Germany West Germany Sweden Belejum Spain Norway Turkey Greece 1-0 1/2 1/2 1-0 0-1 thlJIIt:�:1m11:: :l: :t: i: m�1:trn11t::1:1�t:tm*1mi:r-Jt1uiiWIIt:mmrn::mmtmm:tEIII:�m:�: :itmtmrni 440 Yugoslavia Argentina USA 1/2 Hungary Bulgaria West Germany East Germ.any Romania 1/2 0-1 1/2 Netherlands 0-1 1/2 Austria 1-0 Czechoslovakia 1 963 Match for the World Championship with T. Petrosian Moscow ., Result 0: 1 1/2 1/2 112 1:0 1/2 1:0 1/2 1/2 1/2 1/2 1/2 1/2 0:1 1 :0 1/2 1/2 1 :0 1 :0 1/2 1/2 1/2 44 1 I - · ·0: 1 I · . :. 0,5: 1,5 1.0:2.0 1.5:2,5 2.5:2.5 3.0: 3.0 4.0:3.0 4.5: 3.5 5.0:4.0 5.5:4.5 6.0:5.0 6.5:5.5 7.0:6.0 7.0: 7.0 8.0: 7.0 8.5: 7.5 9.0:8.0 10.0:8.0 1 1.0:8.0 1 1.5:8.5 12.0: 9.0 12.5: 9.5 3rd 1 963 USSR Spartakiad Moscow Preliminary Group A Final Group J\lfIJI:Hfftif@UtmmmII�@Immmutl.UliBiil�!�-tiiiJEtm�mmIEMJ�:�llt�Il@IWIMIIll 1st Round 2nd Round r1 Round 4th Round 5th ound Botvinnik bye Krutikhin - Botvinnik Gipslis - Botvinnik Botvinnik - Shofman Danov - Botvinnik Gcor2ia Kirtzistan Lettland Moldavia Tad.iistikan 1/2 0-1 1-0 0-1 :Ktrn:1mrn@1i1:J:m11m:1m1rnmn�mm1ttt1m1u&11.u1i1mmr1rnrn1tJl:ff1rnirnttt1rnmmmrn11r i•t Round 2ne1 Round 3 ra Round 4th Round 5th Round Kholmov - Botvinnik Botvinnik - Veresov Taimanov - Botvinnik Botvinnik - Stein Botvinnik - Buslaev 442 RSFRS Belorussia Leningrad Ukraina Georgia 0-1 1-0 0-1 1/2 1-0 1963 International Tou rnament Amsterdam 1 964 Moscow Team Championship Moscow NiltttEJftftttlttttfltfflitJEtJMMMl@t�tlllWtllfW@lW�tl%t�J:tfltP8iiidtlfl tHHJJ Botvinnik - Simacln 1/2· Botvinnik - Liberzon 1-0 1 964 Works Team Events, Moscow 443 1 964 USSR Team Championship Moscow ttfollftllJtMltHmmm1mmrntJtHfl@lttmmttJ{lfWHlflttltlltltttI tHBidlittm:lliM:@ 01. Karpov - Botvinnik 1/2 02. Tai - Botvinnik 1/2 1/2 0 3. Geller - Botvinnik 0 4. Botvinnik - Petrosian 1-0 1-0 05 Botvinnik - Smyslov 1-0 Botvinnik - Stein 1 964 1 6th Olympiad, Tel Aviv Preliminary Group A �ittmml l2lftlt :jarmmm =Jtttttr l l$.Mtttt l1filtrmm nttt11 4.o 4.o 4.o irt!IM@l 4.o 4.o lIUV.BKlttlllM 3 .o 3.o t.5 2.5 �llHlWJ llZ!t·'.aittMltltl o.5 1.5 2.5 llif=:-·-:: ··=··=·:·=:=w=ii.iMM� o.o 2.5 �lifiUM1 2.5 J.o 2.5 111.Jittt ll�leJUlilMmrmnmt o.o 2.5 2.s �jilli5Wl l llSWki&iijil{l� o.o ili.Uit o.o n:at1w1 l:m¥ittn1i o.o llZlflNl�l �#.�S.tiiUtMfMUilt II".'!"!'!'!���!".".'!'!'!'!'!'!".'!' ! 444 Final Group 2.5 2.0 4.0 2.0 3.0 2.0 2.5 1.0 2.0 2.0 1 .0 1.5 2.0 2.0 0.5 1.5 1.5 2.0 1.5 2.0 2.0 2.0 1.0 1.0 1.5 1.0 3.0 0.5 1 .0 1.0 1.0 3.0 1.0 1.0 2.0 2.5 2.S 3.0 2.0 3.5 2.0 4.0 2.0 2.S 3.0 1.5 2.0 3.5 2.0 2.0 1.5 3.5 2.5 2.5 2.5 1.0 2.0 0.5 2.0 1.5 1.0 1.0 1.5 1.5 1.0 2.5 2.0 1.5 0.5 2.5 J:�() 3J) 3.5 "3;5 3.0 1.0 3.0 3.0 2.0 3.5 2.0 3.0 2.5 3.0 1.5 2.5 3.0 2.5 3.0 3.5 3.0 1.5 2.5 3.0 2.5 2.5 2.0 2.0 2.5 1.S 2.5 0.5 3.0 1.5 0.5 1.5 2.5 3.5 3.0 2.5 2.5 jj'-�jj jf1.�fj� }j·��� j��,$.�j ���,�j�� �11.#.t jit.¥.t !ll@ �!@t jJ�Jfj 1.5 1.5 IE::11I1: :R:]]JIII::If: : I:tm:@1:1mn1muu.ii.iiiIB.mi.fi:� IIrn:n�:::�EJ:: ]l::]:t: :IIIf: �: :i:j:j:j:fIItrn: 1st Round 2nd Round 3rd Round 4th Round 5th ound 6th Round 7tJJ Round Botvinnik - Medina Botvinnik bye Kirby - Botiinnik Botvinnik bye Letellier - Botvinnik Botvinnik - Max Botvinnik bye 1st Round 2nd Round Jrc1 Round 4th Round 5th Round 6th Round 7th Round 8th Round 9th Round 10th Round 1 1 tb Round 12th Round 1 3th Round Eliskases - Botvinnik Aloni - Boninnik Botvinnik bye Botvinnik - Bobotsov Botvinnik bye Botvinnik - Gli2oric Ciocaltea - Botvinnik Botvinnik bye Botvinnik bye Yanowsky - Botvinnik Sliwa - Botvinnik : -::Botvinnik - Laslo Botvinnik bye Suain 1-0 S.A.R 0-1 Chile Switzerland 0-1 1-0 :ttJ}:: : :�:r: : :i:l:�rn::1:wl:tMr11:::m:i:j:�: :1J@t:: :Jt1m1u1raun::r:ti�:�:i:�:�: rntm:Itt1iti1t:i: :f:�fIIE:: :j:tfftt: 445 Israel 1/2 0-1 Bulgaria 1/2 Yugoslavia 0-1 0-1 Argentina Rumania Canada Poland.. 0-1 1/2 1/2 1 965 International Tournament, Noordwij k 1 965 Moscow Trades Unions Spartakiad Moscow lJ !MfurWifillffiiitW.ffiK1 fMM1@%'MtHlMJMJWiMJMHffrtiW!W1&fil®WMiiWMt%%'iltt# Bohinnik +2-1=1 1 965 ,, Trud" Sports Society Team Events Moscow 446 1 965 rd 3 European Team Championship Hamburg tHlfltHlltMlfllltlflttMlhdlWd.fi�l!lGitiiitrnitmH@@fi!Rfltlli@itHilH!hl.tiWiltW 1/2 0 1. Schmid - Botvinnik 1-0 02 Botvinnik - Schmid 1/2 03. Szabo - Botvinnik 0-1 04. Boninnik - Szabo 1-0 05. Gli2oric - Bomnnik 1/2 06. Boninnik - Gli2oric 0-1 07. Ciocaltea - Botvinnik 0-1 08. Boninnik - Lothar 1 965 USSR Trades Unions Spartakiad Moscow Jffi'WtijjiiiB.llW4!!1 *illW..& ' i1ifil@M£'1@f rmmwJMMlhlI@!lilll!WWtOOMl&@@A¥t1MB&'Mfilt1mL Botvinnik 447 +2-1=1 1 965 Match Moscow v. Leningrad Moscow nrnr:w11i1mr11ntwmtttttimm:w1u1rm1rm1mu111nmm1rnmmmmmmmwrnmmrm:�mn�ltimu.rnr1@ 0-1 Tolush - Botvinnik 1-0 Botvinnik vs. Tolush Total +2-0=0 1 966 Moscow Team Championship Moscow @m1mmWtu1�11.-•mmn1mmmmmm:rmmmm::mrn1turrnm rt111R1n111n•1wnFtJmmnrnrn 1/2 Liberzon - Botvinnik 1/2 Bronstein - Botvinnik Botvinnik - Smyslov 1/2 0-1 Yudovich - Botvinnik 1/2 Botvinnik - Petrosian 1 966 ,,Trod" Sports Society Team Events, Moscow 448 1 966 International Tourna ment, Amsterdam 0 1 1 * * * * 0 0 0 * 0 * * * * * * 0 1 1 * 0 * * 1 0 * * 0 0 * * * * 1 * 1 1 966 USSR Team Championship Moscow iltFfFtHlitfrlfittfHWiillll.iliiiUJtQiat..1 EillJ%HJtfFWJltUtltltt� llti.$.illtl 0-1 0 1. Botvinnik - Petrosian 02. Tai - Botvinnik 1-0 OJ. Yudovich - Botvinnik 0-1 1-0 04. Botvinnik - SmyslO'\' 05. Bierbrager - Botvinnik 0-1 06. Botvinnik - Keres 1-0 07. Lutiko'' - Botvinnik 0-1 08. Botvinnik - Geller 1/2 1/2 09. Botvinnik - Stein 449 1 966/67 International Tournament Hastings 1 967 Works Team Events, . Moscow r1@fi@Nff1@%illM!MiW.iMk4Wt¥1\%WllilliMmmmwmrnUMMMMlf£&&.1ill@m•smoom� Botvinnik +1-0=0 1967 Moscow Team Championship, Moscow ll'!l:l a� J �r !'!'!'l '"i& 'l'!l!'! ;:=·:·:1 '!1!'!'! :=·���'!1!'!'! JjJ!'!':l!l ;�m� -w!'!'.'!' l�m ��Hl� Jt@l@t �jJ@@H :jltm t M � •tl@l :j�IIlmr� t::ttlt t!Hfll� �jllllt 1.0 9.5 s.o 1.0 9.s 10.s l!®lif.MU.iiMH: 6.5 1.0 9.5 12.5 jjJ.�OiJ.tlilM lllUt ilS.�Ji:Jiamfij1t1@jJjl� j�§it.ilUtlt1Ml11 ljf��ifiiilfa �ji�Mil.Mlll@11M. 5.5 9.0 8.0 ics.=·=·='.aidkltmttm 12.5 9.5 4.5 . 2.0 3.5 4.5 4.5 4.5 4.5 3.5 t.5 t.s 10.0 4.5 450 l rM!tit j f.Sl!Mll JiZ&mm �jjfii@nm j}l£1MM Jtl.�ltUf l0J$j�lll Il&lftt Ftlilfl lMmwn-.�i.f\Dimiijili.@111fF1F11FttRlfHl!HflJtt11FitJFtfFtF1tHtll\1 Botvinnik - Balmlin Smyslov - B_otvinnik Liberson .,. :Q�fyinnik PetrosiaJi,-:-,Sotvinnik Botvinnik .,. Bronstein · Botvinnik ?? Botvinnik ?? - - 4th 1-0 1/2 1/2 1/2 1/2 1-0 1-0 1 967 USSR Spartakiad, Moscow Moscow's Team: Petrosian, Botvirurik, Smyslov, Bronstein, Kholmov, Vasiukov, Liberson, Dvoretsky, Kushnir, Konopleva, Rubcova Prelimary Group Final Group �-�!'!'!'!'!'!';'�?!'!'!'Jl!I!' It:t:tl1JE mtt@tWHt : �:ttfftlt::Hffl#Hll �'=tHlWtm fl!Et:ttJJ: l!Mai(ii@JltJJlll ffRSIBStlJftHEtll 6.5 J .5 4.o 1.5 5.5 6.o 6.o 5.5 1.0 1.0 5.5 5.5 rnttlrMWi111Witltt 6.0 m111r=== -=-=-=-:1aE1trmrn: 2.5 5.s 1886.miM.if@Fi@It 4.5 5.5 ::1J.SN==-=:Jtn11111111 4.o 45 1 5.5 \i.2W!\:) flltiif f\11�$.\ ::zartn -�!lU!F ::1r1�1 JlM0.i.ittdi:nmm1111t:tn1:11tFJll?fftl¥MtlffWI1lI1M1@tlFJ11FHlt1F@HIWUH11f Muhitdinov - Botvinnik 0: 1 Botvinnik - Khanov 1-0 Levit - Botvinnik 0: 1 Botvinnik ... Polu2aevsky 1:0 Taimanov - Botvinnik 1-0 Botvinnik - Boleslavsky 1/2 1/2 Geller - Botvinnik Botvinnik - Roman 0-1 Total: +5-1=2 1 967 International Tournament Palma de Mallorca * * 0 0 0 * 0 0 * 0 1 * 0 * 0 * 1 1 * * 1 * * * * * * * * 0 * 0 0 0 * 0 * * 1 * * * 0 * 0 * * 0 0 0 0 0 0 0 * 0 * 0 * 0 0 0 0 0 0 1 * * * * * 1 1 * 1 * 0 1 * * 0 0 * * * * 1 0 0 0 0 0 * 1 1 1 * 1 o 1 * 1 * * 1 * * * 1 o * * * * 1 * 0 0 * * * * * * * * 0 * 0 0 * * * 0 0 * * 1 1 * 452 1 1 * 1 * * 1 1 * * 1 0 0 0 * 0 0 1 * 1 * * * * 1 * 1 0 * * 0 * 1 1 0 1 1 1 1 * o * 1 1 1 * * * * 0 * 1 0 1 1 1 1 1 * * * 1 * * 1 * 1 * 1 1 1 * 1 * 1 1 * 1 1 * 1 * * 1 * 1 o * * * 1 * 0 0 0 0 1 1 0 0 1 * 1 1 1 * 1 * * o 1 o 1 1 1 o l @l�ti:� 1 i 1 1 * 1 * * * 1 o * 0 1 1 * :�jfj$: rJ!Uf aitt1:� .:11.a:i : tt.tl.�l ::�utu ]f$.:1t �:&1t: :\&II% JUfl :I6$. ll �iari ::1i1. �j :1av.tm :15.IIm �llSjfij lSl@f 1 96 8 International To urnament Monte Carlo * * * 0 * * 0 0 * * * * * * * * 1 * 0 . * * * 0 0 0 * 0 0 0 0 0 0 0 0 * 0 * * * 1 * * * * * 1 1 * * * 1 * * * * 1 1 1 * * 0 * * 0 1 969 Moscow Trades Unions Spartakiad Moscow �lf.ifiiilMMlli@l@MitIHl\1�1 l1B.B.fi fl S.kllJttltMl t llIIt ��$.mi:itiiMil@l@MllI@lt@nNI o.5:o.5 1.0:0.0 0.5 :0.5 453 j¥.itaiilit&Wllt1lM1lltt ·�temtiiilt i l t 111Uit?itfM lfi.SiB.1@tlUl1�tlllllll 1 969 International Tou rnament Beverwij k 0 * 0 0 0 * * * * * * * 1 0 1 1 1 0 1 0 0 * 0 1 * * 0 0 * * 0 0 * * * * 1 * 0 0 0 * 0 0 0 * * 0 * 0 0 0 0 0 0 0 0 * * 1 1 * 1 1 * * 0 * * 0 0 0 0 0 0 1 1 0 454 * 1 * 1 * 1 * * 1 1 1 1 1 1 * * * 1 1 1 1 1 1 * 1 0 1 1 1 0 * * * 1 1 1 1 1 1 1 1 1 0 1 0 0 * * 1 * 1 0 * * * 1 * * 1 9 69 International Tournament Belgrade * 1 0 1 * 1 1 0 * * 0 0 0 0 * * 0 0 0 0 0 0 0 0 * 0 0 0 * 0 * * 0 * 1 1 1 0 1 0 0 0 * * 0 1 0 0 * * * 1 1 * * * * 1 * 1 * * * 1 * 0 * * 0 * * * * * 0 0 * * * 0 * * * * * 1 * 1 970 International Tournament Leiden 455 1 1 * 1 1 1 1 * 1 * 0 1 * * * 0 1 1 1 1 1 * * * * * 1 1 1 * 0 * "* 1 1 1 1 970 Match USSR v. Rest of the World Belgrade * 1 0 0 0 * 0 * * 1 * * 0 * * 0 * * * * * 1 0 1 1 * 0 1 * 1 * * * 0 1 * * 1 * 1 !lSll.li �ll&mm l!lf@l jll.tl 456 Tournament and Match Results (1 957-1970) Year I+ Event Match for the World Championship with V.Smyslov, Moscow Return Match fo r the World Championship with 1 958 V. Smyslov, Moscow 1 3th Olympiad, Munich International Tournament, Wageningen 1 959 2nd USSR Spartakiad, Moscow Match for the World Championship with M. Tai, 1 960 Moscow 1 4th Olympiad, Leipzig Match Moscow v. Leningrad Return Match for the World Championship with 1 96 1 M. Tai, Moscow 2nd European Team Championship, Oberhausen 1 96 1/62 International Tournament, Hastings 1 962 International Tournament, Stockholm 1 5th Olympiad, Varna 1 963 Match for the World Championship with T.Petrosian, Moscow 3rd USSR Spartakiad, Moscow International Tournament, Amsterdam 1 964 Moscow Team Championship Works Team Events, Moscow USSR Team Championship, Moscow 16th Olympiad, Tel Aviv 1 965 International Tournament, Noordwijk Moscow Trades Unions Spartakiad 'Trod' Sports Society Team Events, Moscow 3rd European Team Championship, Hamburg USSR Trades Unions Spartakiad, Moscow Match Moscow v. Leningrad 1 966 Moscow Team Championship 'Trud' Sports Society Team Events, Moscow 1 957 457 I Place 3 6 13 7 7 3 1 5 1 11 4 2 6 2 8 - 10 4 7 8 5 2 7 3 1 4 3 7 5 2 4 2 2 2 �I - - 6 - 1 5 1 - - 1 5 - - - - - 1 - 1 - 3 1 - - - I 13 5 1 6 4 2 1 6 15 2 2 1 1 3 4 2 1 1 3 1 - 4 1 I I I I Year Event + 1966 Internati.onal Tournament, Amsterdam USSR Team Championship, Moscow 1 966/67" International Tournament, Hastings 1 967 Works Team Events, Moscow Moscow Team Championship 4th USSR Spartakiad, Moscow International Tournament, Palma de Mallorca 1968 International Tournament, Monte Carlo 1969 International Tournament, Beverwijk Moscow Trades Unions Spartaki.ad International Tournament, Belgrade 1970 Match USSR v. Rest of the World, Belgrade International Tournament, Leiden 458 7 5 · S"· 1 3 5 9 5 6 1 5 1 1 = - 1 2 1 1 2 3 - - - 1 1 - - - 3 - 2 Place I I 4 2 7 II-III 8 II 9 I-II 2 7 VII 3 9 III-IV Translator' s notes* 1 (p. 17) After this White can gain the advantage by 4 1 l:.xg6+ �xg6 ( 41 . . . �h5 42 :h6+) 42 .i.xe5+ �h5 43 .llc3 . But 39 . . .lhf6 is clearly better: 40 :xg7+ <Ji;xg7 4 1 lid7+ ltf7 42 l:d5 exf2+ 43 00 e4. 2 (p.27) This allows 3 1 . . :'ilb5. 3 1 t2Jc4, with the same idea, is better, or else 3 1 'i'c4, e.g. 3 1 . . .:xd6 3 2 �xd6 exd6 3 3 e7. · 3 (p. 3 1) Even so, White has to reckon with 27 ... 1fb6, when ... llb5 is threatened. 4 See Shakhmatnoe Tvorchestvo Botvinnika (Moscow 1 966) Vol.2 p.626. (p. 36) 5 (p. 57) 6 7 There seems to be some misunderstanding here. After 45 'ib7 Black wins by 45 . . . 'i'xb7 46 cxb7 d2 47 �xe2 .ih5+, but 44 .i.xg7 or 44 .i.e5 looks good enough to win. (p. 59) Black seems to be alright after 1 5 i.c6_ .i.h3 , e.g. 16 i.xeS 'i'xe8 17 :tel .ic3 18 .ib2 i.xe l 19 'i'd4 i.xf2+ 20 <li>xf2 f6 2 1 'i'xf6 'i'f8. (p.72) 23 'i'd6 'iif7 24 l:tc7 would have won immediately. 8 (p.72) After 38 lill'5 Black has 38 . . . 'ifbl + 39 lth2 'i'xf5 40 ltxd2 'i'f4+, but 38 t2Je6 'i'xe6 39 .:xd2 wins. 9 But 51 � fxg3 52 <atxg3 h5 53 @f3 'ittf5 54 lte3 <it>g4 55 ltd4 ltxh4 56 c!?c5 �g3 57 �b6 h4 58 �xb7 h3 59 a6 h2 60 a7 is a theoretical draw. (p. 95) 10 11 12 (p. 10 1) But then 20.. . lill'8 and if 2 1 ltxf6 'i'h7 when Black wins, as pointed out by Tai, who suggests instead 20 if'f3 . (p. 1 26) But then 18 t2Jxd5 looks very strong. (p. 1 3 3 ) Black looks alright after 14 . . .lDe4 1 5 l:b2 ..ta4 16 f3 11i'xc l 17 :xcl t2Jd6. Instead 14 f3 , denying the black knight the e4 square, looks stronger: 14 . . . tDh5 ( 1 4 . . . l:tbS 15 t2Jg6) 1 5 :n t2Jxr4 16 l:xc2 f6 ( 16 . . . lib4 17 tDxc6) 17 llxc3 ttJe2+ 18 � ttJxcl 1 9 l!hxc6. * A computer was used to help in checking the analysis. 459 13 14 15 (p. 1 75) In fact after 33 . . . �g7 34 l:lxf5 Black appears to have a virtually forced win: 34 . . . 'i'e3 3 5 h4 (3 5 .i.f2 :xrs; 3 5 tiJe2 :txf5 36 'i'xf5 'i'xe2+) 3 5 . . . 'ifgl + 36 �h3 'i'hl+ 3 7 �g4 (37 .th2 i..e 5) 37 . . . .i.f6 (threatening 3 � . . . 1:.h8) 3 8 l:tf2 :hs 3 9 l:th2 (39 'iif5 tiJeS+) 3 9. . . 'i'gl 40 ttle2 'ii'e 3 4 1 'i'f5 :xh4+ 4 2 i..xh4 ttle5+. (p. 1 76) After 46 . . .<t>h7 White cannot play 47 'i'xc7+ in view of 47 . . . i..g7. (p. 1 90) 14 . . . ttld4 wins on the spot! Presumably the move order 13 i..xe6 .i.xe6 14 exf6 'ilxf4 etc. is intended. 1 6 (p. 1 90) The immediate 16 .i.b5+ looks more decisive: 16 . . . axb5 1 7 ttlxb5 'ifb6 1 8 .i.xc5 1i'xc5 1 9 lbc 1 . 1 7 (p. 1 9 1) 2 4. . . 1i'g5 looks stronger. 18 . (p. 1 99) Tue tactical 10 l:xh4 'ifxh4 1 1 tZ:lxd5 also looks very strong. 1 9 (p.202) White has a clear win with 59 <:Ji;e7 ztd2 (or 59 ... h2 60 d8'i' :e2+ 6 1 :e6 l:xe6+ 62 �xe6 hl'i' 63 'i'xg5+; 59 . . . �e2+ 60 :e6; 59 .. .lk? 60 l:d6 h2 6 1 l:dl) 60 l:te6 h2 6 1 :e l . 20 21 22 23 (p.226) 34 . . . %:.g8+ is much stronger: Black wins after. 35 <tin lig2+ 36 <ii?g l t£ixc4 3 7 lhc4 i.d5+. (p.243 ) Mark Dvoretsky claims a win for Black by 62 . . . g4+ ! 63 �xg4 (63 <it>e2 <ite4) 63 . . . .i.c3 ! (threatening 64 . . . e2 or 64 . . . <itd3) 64 i.b5 a4! 65 �f3 a3 66 h7 @c5. (p.252) A citation from Voltaire' s Dictionnaire Philosophique, (p.253) After 49 . :f3 it is not altogether clear how White can stop the black pawns. . . 24 (p.260) It would seem that the author was right to be afraid of the check; after 24 l:h2 there is no good defence against 25 l:te2+. 25 (p.26 1) But it would seem that White could have drawn by 22 .i.h6+ 'iti>g8 23 'i'xe5 ifxh l + 24 lte2 'i'xh6 25 'ife6+ <i!rh8 26 'i'e5+ with perpetual check, as 26 . . .'i'g7 loses to 27 l:th l + 'itg8 28 ..-e6+ ..-f7 29 l:h8+. 460 26 (p.276) 2 1 .. .'ifa6 seems an adequate reply. 27 (p.280) 25 . . J�fB is a tougher defence. 28 (p.286) 25 . . liJgf3+ 26 �g2 lbxd2 27 !ib8+ <i;;f7 28 :xd8 hihxdS 29 .i.xd2 l:ab8 leaves Black with a material advantage, but 25 J.xb5+ ltJxbS 26 :I.xb5 looks fine for White. . 29 (p.289) White doesn't appear to lose his knight after 38 �f3 ..i.h4 (3 8 . . . .i.f4 39 ltJe6) 3 9 <iti>g4 ..i.g5 40 ltJe8. 30 (p.296) But after 32 tiJd4 there appears to be no satisfactory defence to the threats against e6 : 32 . . . .i.xe4 3 3 iif7. 31 (p.296) By 3 1 'iiil d5 32 'i'f8+ <iti>c7 33 \ic5+ <iti>b8 34 1i'xa5 White can win a piece, although after 34 . . .'i'xh2 Black wins the g2 pawn, with .Perhaps reasonable compensation. 32 (p.296) After 34 . . . <itid7 3 5 'i'xa5 .:txc2 36 <iti>xc2 ifc4+ Black regains his pawn, but with 34 'W/g7 fie7 35 'i'g6+ <itid8 3 6 l::[f7 White can win the queen. 33 (p.300) Black can answer this with 17 . . . lbxc4. Therefore 17 1fif3 liJxd3 1 8 'i'xd3 seems better - cf. the note to White's 16th move! 34 (p. 328) 25 'i'xf6 :xd7 26 'ifxe6+ :Lf7 27 ild5 followed by 28 'i'g6+ and 29 'f!lxg7 mate is more incisive. 35 (p.368) Black has a forced win by 39 . . . g5 40 g3 l!f3 . 36 (p.375) 47 l:tel l:.xe l 48 �xe l is even better, with a won pawn ending for White. 37 (p.3 84) 28 . . . 'fllxb3 29 ifd4+ liJe5 looks unclear. Therefore simply 27 lbe6 seems better, with irresistible threats. 38 (p.3 85) It would appear that White has nothing better than perpetual check after 30 . . <it>f7 3 1 ttJg5+ (or 3 1 'i'g7+ @e8) 3 1 . . .�g8. . 39 (p.388) 34 . . . :d l +! is more immediately decisive. 40 (p.406) 1 8 .. Jk2+ was also possible: 19 l:txc2 (or 19 <t>fl l:txb2 20 ..i.xa4 l:.xb7) 19 . . . 'i'xc2+ 20 <ittfl 'i'xb2 21 'i'xe7 \ixb5+. 46 1 41 (p.4 1 1) 38 �3xc2 is possible, when White is a clear pawn up. 42 (p.426) In fact this can be met by 1 9 ..ta4 20 lhxa4 l:a8, when Black regains the piece. . . . 462 Index of Openings . . Open Games Ruy Lopez. C77 - 297. Semi-Open Games Caro-Kann Defence. B U - 257, 274; B 1 2 - 289, 293; B l 4 - 269; B 1 8 215, 277, 329; B l 9 - 345, 3 8 1 . French Defence. COS - 365; C l I 337; C 1 5 - 256 ; C 18 - 299. Pirc-Ufimtsev Defence. B06 - 3 36, 378; B07 - 3 16, 3 1 8, 3 3 1 , 374; BOS - 304, 353, 370, 376; B09 - 3 27, 334. Sicilian Defence. B25 - 368; B34 265 ; B36 - 352; B37 - 3 39; B38 369; B63 - 347; B72 - 260, 26 1, 30 1 ; B75 - 298, 3 1 5; B88 - 281. Closed Games Catalan Opening. D l l - 280. English Opening. Al4 - 32 1 ; A l 6 263, 264; A20 - 37 1 ; A2 1 - 306; A22 - 322; A23 - 288; A24 - 303 , 351; A2 6 - 343, 362: A29 - 27 1 , 302, 33 3 , 344, 363 , 366; A3 5 - 309; · 04 1 - 3 1 9. Griinfeld Defence. D70 - 3 50; 075 3 4 1 ; D79 - 254, 358; 083 - 328; 098 - 307. King's Indian Attack. A07 - 284, 338. King' s Indian Defence. A42 - 270, 375, 377: A44 - 283 ; A48 - 348; A56 - 3 26; A6 1 - 276; E6 1 - 349; E62 - 252� 253, 3 6 1 ; E66 - 273; E69 - 295 ; E80 - 259; E81 - 258, 292, 3 80; E82 - 279, 323; E85 29 1 ; E87 - 294, 3 24, 3 54, 355; E90 - 3 57, 372. Nimzo-Indian Defence. E24 - 278, 2�7; E3 5 - 308; E42 - 3 3 5; E43 272, 3 1 7, 3 56; E45 - 255; E48 346; E5 1 - 285, 286, 364, 367. Queen' s Gambit. D3 1 - 3 13, 373, 379; D32 - 332; 034 - 296; 036 - 330, 342; D46 - 282. Queen's Gambit Accepted. D24 - 267; D27 - 3 10, 3 1 1 , 3 14. Queen's Indian Defence. E l 4 - 305; E 17 - 360 ; E I S - 3 1 2. Reti Opening. A05 - 266; A12 - 262; A14 - 300, 359; E l 7 - 320. Slav Defence. D13 - 268, 325; D l 4 290, 340. C77 etc. are Encyclopaedia ofChess Openings codes Numbers refer to ga.:uies 463 Already published : Mikhail Botvinnik : Botvinnik's Best Games Volume 1 1 925-1 941 Mikhail Botvinnik: Botvinnik's Best Games Volume 2 1 942-1 956 464 £: 29 · 99 : .�