"Paul Keres: The Road to the Top" by Paul Keres, 1996 American Batsford Chess Library, Figurine Algebraic Notation, Softcover, 256pp., $23.95 "Paul Keres: The Quest for Perfection" by Paul Keres and John Nunn, 1997 American Batsford Chess Library, Figurine Algebraic Notation, Softcover, 272pp., $23.95 It has become commonplace to refer to the late great Estonian grandmaster Paul Keres as one of the strongest (perhaps even THE strongest) player never to have become world champion. The basis for this statement is not only emotional but factual. No doubt, particularly in the late 1930s or 1940s, had he been given the opportunity to play Alekhine, he would have defeated the world champion. The problem is that his career as a world class player was always in the shadow of Botvinnik. Before his untimely death in 1975, Keres had written a multi- volume series with a collection of wonderfully annotated games about his career. The reader may recognize the original titles: "Grandmaster of Chess: The Early Years of Paul Keres;" "Grandmaster of Chess: The Middle Years of Paul Keres;" and "Grandmaster of Chess: The Later Years of Paul Keres." The set had neither the magnet of the Fischer name nor the high profile of Capablanca, but it still was an outstanding series. It was, however for the most part, overlooked, or, at least not given the attention it deserved. The American Batsford Library (the U.S. counterpart of Batsford Publishing in England), has re-issued the series in a two- volume set, with some editing and additional material added by British GM John Nunn. It is outstanding. The first volume is entitled "Paul Keres: The Road to the Top" and deals with Keres' career until 1950. There are fifty-one annotated games by Keres, with additions, corrections and comments, all given in the form of footnotes, added by Nunn where he sees fit. The second volume is entitled "Paul Keres: The Quest for Perfection" and concerns itself with the balance of Keres' career, presenting fifty-two annotated games. The original series presented games through 1962. Nunn has added five games to this part of the book, drawing from the original Russian edition that contained them (the English edition did not) and then added fifteen games himself to complete the series. The result is a marvelous presentation of the games and career of one the great world-class grandmasters of all time. Keres, particularly in his youth, was noted for his slashing attacking style. While never quite losing his taste for this style of play, as he became more mature, his play became more positionally and strategically based. This two-volume series chronicles the evolution of Keres' style and his best games as he does battle over a 45-year playing career against the world's strongest. From the first volume... Stahlberg - Keres, Bad Nauheim 1936, Nimzo-Indian Defense 1 d4 e6 2 c4 Bb4+ I often employed this move-order at the time, even against some of the world's best players, and with quite good results. The idea of this order of moves is that after 3 Nc3, Black, in addition to the Nimzo-Indian Defence 3...Nf6, can also choose to transpose into the Dutch Defence by means of 3...f5, a defence in which the early development of White's bl-knight is not held to be the best. But the chief reason for my choice lies in the fact that quite often positions that had been but little investigated by theory tended to occur and in consequence players would be forced to think for themselves from the very first moves. 3 Nf3 In many games my opponents played 3 Bd2 to which I usually replied 3...Qe7, transposing into positions similar to the Bogo-Indian Defence. The text-move is more energetic and leads to more interesting positions. 3...c5 4 e3 In the same tournament Alekhine played 4 dxc5 against me, when great complications arose from the continuation 4...Bxc3+ 5 bxc3 Qa5 6 Nf3. Black could very well answer the text-move with 4...f5 but instead decides to transpose into the Nimzo-Indian Defence. 4...Nf6 5 a3 Theory recommends 5 Ne2. Now we reach the Saemisch system. 5...Bxc3+ 6 bxc3 0-0 7 Bd3 d5 Nowadays this advance is seldom played, since White can select a system of development against which Black experiences great difficulty in gaining any active counterplay. 7...Nc6 or 7...d6, aiming at ...e5, is more usual. 8 cxd5 exd5 9 Ne2 This method of developing the knight, which inaugurates a system evolved by Botvinnik, is held to be best for White. His further strategic plan is as follows: first to complete his development by 0-0 and Ng3, and then to play f3 and, after due preparation, carry out the advance e4. Practical experience shows that it is very difficult for Black to obtain active counterplay against this plan, and therefore this variation is rarely seen in modem tournament practice. At the time that this game was played, Botvinnik's system had not yet been worked out and so both players were faced by new problems over the board. Therefore, it is not surprising that the following play is not carried out in accordance with the latest word of modern theory. 9...b6 The idea behind the text-move, to exchange off White's d3-bishop, is one of the best in this position. 10 0-0 Ba6 11 Bc2? Today this move can be given a question mark since it is common knowledge that White can assure himself a promising position by 11 f3 Bxd3 12 Qxd3 Re8 13 Ng3 Nc6 14 Bb2 followed by Rae1. With the text-move Stahlberg wants to keep his bishop to support the advance e3-e4, but the black bishop on a6 is too well placed and noticeably impedes White's plans. In addition White loses an important tempo, so that Black obtains an advantage in development. 11...Nc6 12 Rel This move should also be criticized since in the subsequent play the rook is badly placed on el. Better, as recommended by Stahlberg, was 12 f3 followed by Rf2. 12...Re8 13 f3 Rc8 14 dxc5? This anti-positional exchange results in a speedy disadvantage for White. If the variation 14 Ng3 cxd4 15 cxd4 Nxd4 16 Bxh7+, and then 17 Qxd4, does not appeal to White then he should be content with the loss of a couple of tempi and return with his bishop, 14 Bd3, so as to carry out the advance e4 if possible. Perhaps Stahlberg had not taken into consideration the possibility of 16 Bxh7+, and therefore deemed an exchange on c5 to be indispensable preparation for Ng3. 14...bxc5 15 Ng3 (See Diagram) 15...d4! Black could also get a good position by means of 15...Ne5 but this central thrust is much more energetic. Usually, opening the centre when one's opponent has the two bishops is a dubious operation, but here this advantage is counter-balanced by Black's superior development, so that Black once again holds the initiative. 16 exd4 White would not have fared better with 16 cxd4 cxd4 17 e4 avoiding the opening of the centre. There could follow 17...d3 18 Ba4 Nd7 or immediately 17...Nd7, when Black would have obtained an overwhelming middlegame thanks to his strong passed pawn. 16...cxd4 Black could also first exchange rooks by 16...Rxe1+ 17 Qxe1 and then play 17...cxd4 but in that case after 18 cxd4 Nxd4 White would have at his disposal the good square dl for his bishop. After the text move White must exchange on e8, since 17 cxd4 would lose material on account of 17...Qxd4+!. 17 Rxe8+ Qxe8 18 cxd4 Nxd4 19 Ba4 White is forced to place his bishop passively on the edge of the board since 19 Qxd4? is, of course, impossible because of 19...Qel+ and the line 19 Bd3 Bxd3 20 Qxd3 Qel+ 21 Qf1 Qxf1+ loses a piece after 22...Nb3. 19 Be4 is no better, since White would be in a horrible position after 19...Rd8. 19...Qe5 Threatening 20...Nxf3+ and thus practically forcing White's reply. 20 Rbl Nd5! Black's two centralised knights have an overwhelming effect. 21 Bb2 White has nothing better against the threat of 21...Nc3, since the attempt at centralisation by 21 Ne4 loses material after 21...Be2 and ...Bxf3. 21...Nc3 22 Bxc3 Rxc3 23 Khl (See Diagram) In the event of 23 Ne4 Black could either have captured on a3 or else played 23...Rd3. The text-move temporarily protects the a-pawn because 23...Rxa3? would now be a blunder because of 24 Qxd4 !. At the same time the king is removed from the threatened diagonal gl-a7. 23...h5! Frustrating any possible mating ideas on the back rank and threatening the unpleasant ...h4-h3. White cannot any longer hope to organise a satisfactory defence since his scattered pieces are entirely lacking in co-ordination. After 24 Ne4, for example, there could follow 24...Be2 intending a sacrifice on f3. 24 Bd7 Rd3 Here Black misses a chance of winning quickly by 24...h4 25 Ne4 Be2! followed by ...Nxf3, which would completely demolish White's kingside. The game was played in the last round and if I won it I could still overhaul Alekhine. It is, therefore, very natural that I was a little excited and so did not manage to find the strongest continuation. The text-move is also very strong, although now White can defend himself stubbornly. 25 Qa4 After 25 Qel the simplest way to win is 25...Qxe1+ 26 Rxe1 g6 since White loses the a-pawn on account of the threat of 27...Nc2. Black can also interpose the move 26...h4 which forces White to play 27 Nf1 because of the possibility of...Nc2. With the text-move White wins a tempo by attacking the bishop and would like to utilise the move to build up a defence. 25...Bb7 26 Ne4? Allows Black to bring off a pretty finish, but it is very difficult by now to find an adequate defence. The best practical chances are afforded by 26 Qc4. Then the sacrifice 26...Rxf3 27 gxf3 Bxf3+ 28 Kg1 Qe3+ 29 Kf1 would lead to nothing and the complicated variation 26...Re3 27 Rxb7 Rel+ 28 Nf1 Ne2 29 Rbl Rxb1 30 Qc8+ Kh7 31 Qc2+! g6 32 Qxb1 Qd4 33 Ng3 only results in an approximately equal ending. But in this variation the simple protection 26...Qe3! ensures Black a decisive attack, since after 27 Nf1, or 27 Nf5, Black's 27...Qe2 creates threats on f3 and the back rank that cannot be parried. [fn 1 There is a serious flaw in this line, since 27 Be8! leaves Black with nothing better than 27...Bd5 28 Qxd5 Qxe8, when 29 Ne2! leads to the win of a pawn for White (although this would probably be insufficient for victory). Of course, Black is in no danger of being worse after 26 Qc4, for example 26...Rb3 27 Rd1 Re3! is just equal, but it seems that Black's error on move 24 might have been more serious than Keres imagined.] Also 26 Bh3 would have afforded White no salvation since after 26...h4 27 Nf1 (or 27 Ne4 Bxe4 28 fxe4 Rxh3!, etc.) 27...Qe2 [fn 2 This line is also flawed, as after 27...Qe2 White can defend quite easily by 28 Qd7!, threatening the b7-bishop and at least perpetual check on d8 and h4.] there comes a decisive sacrifice on f3. 26...Bxe4 27 fxe4 Nf3! 0-1 After 28 gxf3 Rd2, mate on h2 is not to be prevented. The games and career of Keres have been overlooked, particularly in the West, for much too long. These two books should become a permanent part of every serious player's library.