Interesting, If a Bit Austere by Taylor Kingston Sokolov's Best Games, by Ivan Sokolov (with John Nunn), 1997 Cadogan Chess, Figurine Algebraic Notation, Softcover, 157 pp., $19.95. This book's title could cause some confusion. Its author and subject is GM Ivan Sokolov, born 1968 in Yugoslavia, who as of July 1999 stood 27th on the FIDE rating list at 2656. He should not be confused with Russian GM Andrei Sokolov (born 1963) who was once among the world's top 5 but who since his loss to Karpov in the 1986 FIDE Candidates final has faded out of the top 100. While Andrei rose high and fell, Ivan Sokolov has been less famous but more consistent since gaining the GM title in 1987. He does not often win major events but usually scores a respectably high place. He has not quite broken into the ranks of the elite, never qualifying as a FIDE or PCA Candidate, and in the 1997 FIDE knockout championship he was seeded into the second round but lost immediately to Epishin. Still, he has many noteworthy accomplishments to his credit, such as the championships of Yugoslavia and Bosnia-Herzegovina, 1st place at Portoroz/Rogaska 1993 (category 12), 2nd (behind J. Polgar) at Madrid 1994 (category 15), equal 2nd (with Kamsky, behind Karpov) at Groningen 1995, and clear 2nd at New York 1996 (behind Adams but ahead of Korchnoi and Salov). His best success within the time frame of the book is probably his clear 1st in the 1995 Dutch Championship, with a score of 9«/11. He has stood as high as 12th on the FIDE rating list. The book contains 50 games, all wins by Sokolov, over the period 1988-1996. Each is introduced by GM John Nunn with a few paragraphs giving the context of the game and discussing its major strategic or tactical points. Annotations are by Sokolov. The result is a rather austere book which will have little appeal to the average player but which serious students of at least expert or master strength could well enjoy and learn from. A book like this stands or falls based mainly on the quality of its games and annotations. Of the quality of the games there can be little doubt: Sokolov's victims include such major names as Anand, Adams, Topalov, Yusupov, Gelfand, Beliavsky, Timman, van Wely, Kamsky, Akopian, Lautier, and Korchnoi. The annotations are oriented toward advanced players, and they concentrate mostly on middlegame complications, a Sokolov forte. Some excerpts from Sokolov-Topalov, Wijk aan Zee 1996, will give an idea of his annotative and playing style. In this position (See Diagram): Sokolov played the TN 18 Qb3, and comments: "With the text White keeps the queens on the board and quickly castles queenside, believing that in the coming complications his king will be safer than its black counterpart. If White tries a similar idea with 18 Qa4+ b5 19 Qf4, Black would have decent counterplay after 19...Rhe8 20 0-0-0 Qe4. 18 Qg4+ f5!? 19 Qf3 is interesting, while previous theoretical discussion had been focused on the ending that arises after 18 Qf3 Qxf3 19 Nxf3, where White had failed to prove any serious advantage after 19...Rae8 20 Kf2 Re4! 21 Bxc5 Rhe8 22 Re1 Rxe1 23 Nxe1 b6 24 Bxb6 Kd6 (a point being 25 c4 Re4) Thuesen-de Firmian, Farum 1993, or 19...Rhe8 20 Kf2 Kd6 21 c4 f6 Illescas-Topalov, Alcobendas (4) 1994." The game then continued 18...b5 19 0-0-0 Rhe8 20 Bxc5 Rac8 21 Bd4 Qg2 22 Qa3! Qxg3! 23 Qxa6 Rxc3+! 24 Kb2!, reaching this position (See Diagram): Topalov here played 24...Rcc8, but Sokolov comments "Black had an amazing possibility that might have saved the game, viz. 24...Qg2+!! 25 Kxc3 (25 Ka1? Qxd5) 25...Rc8+ ...". He goes on to analyze 26 Kb4?, 26 Kb3?, and 26 Bc5?, and finally concludes "26 Kd3 was analyzed by Burgess as good for White: 26...Qc2+ 27 Ke3 Re8+ (27...Qxd1 28 Qxb5+) 28 Kf4 Qf5+ 29 Kg3 Qg5+ 30 Kf2 Qh4+ 31 Kf1 Qf4+ 32 Bf2 Qc4+ 33 Kg2 Qg4+ 34 Bg3." From these excerpts it should be apparent that Sokolov is writing for the serious rather than the casual player, that he strikes a good balance between general explanation and technical detail, that his style is straightforward and neither particularly dry nor colorful, that he is unafraid of and able to handle tactical complications, and that unlike some annotators he does not pretend that he saw everything at the board. Perhaps not immediately apparent is that he is becomingly modest the position after 34 Bg3 in the above line is a lot better than just "good for White" (about +6.03 says Fritz4, in which case one might quibble that 24...Qg2+ objectively deserves only "!?" rather than "!!"). As an interesting aside, GM Alexander Shabalov annotated a game in this line, Remlinger-Shabalov, Las Vegas 1993, in the 10/93 issue of Chess Life. Shabalov claimed that at move 18 "White couldn't escape from trading queens (believe me, no way) ...", and that Black's play constituted "a sort of requiem to the Taimanov System in the Benoni." Sokolov's play, particularly 18 Qb3, appears to refute Shabalov's claims. In the openings, familiarity with many lines is assumed; Sokolov's emphasis there is on theoretical innovations. These he explains in a well-written, logical, objective fashion. An example is Sokolov- Luther, Aalborg 1991: 1 d4 Nf6 2 c4 e6 3 Nc3 Bb5 4 Qc2 0-0 5 a3 Bxc3+ 6 Qxc3 b6 7 Bg5 Bb7 8 Nf3 d6 9 Nd2 (See Diagram): Sokolov comments: "I worked out this interesting set-up in 1991. In general in this line (7 Bg5), if White opts for the strategy of setting up a full centre (f3 and e4), it often takes quite a while to develop the knight from g1 and at the same time it would be useful for White to have better control over the square e4. With the knight manoeuvre witnessed in this game White tries to solve both of these problems in the most efficient way. The drawback of this concept is the additional loss of time, should White need to position his knight on c3 if Black hits the white centre immediately with ...c5: instead of Ng1-f3-d2-b1-c3, a shorter route is Ng1-e2-c3." Another example is in the Open Spanish, in the game Anand- Sokolov, European Clubs' Cup, Lyon, 1994, where after 1 e4 e5 2 Nf3 Nc6 3 Bb5 a6 4 Ba4 Nf6 5 0-0 Nxe4 6 d4 b5 7 Bb3 d5 8 dxe5 Be6 9 Nbd2 Nc5 10 c3 d4 11 Ng5 (See Diagram): Sokolov's TN 11...Bd5!? caused the normally quick-moving Anand to think for almost an hour. Sokolov has devised several innovations in the Classical (4 Qc2) Nimzo-Indian and is considered a great authority on it. Ten of the book's 50 games feature that variation, all with Sokolov on the white side. Those who play the line would benefit from the insights he offers. In four other games he plays the Rubinstein line (4 e3). Sokolov appears to be very much a 1 d4 player; as White he makes no other first move among these games. As Black he leans toward the Slav against 1 d4 (though there is one very interesting game with the Blumenfeld Countergambit), and he replies to 1 e4 with 1...e5. While the overall emphasis is on opening innovation and middlegame calculation, Sokolov is also not without skill in the endgame. Here against Yusupov at Nussloch 1996 (See Diagram): he found the unusual key move 43 Ba3!, after which "Black cannot bring his king closer to the action with 43...Kg7 due to 44 c6! bxc6 45 Bd6 Bxd6 46 Nf5+. In any case 44 c6 with Nf5 and Bd6 to follow is a deadly threat." While the games and annotations are impressive, some key things are missing from this book that with only a little more effort would have improved it considerably. There is no biographical introduction, and we learn next to nothing about Sokolov as a person, just bits and pieces in Nunn's game intros. There are no photographs. Most serious from an historian's viewpoint is the lack of any tournament tables or career record. These are commonly, almost invariably included in single-player collections, and their absence here is puzzling. Also, with someone of Sokolov's obvious tactical ability, why not add a few "greatest hits" combinational positions, like those featured in Vishy Anand's recent book? So while those omissions make for the austere quality mentioned earlier, on the whole this is a collection of interesting games with annotations that strike a good balance between strategic explanation and analytical detail, though at a level that will probably fail to engage anyone with only a casual interest or an Elo rating of less than, say, 2000. While this is not a book we can rave about, Sokolov's Best Games is recommended for serious students interested in some unpretentiously good work by a lesser-known but successful modern grandmaster.