Chess

TODAY AND on Monday, two reviews of books by former World Correspondence champions, both published by Gambit but in all other respects as different as you could imagine.

Readers may have noticed that I seldom criticise books. The explanation is simple: I usually don't bother with ones I don't like. But The System by Hans Berliner (Gambit pounds 14.99) has made such an impression that I must make an exception.

I first delved into this on a train on the way to Gatwick Airport and it so incensed me (I admit I was also tired) that I went three stops too far and nearly missed my plane. Later, I showed it round to my Bundesliga team and it attracted comments and ridicule of great ferocity.

The reason for this is its extraordinarily arrogant, indeed almost Messianically bombastic tone. Berliner, in his time, was an unprecedentedly successful champion with a career record of 94 wins, 10 draws and just one loss. He also created a fine computer program, Hitech. But his stated belief in the foreword that chess will be solved by 2030, and his attempts to refute major openings such as the Grunfeld and Queen's Gambit Declined, are utterly at variance with the modern, flexible understanding of chess. One reason for this is that much of the book was written 20 or 30 years ago, and his purely analytical approach - though understandable in a correspondence and computer man - makes almost no concessions to positional understanding.

I'll start with his assertion that "White's correct first move is 1 d4 because it controls three central squares while no other move controls more than two". Leaving aside arithmetical considerations - how does he define a central square to get exactly three? - this is extraordinary special pleading. The move, after all, also weakens e4 and as a result he spends much of the rest of the book trying to make f3 work in sundry positions.

Next there's the claim that after 1 d4 Nf6, 2 c4 is better than 2 Nf3. Of course, if White's position is sufficiently good then it may be. But, a priori, you don't, for example, know whether the weakening of e4 and the early exposure of the b4-e1 diagonal exploited by the Nimzo-Indian - 1 d4 Nf6 2 c4 e6 3 Nc3 Bb4 - is a serious matter. (And Berliner, admittedly in old notes - game 12 - confessed that he didn't know how to meet this.)

Some of Berliner's specific analysis, especially against the Grunfeld, is extremely interesting. But he simply doesn't consider a sufficiently broad range of lines to convince the modern eye of his very strong assertions - for example, after 1 d4 Nf6 2 c4 g6 3 Nc3 d5 4 cxd5 Nxd5 5 e4 Nxc3 6 bxc3 Bg7 7 Bc4 c5 8 Ne2 0-0 9 Be3 Nc6 10 Rc1 cxd4 11 cxd4 Qa5+ 12 Kf1 Ilya Gurevich's Qa3!, which he doesn't mention, is now the main line.

Although I find parts of this book disturbingly compelling, I'm happily far from convinced that the chess Messiah has come.