GRANDMASTER Andy Soltis is a fascinating chap. He was one of the generation of talented American players who came to the fore following Bobby Fischer’s victory over Boris Spassky in their 1972 match for the World Championship, but has managed to combine a distinguished chess career with the duties of a professional journalist.While not working for the New York Post, or playing chess, Soltis has found the time to write a huge number of chess books. These include tomes on almost every opening you can think of, always accurate and well explained, but my favourites are his more idiosyncratic efforts. Confessions of a Chess Grandmaster<$>, an autobiography which mostly chronicles his long and eventually successful quest for this title, will certainly remain on my list of the ten best chess books of all time.

Andy and his wife Marcy (also a journalist) are regular competitors in the Bermuda Open, and indeed he mentions our tournament very favourably in the above book. They were unable to play in the 2005 edition because of a commitment to their newspaper, which consisted of test-driving a Caribbean cruise (nice work if you can get it!), but we hope to see them back here soon.

GM Soltis came into my mind because I have just discovered a fascinating book which he wrote in 2004, entitled Rethinking the Chess Pieces. This is a detailed analysis of such topics as which pieces are best suited to certain tasks, which combinations work well together, etc. While the usual formula for the relative values of the pieces (pawn=1, knight=bishop=3, rook=5, queen=9) is not ready for the dustbin, it falls far short of telling the full story.

Chapter 5, The Personalities of the Pieces, includes a section on the way in which a rook can make the life of a knight intolerable on an open board. This is illustrated in its starkest form by the position shown in Diagram 1, which was analysed as long ago as the 7th century AD.

Readers who are paying attention may wonder how this can be possible, since the modern game of chess did not come into existence until well after that date. The answer is that this was taken from the game of shatranj, an early version of chess in which, as luck would have it, the moves of the king, knight and rook were already as we know them today.

The black knight is badly placed, cut off from the protection of its king. The white pieces, by contrast, are well positioned in the middle of the board. Even so, it initially seems unlikely that it is possible to force a win, but by a series of instructive manoeuvres this can indeed be done.

White starts with 1. Rf5 (I wanted to put “check”, since we are hunting down the knight just as if it were the king). Now the toughest defence is 1. . . . Nd2, we shall look at the alternatives later. There follows 2. Rf4, reaching Diagram 2. The knight must move again, since after, say, 2. . . . Ka6 the white king just walks in and wins it with 3. Kb4!, Kb6 4. Kc3, Nb1+ 5. Kb2, Nd2 6. Kc2. So the continuation is 2. . . . Nb3+ 3. Kb4, Nc1 (the knight is on the edge of the board, never a good sign) 4. Kc4, and this is the critical position shown in Diagram 3.

White threatens to win the knight with 5. Rf2 and 6. Rc2. The conclusion is 4. . . . Ne2 5. Rg4, Kc7 (the knight cannot return to c1 for the reason just given) 6. Kd3, Nc1+ 7. Kc2, Na2 8. Ra4 and in Diagram 4 it’s over.

Having seen these winning plans, it should not be difficult to work out how to deal with other first moves by Black. After 1. . . . Nh2?, for instance, 2. Rf4 traps the knight immediately and the king can win it at leisure. 1. . . . Nh4 is a little tougher, but 2. Rf6 cuts off most escape routes and White wins easily with 2. . . . Ng2 3. Kd4, Ne1 4. Rf2!

See if you can handle the other possible defences, including Black’s attempt to use his king with 1. . . . Nh4 2. Rf6, Kc7 3. Kd4, Kd7 4. Ke4, Ke7.