Shatranj problem, March 2015

Saturday, 7th March 2015

There are two solutions to February's problem. The first one is: 1. Queen e6-d7, Rook c8-d8; 2. King g6-f6, Rook c8-a8; 3. Rook e7-e6, Rook a8-b8; 4. Rook e6-c6, Rook b8-a8; 5. Rook c6-c8 (check), Rook b8-c8 (takes rook); 6. Queen d6-c8 (takes rook). Black is left with a bare king, and is unable to reduce white to a similar predicament this turn: in shatranj, this is therefore a loss for black. In move 2 black could have moved the rook to b8 instead, moving back and forth between b8 and a8 in the same manner until required to capture the white rook.

A Reminder of the Rules of Shatranj

This is a brief reminder, for chess players, about how shatranj, or mediaeval chess, differs from the modern game.

The king, rook and knight move as the modern pieces do. The pawn only ever moves one step forward, never two. The bishop moves exactly two squares diagonally, and can jump over a piece in the way. The queen moves diagonally to an adjacent square, much like a draughts king.

Castling and en-passant moves are unknown. A pawn always promotes to a queen. Stalemate is a loss for the trapped player, and if one side is reduced to only a king then that player has lost the game.

The diagram shows the full setup of the pieces, and illustrates the symbols used for each of the pieces in the mediaeval game. For a full discussion of the game, including a full set of rules, see the Shatranj page.

The second solution can be forced by black but has the same eventual result. It is: 1. Queen e6-d7, Rook c8-a8; 2. King g6-f6, Rook a8-d8; 3. Rook e7-h7, King f8-g8; 4. Rook e7-g7 (check), King g8-h8; 5. Rook g7-e7 winning as in solution 1 from moves 3 onwards.

This month's problem comes from a real game, between Rabrab (white) and Na'im al-Khadim. There were three different recorded solutions that gave white a win, but the shortest occurs on the fifth turn. Can you find it?