Single Comment

According to Murray on page 57, al-'Adli, in the ninth Christian century, reported that in India a stalemated player won, which he contrasted with the rule with which he and his readers were familiar. According to that same account, the elephants stood in the corner squares, and had the move of the dabbabah rather than the alfil; the rooks stood on the c and f files. A player who bared his opponent's king won, even if the opponent could return the compliment on the next move--a rule that was also current in the Hejaz, although the rest of the Moslem world held that if the opponent could even the score on the next move, the game was tied. <p> As for the crosswise arrangement of the kings, the arrangement shown in Hans Bodlaender's diagram is that given by Murray on page 80 as used in recent times in India in the varieties of chess that were apparently of native descent (distinguished from shatranj, introduced by the Persian conquerors, and European chess, introduced by the British, French, and Portuguese conquerors). In ancient times there may not have been a fixed rule. As for pawn promotion, the rule given in a work 'written about 1600 or 1700,' as Murray says, seems to me to say that a pawn reaching the ultimate rank on the a, d, e, or h file is promoted to counsellor, and a pawn reaching the ultimate rank on the b, c, f, or g file is returned to its square of origin with the rank of counsellor (ferz). This is on page 64, and the text is as vague as the date. As for stalemate, 'When a king is imprisoned without standing in check, and no other of his pieces can move, he may slay the piece of the enemy in his vicinity which imprisons him.' So the stalemated player does not win. Just before this we read: 'It is not proper to protect another piece rather than the King. The slaying of the King is yet considered proper. Imprisonment is counted as a defeat of the King. If the King is left entirely alone it is reckoned a half-victory, if he is checked 64 times in succession he is also held to be defeated.' I think we might fairly understand that as perpetual check. The game begins with each player moving his counsellor and counsellor's pawn two squares forward, 'Also another piece which goes one square distant is advanced at the same time...' apparently another pawn. <p> On page 81 Murray describes Hindustani chess, one of the three native varieties current in the nineteenth century, when all the pieces had the moves current in Europe. In this game a pawn reaching the last rank is promoted to the master piece of that rank, except that on either central file promotion is to vizier (Q). Further, a pawn can only be promoted if the appropriate piece has already been lost; a player can never have more of any sort of piece than he started with. Murray specifies that the color restrictions of the elephant, now moving as a bishop, must be observed. This means that a pawn cannot be promoted to elephant on the c file until that player has lost his elephant that started on the f file. A pawn that cannot be promoted cannot be advanced to the eighth rank; it must remain where it is, an immobile target. Whether it can offer a threat that cannot be executed is a subtlety that seems not to be addressed. Logic suggests to me that it cannot. On page 82 Murray describes various conclusions to the game. Checkmate is a win. Stalemate is not allowed; a move that inflicts stalemate must be retracted, and another move played. Capture of all of a player's pieces (pawns are ignored) is a half-win. When both players are reduced to a king and a single companion, the game is drawn. Perpetual check is a draw. It is in this game, not the earlier ones described above, that the king has the privilege of making one knight-leap, provided he has not been checked. According to one questionable source, the king cannot capture on that leap. Conclusion: John Gollon, sitting in the Cleveland Public Library, reading its copy of Murray's _History of Chess_ and taking handwritten notes, confused three sets of rules. Fergus, I hope you're not too disappointed. This page, with its seventy comments, should probably be scrapped.