Chess960, now adopted by FIDE, is too entrenched to be changed. As implied by his article title, I had invented a new variation called Chess480 which uses the Castling rules I had naively hoped would be adopted by Chess960. Chess480 is played worldwide, on several chess servers, alongside and in some cases instead of Chess960. Where it is played instead of Chess960 it is because of the complications in implementing the castling rules.
In this article I will address each of O'Shaughnessy’s points against Chess480’s Castling and submit further evidence that the Castling rules for Chess960 are flawed.
O'Shaughnessy’s first assertion is that when the King starts on b, c, f and g files in Chess480 he has “unsavoury castling options”. The assertion is that Castling to the far side would bring the King into the center of the board, while Castling near side leaves him in the same wing that he started. While it is true that far side castling brings you to the center of the board in Chess480, it is unlikely a savoury option to Castle far side in Chess960 or in Chess480.
Position #538:
nrkqbbnr/pppppppp/8/8/8/8/PPPPPPPP/NRKQBBNR
In this position Castling with the h-file Rook would require moving 6 pieces (two pawns, two bishops, and a knight) before Castling was possible. This assumes that the opponent doesn’t interfere with your plans. Here is a possible opening using the most reserved of moves.
Chess960
nrkqbbnr/pppppppp/8/8/8/3BPP1N/PPPPQBPP/NR3RK1
The movement of four squares by the King (which, by the way, is not the longest distance he may be asked to travel in Chess960 Castling) does make him act more like a Rook than a King. 1
Chess480
nrkqbbnr/pppppppp/8/8/8/3BPP1N/PPPPQBPP/NR1RK3
Contrast that to Chess480 and we see that the King moves the more customary two squares.
Neither position feels ‘unsavory’ to me, however it seems more likely that moving such a large number of pieces for the express purpose of Castling would be easily thwarted by an astute opponent. I find it highly unlikely that in either variant this opening would have much appeal.
I choose to illustrate this point with position #538 to demonstrate a particular quirk of Chess960. I call it “The Rook’s Leap”, which involves no movement of the King at all during Castling. When the King starts on either the c-file or g-file, he may Castle without moving in Chess960. The only pieces to move in these cases would be the Rook, which leaps over the King. Almost 29% of Chess960 game positions start with the King having the option of Castling without moving. I find that unsavory.
O'Shaughnessy asserts that the “purpose of castling is to vacate the king from the often volatile and dangerous centre to a (hopefully) safe haven, and to enable communication between the rooks.” Two points should be made here. First Chess960 can’t claim to do a very good job of this when there are positions where the King doesn’t even move during Castling. Secondly, there is no more ‘purpose’ in Castling than there is a purpose to moving your pawns two spaces on the first move. It is an evolved rule. A convenience of a common convention of the King’s Leap.
As Vladimir Vukovic points out in his book The Art of Attack In Chess page 59:
“At the time when modern chess was in its infancy, i.e. after the reforms of the rules in the last quarter of the fifteenth century... The castling move itself was at the time only in the process of being created (some were still using the mediaeval king’s jump, while others were castling but in two moves - i.e. first Rook h1-f1, and then on the next move King e1-g1, etc.)...”
Castling was actually a simplification and unification of these moves. It was taking the combination of the King’s Leap and a fairly common movement of the Rook and combining them into a single move. A convenience, not a design to give the game a layer of strategy as apparently suggested by O'Shaughnessy. The strategy and tactics of Castling came later, as they will with Chess960 and Chess480. In Vukovic’s book he further states that the King’s starting position in the center of the board is the reason Castling is, in his opinion, nearly a requirement.
Page 50:
“...[the King’s] position in the centre of the back rank stood out immediately as a fault in the game’s structure. On the one hand, it had to be safeguarded against the formidable new forces, and on the other it had to be removed from the centre so as not to hamper them. For this reason it was necessary for the king to get away from the centre as quickly as possible, and it was to this end that the king’s double move was built on and transformed into the full castling move. … We start the game of chess today with the pieces placed in the ancient order which derives from chaturanga, and then we switch over by castling to a new position, which is better suited to the alterations made in the game’s rules.” 2
In 57.5% of Chess960/Chess480 starting positions, the King is already removed from the center of the board and the ancient order derived from chaturanga. One might say that Vukovic’s point invalidates even the need for Castling in these cases, or at least reduces its tactical usefulness. In the other 42.5% of openings, both variations do an adequate job of removing the King from the center. So the question remains, why make the move more complicated than it needs to be?
Another oddity of Castling in Chess960 is that in some positions you must lose Castling rights to one side of the board in order to Castle to the other. The final position #959 is just such a position.
Starting Position #959
rkrnnqbb/pppppppp/8/8/8/8/PPPPPPPP/RKRNNQBB
Note here that for the King to Castle using his a-file Rook he would need to move (and thus losing Castling rights with) the Rook on the c-file. As Rooks are normally the last pieces developed it seems most unnatural to first require its early movement to Castle, but to compound that movement with a sacrifice of Castling to that Rook’s side adds a unnecessary complication. There is never a time in Standard Chess nor in Chess480 where one is required to give up Castling rights for one Rook in order to Castle with the other. This is a peculiarity one will find only in Chess960.
Castling in Chess480: An appeal for sanity ends with a section deriding the mirrored nature of Chess480 making the point that #534 (the mirror of Standard Chess) would play exactly like Standard Chess. This is as obvious as stating that position #518 is, in fact, Standard Chess. It is true that Chess480 roughly doubles your chances of playing a Standard Chess game, but the odds of that are only 1 in 480 and half of those times the advanced Standard Chess player will feel a strange sense of je ne sais quo because the opening is mirrored.
A small price to pay for simplicity.
1. The official FIDE rules for Chess960 suggest that you give the intention of a Castling move. One of two methods have you first removing your King from the board, moving your Rook and finally replacing your King to the board to finish Castling. This avoids confusion about your intentions should you have moved your King first.
2. Emphasis mine. It is worth noting that Vukovic’s points are all predicated on the starting positions of Standard Chess. Almost the entity of The Art of Attack In Chess is dedicated to cracking open the Castled positions of Standard Chess. He implies that Castling isn’t just an option, but almost a necessity in Standard Chess. Would he feel the same about Chess960/Chess480?
Interesting follow-up article, thanks John.