Page 1 of 2

Re: Castling long

Posted: Mon Feb 07, 2011 11:09 am
by teddyosbourn
:?

Re: Castling long

Posted: Mon Feb 07, 2011 11:09 am
by Roger de Coverly
Warren Kingston wrote:Have been playing chess a long time, and never knew this (bold bit). Just played a game and this happened and it threw me totally. I had a bishop covering the diagonal for a castle long and this came in to effect.
Korchnoi is reported to have asked the arbiter about the legality of castling under similar circumstances during the 1974 match with Karpov.

Re: Castling long

Posted: Mon Feb 07, 2011 11:14 am
by Jon D'Souza-Eva
I'm astonished that you didn't know that Warren! It's more common for players to mistakenly think that it is illegal to castle long if b1/b8 is attacked (i.e. the rook, but not the king, passes through a square which is attacked). And of course there was the Korchoi vs Karpov game...

In game 21 of their 1974 match, Karpov played 17. ... Bxd5, attacking Korchnoi's rook on h1. Korchnoi was unsure of the rules and asked the controller if he was allowed to castle. Upon being told it was legal he played 18. O-O and Karpov resigned a move later! I suspect Korchnoi knew the rules perfectly well and just did this to wind Karpov up.

[Event "Karpov-Korchnoi Candidates Match"]
[Round "21"] [Result "1-0"]
[White "Korchnoi, Viktor"]
[Black "Karpov, Anatoli"]
1. d4 Nf6 2. Nf3 e6 3. g3 b6 4. Bg2 Bb7 5. c4 Be7
6. Nc3 O-O 7. Qc2 c5 8. d5 exd5 9. Ng5 Nc6 10. Nxd5 g6
11. Qd2 Nxd5 12. Bxd5 Rb8 13. Nxh7 Re8 14. Qh6 Ne5 15. Ng5 Bxg5
16. Bxg5 Qxg5 17. Qxg5 Bxd5 18. O-O Bxc4 19. f4 1-0

Re: Castling long

Posted: Mon Feb 07, 2011 12:26 pm
by Matt Mackenzie
According to Korchnoi, the inquiry was genuine!

He said that he had been playing chess for over thirty years by then (this was, remember, back in 1974!!) and "it had simply never come up before".

To judge by another strong Soviet GM, Averbakh (who [in]famously objected to C J S Purdy castling in a similar fashion) he might just be telling the truth, too :D

Re: Castling long

Posted: Mon Feb 07, 2011 3:13 pm
by IanDavis
So you can castle Queenside as well? :lol:

Re: Castling long

Posted: Mon Feb 07, 2011 3:34 pm
by IM Jack Rudd
There were two instances of illegal castling at Gibraltar this year on consecutive days: in the first, black castled queenside when the white queen was covering d8; in the second, black castled queenside having already moved his king. (The first wasn't spotted; the second was.)

Re: Castling long

Posted: Mon Feb 07, 2011 3:51 pm
by Sean Hewitt
IM Jack Rudd wrote:There were two instances of illegal castling at Gibraltar this year on consecutive days: in the first, black castled queenside when the white queen was covering d8; in the second, black castled queenside having already moved his king. (The first wasn't spotted; the second was.)
Oh dear. Because then you have to move your King, which is probably not pleasant!

Re: Castling long

Posted: Mon Feb 07, 2011 3:58 pm
by Jon D'Souza-Eva
A year or so ago the father of a reigning British Champion asked me why he couldn't enter his daughter's latest game into Chess Base. It turned out that she had moved her king, then later returned it to its starting square and then later still castled! Chess Base wouldn't allow the illegal castling move, which I think is a bug - at club level unspotted illegal moves are surprisingly common.

Re: Castling long

Posted: Mon Feb 07, 2011 4:01 pm
by IM Jack Rudd
Jon D'Souza-Eva wrote:A year or so ago the father of a reigning British Champion asked me why he couldn't enter his daughter's latest game into Chess Base. It turned out that she had moved her king, then later returned it to its starting square and then later still castled! Chess Base wouldn't allow the illegal castling move, which I think is a bug - at club level unspotted illegal moves are surprisingly common.
It's not a bug, it's a feature - the programme is designed to work that way.

The two illegal castling situations at Gib:
http://www.gibraltarchesscongress.com/g ... rtrd4.html
http://www.gibraltarchesscongress.com/g ... rtrd5.html

Re: Castling long

Posted: Mon Feb 07, 2011 7:22 pm
by Michael Jones
Tim Krabbé's site (http://www.xs4all.nl/~timkr/records/records.htm) mentions an instance of a player castling on both sides in the same game. His opponent apparently didn't realise anything was amiss.

Re: Castling long

Posted: Tue Feb 08, 2011 4:50 pm
by Richard Thursby
Some people have tried to argue that for example, if white promotes a pawn to a rook on e8 then it is possible, subject to the usual restrictions, to castle by moving the king to e3 and the rook to e2. I don't agree with this, since the pawn has been promoted to a rook, not replaced by a rook (even though that happens in practice on the board), so the rook on e8 was originally a pawn on the second rank so has moved to e8 and hence can't be used for castling. Apart from the diagrams, the laws of chess (2009) don't rule this out.

I believe the approved method of entering an illegal move in chessbase is to produce two (or more, as necessary) separate files for the game.

Re: Castling long

Posted: Tue Feb 08, 2011 5:22 pm
by IM Jack Rudd
The text of the rule on castling states:
The FIDE Laws of Chess wrote:This is a move of the king and either rook of the same colour along the player’s first rank, counting as a single move of the king and executed as follows: the king is transferred from its original square two squares towards the rook on its original square, then that rook is transferred to the square the king has just crossed.
(Emphasis mine. That clause is in there specifically to prevent castling with a promoted rook.)

Re: Castling long

Posted: Tue Feb 08, 2011 6:54 pm
by Justin Hadi
I heard some titled player won a bet based on that trick before the rules were changed. Probably somebody mentioned it to FIDE rather than an attenpt to castle like that in a real game.

Re: Castling long

Posted: Tue Feb 08, 2011 7:08 pm
by Christopher Kreuzer
IM Jack Rudd wrote:The text of the rule on castling states:
The FIDE Laws of Chess wrote:This is a move of the king and either rook of the same colour along the player’s first rank, counting as a single move of the king and executed as follows: the king is transferred from its original square two squares towards the rook on its original square, then that rook is transferred to the square the king has just crossed.
(Emphasis mine. That clause is in there specifically to prevent castling with a promoted rook.)
Does that wording still leave open the possibility of castling with a rook of your own colour, created by promotion of one of your opponent's pawns on the right square? This being, of course, before FIDE changed the rules to make clear that you can only promote to pieces of your own colour... (spoilsports!).

Re: Castling long

Posted: Tue Feb 08, 2011 9:21 pm
by Andrew Bak
Richard Thursby wrote:Some people have tried to argue that for example, if white promotes a pawn to a rook on e8 then it is possible, subject to the usual restrictions, to castle by moving the king to e3 and the rook to e2. I don't agree with this, since the pawn has been promoted to a rook, not replaced by a rook (even though that happens in practice on the board), so the rook on e8 was originally a pawn on the second rank so has moved to e8 and hence can't be used for castling. Apart from the diagrams, the laws of chess (2009) don't rule this out.

I believe the approved method of entering an illegal move in chessbase is to produce two (or more, as necessary) separate files for the game.
I seem to remember that Sam Loyd produced a problem using this ambiguity in the rules, but I unfortunately have no source for this, I just have my dodgy memory...