16

According to the FIDE Laws of Chess:

4.4 If a player having the move:
4.4.1 touches his king and a rook he must castle on that side if it is legal to do so
4.4.2 deliberately touches a rook and then his king he is not allowed to castle on that side on that move and the situation shall be governed by Article 4.3.1

Yesterday in the semi-finals of the Chess960 World Championships in the following position Nepomniatchi tried to castle kingside by first picking up his rook to move it off the king's destination square (g1) to f1, the rook's destination square and then moving the king to g1.

The arbiter ruled in accordance with 4.4.2 that he was not allowed to castle kingside and the king must stay on e1 and he got a bad position. Later, after an appeal, this was overruled and they replayed the game.

 [title "Neponiatchi - So, 2019 Chess960 1/2 final"]
 [fen "2nr1r1k/pb1n1pqp/1p2p1p1/1N1pN3/5P2/1P4P1/P2PPQ1P/2R1K1RB w - - 0 1"]

Jonathan Tisdall, on Twitter, reported that:

According to organizer Jøran Aulin-Jansson, the rook being on g1, the king destination, caused enough confusion to overturn the strict ruling

This raises the question: In such situations how can the player legally castle on the side where the rook starts on the king's destination square?

One humorous, though impractical at short time limits, solution was suggested here.

Even more impractical was MrDodgy's suggestion:

A true pro would pick up the rook from g1, toss it up in the air, slide the king to g1 and the rook would land dead centre of f1. Press the clock, lean back in chair. Flex.

Still illegal.

  • 2
    In all fairness though, MrDodgy’s suggestion is pretty funny... ;D – Rewan Demontay Oct 30 at 11:25
  • 2
    How would it have been legal to castle in the first place? The Rook wasn't on its standing position. – vsz Oct 31 at 7:06
  • 8
    @vsz Chess960 starts the game with the back row somewhat scambled, it is assumed that in this game the two pieces have not yet been moved from their starting positions. – Jasen Oct 31 at 7:36
  • Castling in chess960 always looks odd. First time I encountered it, I was surprised it was even legal. – Mast Nov 1 at 17:00
  • When capturing a piece, I will usually pick up my own piece, and simultaneously place it on the square of the piece I'm capturing while picking up the piece I'm capturing. I can't see any reason the same action can't be employed to place the king on the square occupying the rook while simultaneously picking up the rook – Darren H Nov 2 at 15:19
17

The FIDE Laws of Chess actually cover this with a recommendation without mentioning the problem.

II.3.2.5.1 When castling on a physical board with a human player, it is recommended that the king be moved outside the playing surface next to his final position, the rook then be moved from its starting position to its final position, and then the king be placed on his final square

The first time I read this I though "What a daft thing to suggest? Who is going to do that?" but now I see the point.

The FIDE Laws of Chess go on to state:

II.3.2.7 Notes
II.3.2.7.1 To avoid any misunderstanding, it may be useful to state "I am about to castle" before castling.

Suppose Nepomniatchi had read that before and said those words before castling, would the arbiter have allowed him to castle the way he did?

So, it looks like this was thought of when the rules were written and they were written accordingly.

For practical play there is a simpler solution. According to 4.4.1 touching the king then the rook must be followed by castling. In cases where either the rook is on the king's square or the king is on the king's square just touch king then rook and then make the move. This solves all the associated problems.

  • 3
    off topic, but why would you wait? If you know the answer, just post it – Ant Oct 31 at 20:12
11

Touch your king, then your rook. By rule 4.4.1, you are now required to castle. You can now move the king and rook to their proper squares however you like, since you are already forced into a specific move and any subsequent piece touches can't change that.

9

I saw this in the chess news on a couple of sites, and it seems pretty simple to me. Pick up the K, and move it to g1, simultaneously pushing the rook off the g1 square. Then move the rook to its proper square.

In any case, you should pick up the K first, and then if you also want to pick up the rook, grab it second, and move them simultaneously. Either way, you are in compliance.

  • 3
    How would you move them simultaneously? Using two hands to castle is illegal. – Federico Poloni Oct 30 at 23:12
  • 5
    @FedericoPoloni with a little finesse you can manipulate two pieces with one hand, without dropping anything at the wrong moment :) – hobbs Oct 31 at 0:53
  • 1
    @hobbs Great, now clumsy people are more likely to make an unintentional blunder while castling... – Michael Oct 31 at 2:26
  • 5
    Exactly this. You do it just as you would when capturing a piece. Pick up your own piece, place by opponents piece, pick opponents piece up and finishing your one handed movement, reposition your piece in the center of the square. The only added move is now positioning the "captured" piece at its place, not beside the table. (And adjust with j'adoube if necessary before clocking) – Stian Yttervik Oct 31 at 9:33
  • 1
    @Michael On the scale of coordination demands in the world of game and sport, holding two pieces in one hand seems a pretty small ask. – J... Oct 31 at 19:30
4

There are two problematic cases in Chess960 castling:

(1) rook starts on king’s destination square.
(2) king starts on king’s destination square.

It is not beyond the wit of man for the FIDE Laws to handle those cases explicitly. Maybe a future version of the rules will do so.

However the most important FIDE rule here is:

Too detailed a rule might deprive the arbiter of his freedom of judgement and thus prevent him from finding the solution to a problem dictated by fairness, logic and special factors.

It must be difficult to rule in the excitement of such a situation. But had not GM So generously agreed to draw the game, the ruling could have effectively terminated the semi-final. The moral for players is to remain calm, and exercise opportunities for appeal.

  • Isn't the king's destination square two squares in the direction of the rook? – Peter Taylor Oct 31 at 8:13
  • @PeterTaylor In 960, not necessarily. In some cases that will be off the board! – D M Oct 31 at 11:17
  • @Peter Taylor, starts on b file to g file but castles just to c file or g file - i.e. the same squares as regular chess! – Laska Oct 31 at 14:50
2

Stian Yttervik made a comment that I think deserves to be expanded into a full answer. (I have changed the wording somewhat. That is not Stians fault)

Just pretend that the King is capturing the Rook.

In other words: Move the King first. Use the King to gently push the Rook off-center and firmly place the King where it should be. Then move the Rook where it should be. Hit the clock.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.