A Chess paradox

[E Michael White]

[Date "2011.11.25"] [White "A"] [Black "B"] [Result "*"] [SetUp "1"] [FEN "3Q3Q/8/8/3Pp3/3k4/8/8/5K2 w - e6 0 51"] [PlyCount "1"] 51. dxe6+ *

Double check with 2 queens
White could win more quickly with Qc7

[Geoff Chandler]

Hi Michael

I had my own position (same ep theme) ready in waiting for
anyone who said that two Queens cannot give Double Check either.
I'll use yours.

[FEN "3Q4/4p3/8/3P4/3k4/8/8/5K1Q w - - 0 1"] 1. Qhh8+ e5 2. dxe6+*

[E Michael White]

Maybe one of us went off topic. PCAM decision awaited.

[Nick Ivell]

As a junior I was always puzzled by a check given by a pinned piece, for example a rook giving check but itself pinned by a queen. Was it really check, as the rook cannot move? I rationalised it to myself as follows: capture of the king ends the game! So the notional RxK comes before the even more notional QxK...

I probably have not explained myself clearly. As a junior I had some strange ideas!

[Christopher Kreuzer]

As a junior I was always puzzled by a check given by a pinned piece, for example a rook giving check but itself pinned by a queen. Was it really check, as the rook cannot move? I rationalised it to myself as follows: capture of the king ends the game! So the notional RxK comes before the even more notional QxK...

I probably have not explained myself clearly. As a junior I had some strange ideas!

No, it is quite clear. If anyone can be bothered to look it up, I had a position in round 1 of the 4NCL (division 3) where my final move delivered checkmate because Black was unable to interpose with a bishop that was pinned to the Black king by my rook, which was itself pinned to the White king by the Black queen! (I had similar thoughts to yours - was my rook really pinning his bishop if it was pinned itself?) Now there is a challenge - is there a limit to how long a chain of pins (to the king) you can set up in a composed position?

Actually, I'd better give the game - click to the final position to see what I was describing above (if you play through the game, the double knight sacrifice is flashy but not quite sound - I was lucky that my opponent missed that it was mate in the final position):

[Event "4NCL"] [Site "Barcelo Hotel, Daventry, ENG"] [Date "2011.11.12"] [Round "146.5"] [White "Kreuzer, Christopher RL"] [Black "Armour, Philip"] [Result "1-0"] [WhiteElo "1953"] [BlackElo "1908"] [PlyCount "55"] [EventDate "2011.??.??"] [WhiteTeam "Cambridge University 3"] [BlackTeam "Guildford 3"] 1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. c4 Nb6 5. f4 g6 6. Nf3 Bg7 7. Be2 O-O 8. O-O c5 9. dxc5 dxc5 10. Be3 Qc7 11. Nc3 Be6 12. Qb3 Na6 13. Ng5 Bc8 14. Nb5 Qb8 15. Ne4 Nd7 16. Bg4 Nxe5 17. Bxc8 Rxc8 18. fxe5 Qxe5 19. Ng5 Bf6 20. Bf4 Qxb2 21. Qh3 h5 22. Nxf7 Kxf7 23. Nd6+ exd6 24. Bg5 Qd4+ 25. Kh1 Qxa1 26. Qd7+ Kg8 27. Qe6+ Kf8 28. Bh6# 1-0

[IM Jack Rudd]

Cool game, Chris. You interested in annotating it?

[Christopher Kreuzer]

Cool game, Chris. You interested in annotating it?

Possibly... (I really have no time before 1st December). I'd prefer it if someone who knew what is happening at various points annotated it!

[chrisobee]

As a junior I was always puzzled by a check given by a pinned piece, for example a rook giving check but itself pinned by a queen. Was it really check, as the rook cannot move? I rationalised it to myself as follows: capture of the king ends the game! So the notional RxK comes before the even more notional QxK...

I probably have not explained myself clearly. As a junior I had some strange ideas!

No, it is quite clear. If anyone can be bothered to look it up, I had a position in round 1 of the 4NCL (division 3) where my final move delivered checkmate because Black was unable to interpose with a bishop that was pinned to the Black king by my rook, which was itself pinned to the White king by the Black queen! (I had similar thoughts to yours - was my rook really pinning his bishop if it was pinned itself?) Now there is a challenge - is there a limit to how long a chain of pins (to the king) you can set up in a composed position?

Actually, I'd better give the game - click to the final position to see what I was describing above (if you play through the game, the double knight sacrifice is flashy but not quite sound - I was lucky that my opponent missed that it was mate in the final position):

[Event "4NCL"] [Site "Barcelo Hotel, Daventry, ENG"] [Date "2011.11.12"] [Round "146.5"] [White "Kreuzer, Christopher RL"] [Black "Armour, Philip"] [Result "1-0"] [WhiteElo "1953"] [BlackElo "1908"] [PlyCount "55"] [EventDate "2011.??.??"] [WhiteTeam "Cambridge University 3"] [BlackTeam "Guildford 3"] 1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. c4 Nb6 5. f4 g6 6. Nf3 Bg7 7. Be2 O-O 8. O-O c5 9. dxc5 dxc5 10. Be3 Qc7 11. Nc3 Be6 12. Qb3 Na6 13. Ng5 Bc8 14. Nb5 Qb8 15. Ne4 Nd7 16. Bg4 Nxe5 17. Bxc8 Rxc8 18. fxe5 Qxe5 19. Ng5 Bf6 20. Bf4 Qxb2 21. Qh3 h5 22. Nxf7 Kxf7 23. Nd6+ exd6 24. Bg5 Qd4+ 25. Kh1 Qxa1 26. Qd7+ Kg8 27. Qe6+ Kf8 28. Bh6# 1-0

Yes, the point, I believe, is exactly that 28) Bh6 delivers checkmate immediately so at that point the game is over and no number of pieces being pinned would counteract that.
For the game itself I haven't seen 8 )... c5 before but anyone with a better opening knowledge might say if it is a novelty or not ? I would certainly favour 8 )... Be6 or dxe5 in this position. Skipping quickly on to the double knight sacrifice it is fair to say I would probably have played neither ! However, 22) Nxf7 leaves white with a won position after Kxf7 though the more solid 22) Rae1 putting pressure on e7 and, depending on the response by Black, leaving Nxf7 as a potential threat would have been an alternative though 23) Rae1 is still very strong for White.
23) Nxd6+ seems to hand the initiative to Black with 23).........exd6 the simple and best response. White can still put pressure on the Black position but 24) Rab1 looks better than Bg5.
Following the moves 24).... Qd4 + 25) Kh1 Qxa1 26) Qd7 Kg8 27) Qe6 + then Black should simply play 27)... Kh7 and after, for example, 28) Qf7+ Bg7 29) Rxa1 ( I don't seek anything better for White) then 29)... Rc7 leaves Black with a much better position.
Of course 27)............. Kf8 ?? is horrible and loses to the simple Bh6#.
Fair to say it was not a boring game

[Stewart Reuben]

Nick's confusion as a junior is one of the most frequently asked chess questions. His method of solving the 'paradox' is excellent.
There used to be a number of true paradoxes in the Laws.
Until 1984 it was possible to win on time with a bare king. 6.9 took a long time to evolve.
A king could legally remain in check from THREE pieces. That anomaly was pointed out by the readers of Chess Magazine.
Promotion. In 2005 3.7e was changed by adding the word NEW in the sentence a new queen, rook, bishop or knight of the same colour. Prior to that you could have promoted to one of your own pieces already on the board.

I have been asked whether this process is now complete. My response is, I do hope not. However, I do disagree with the last sentence of 9.2. If you can look into the future and see a position is drawn, why not be able to look one move into the future and see that a player will never be able to castle.
That may confuse some. Try this position:
White Qd4 Kc1. Black Ra8 Ke8 Rh8. 1 Qe5ch Kf7 2 Qf5ch Ke8 3 Qe5ch. The Laws say this is the first time this position has occurred because Black could have castled until he moved his king.

[Kevin Thurlow]

"A king could legally remain in check from THREE pieces. That anomaly was pointed out by the readers of Chess Magazine."

One reader actually - me!

"Promotion. In 2005 3.7e was changed by adding the word NEW in the sentence a new queen, rook, bishop or knight of the same colour. Prior to that you could have promoted to one of your own pieces already on the board."

That would be a great way of bringing another piece into the attack.

[Alex Holowczak]

"A king could legally remain in check from THREE pieces. That anomaly was pointed out by the readers of Chess Magazine."

One reader actually - me!

I'm intrigued. Do explain?

[Alex McFarlane]

The wording was along the lines of -
A king is in check if attacked by one or two of the opponent's pieces. Hence if attacked by three it wasn't in check.

The other potential paradox was promoting the king's pawn to a rook and then castling with that new rook!

[Jonathan Rogers]

The rule used to be that a king was in check when attacked by one or two pieces, so if you managed to attack a king with three pieces simultaneously, then you wouldn't be giving check at all. The other guy might just ignore it and checkmate you, for example.

I am trying to remember how it is possible to to this. No doubt Kevin will help. Or Jack.

[Jonathan Rogers]

The former rule permitting promotion to an opponent's piece surely had potential relevance. Imagine a Dragon type position, where White's king was on a2 and a White pawn on b3; Black has a pawn on b4, another pawn on b2, a rook on c2, and Bishop on the open long diagonal, say g7. Black may be in danger of getting mated himself, and has no time to play ...Rc1 or to continue his own attack in other way, but hey presto, b1= White knight would be checkmate!

[andrew martin]

Phew! Who would be an arbiter