Christmas puzzle solutions

by Frederic Friedel
1/12/2020 – Last year we rivived our twenty-year tradition of giving you unusual puzzles for the Christmas holidays — puzzles that cannot be easily solved with a computer (the point of this endeavour), tasks which require you to think all by yourself. Sadly the greatest problematist, Pál Benkö, who faithfully contributed to this endeavour, could only make one contribution, for Christmas day, before passing (at 91 last August). We hope you enjoyed the puzzles and were able to solve some yourself.

ChessBase 15 - Mega package ChessBase 15 - Mega package

Find the right combination! ChessBase 15 program + new Mega Database 2020 with 8 million games and more than 80,000 master analyses. Plus ChessBase Magazine (DVD + magazine) and CB Premium membership for 1 year!


Christmas Puzzle Solutions

Originally we planned to provide the solutions at the bottom of each puzzle page, around two weeks after publication. Not a good idea: it forces readers to keep returning to the original pages to check if the solutions have appeared. So today we bring you all the solutions on a separate page. You probably know full well that you can move pieces on our chess diagrams, and that the replayer will in fact allow you to start a chess engine (fan button) to assist you in your analysis. We wish you enjoyment with the solutions and look forward to your thoughts on these unusual puzzles.

December 25th 2019: Pal Benko's last Merry Christmas Puzzle

In July 2019, shortly before his passing, Pal Benko, who has always helped me with my Christmas Puzzles, sent me a first position for the 2019 season. He advised me to start with a famous problem, one that looks easy but needs some thinking outside the box.


The trick with the solution is: if it is White to move the previous move by Black must have been with one of the rooks or the king. If it was with the rook on a8 then Black cannot castle queenside and is mated with 1.♕g5 ♚f8 (or 1...♚d8 2.♕d5 ♚c8 3.♕xa8#) 2.♕xe7 Kg8 3.Qf7#. If on the other hand it was the rook on h8 that last moved then 1.♕d4 ♜g8 2.♕d7+ ♔f8 3.♕xe7# fulfils the task. But if we assume that it was the black king that last moved, then neither castle is possible, and White can mate in three with three different key moves: 1.♕c5, 1.♕g5 and 1.♕d4.

During a recent visit to Hamburg the 12-yearl-old chess prodigy Dev Shah, inspired by Pal Benko's beautiful three-mover, decided to try composing chess studies. We gave you two examples which Benko had evaluated: "Good initial effort by the boy!"

Young talent Dev Shah working with endgame expert GM Karsten Müller.

Solutions to Dev Shah's problems:

December 26, 2019: What were the previous moves?

During the Christmas puzzle week we present you with puzzles that cannot be easily solved with a computer (the point of this endeavour). We give you tasks which require you to think all by yourself. On "boxing day" we had problems in which you were tasked with finding the previous move(s).

Solutions to the problems:

December 27, 2019: Réti for beginners

On a train trip I met two young boys, chess novices who had become bored playing each other. So I gave them a few simple (but famous) positions and taught them the Reti-manoeuvre. They were delighted. Here are the solutions to the studies I showed them:

December 29, 2019: Missing pieces

The following twin problem were by the famous English problemist Thomas Rayner Dawson. The two positions are clearly mirror images of each other, but for concrete reasons they do not have mirrored solutions. Could you tell why? Dawson was original if anything, and this is a puzzle he posed in 1927.

In each of these two positions the white queen has fallen off the board. Your task was : replace the queen and stalemate Black in one move. Solution: since the pawns are undisturbed the white queen must have, in both cases, fallen off from a square to the left of the king (how could she have got to the right?) So: in the first diagram the queen should be replaced on the square e1 and the move 1.f4 stalemates the black king. In the second diagram the queen is replaced on a1 and the move 1.a4 delivers stalemate. Were you able to figure this out?

Missing Kings

Here are more missing pieces. In the first diagram (author unknown) both the kings have fallen off. Put them back on the board in such a way that White, on the move, can deliver immediate mate, i.e. mate in one move. Fairly simple: place the black king on h1 and the white king on f3. Then White can play 1.♔xf2#.

In the second problem by Dr Karl Fabel, Deutsche Schachblätter 1950, the black king has fallen off. You should place it somewhere on the board so that White can immediately mate in a single move. Solution: you may have been tempted to place the king on c1 and castle. But the black king can only have got to c1 if the white king had moved out of the way – and in that case White cannot castle. The correct solution is to place the black king on f3 and then castle. Perfectly legal.


Finally, we had two problems by Karl Fabel. In the first you had to construct, by entering legal moves for White and Black, who are naturally cooperating, a position in which Black has only the king on a4 and White has all his pieces on their original squares. The second is a study that is longer than expected. Here are the solutions:

Would you believe it — in the second simple-looking position White needs no less than 47 moves to win? Karl Fabel's problems are always ingenious, full of surprises.

Editor-in-Chief emeritus of the ChessBase News page. Studied Philosophy and Linguistics at the University of Hamburg and Oxford, graduating with a thesis on speech act theory and moral language. He started a university career but switched to science journalism, producing documentaries for German TV. In 1986 he co-founded ChessBase.
Discussion and Feedback Join the public discussion or submit your feedback to the editors


Rules for reader comments


Not registered yet? Register

Frits Fritschy Frits Fritschy 1/14/2020 12:03
By the way if anyone interested, the solution to my own little problem in the comments on 26/12: The last seven moves could only have been 1 Ba2-/xb1+ f7-f5 2 g5xf6 e.p.+ Kg6xf6 3 d4-d5+ e7-e5 4 d5xe6 e.p.+. Of course that the first move could also have been a (any) capture, is a flaw. This can be helped bij removing the rook on g1 and adding pawns on a2 and e4, the solution now being 1 e4-e5+ f7-f5 2 e5xf6 e.p.+ Kg6xf6 3 d4-d5+ e7-e5 4 d5xe6 e.p.+. Sorry, can't do any better.
Frits Fritschy Frits Fritschy 1/13/2020 11:27
If someone doesn't understand a joke, you shouldn't explain it. You just get it, or you don't.
brian8871 brian8871 1/13/2020 06:22
For those who are interested in the problem mentioned but not shown here, the setup is 8/2p5/8/8/1p1p1p2/1P1P1P2/8/2kBK3--White to play and mate in 47. The solution is listed below.

1. Ke2 Kb2 2. Kd2 Kb1 3. Bc2+ Kb2 4. Kd1 Ka1 5. Kc1 Ka2 6. Bd1 Ka1 7. Kc2 Ka2 8. Be2 Ka1 9. Bf1 Ka2 10. Bh3 Ka1 11. Bg2 Ka2 12. Bf1 Ka1 13. Be2 Ka2 14. Bd1 Ka1 15. Kc1 Ka2 16. Bc2 c6 17. Bd1 Ka1 18. Kc2 Ka2 19. Be2 Ka1 20. Bf1 Ka2 21. Bh3 Ka1 22. Bg2 Ka2 23. Bf1 Ka1 24. Be2 Ka2 25. Bd1 Ka1 26. Kc1 Ka2 27. Bc2 c5 28. Bd1 Ka1 29. Kc2 Ka2 30. Be2 Ka1 31. Bf1 Ka2 32. Bh3 Ka1 33. Bg2 Ka2 34. Bf1 Ka1 35. Be2 Ka2 36. Bd1 Ka1 37. Kc1 Ka2 38. Bc2 Ka1 39. Bb1 c4 40 dxc4 d3 41. Bxd3 Ka2 42. Kc2 Ka1 43. c5 Ka2 44. c6 Ka1 45. c7 Ka2 46. c8=Q Ka1 47. Qa8#
JNorri JNorri 1/13/2020 02:18
The above Cant Castler by Loyd is from 1888; he had shown the idea in very basic form in 1859. The en-passant was pioneered by C. Schmidt in 1852 in an illegal position and 1858 correctly by Max Lange. Many others joined in, before Loyd added some influential problems.
lajosarpad lajosarpad 1/13/2020 12:26
@Grey Hiker you are right in stating that at chess problems we may assume that castle is possible unless it is provably impossible. However, in the first problem we have to evaluate the possibility of two separate castles. Let's take the rule of assuming that castling is possible. If we apply this rule for the queenside castle, then, under that assumption we can prove that the kingside castling is impossible. If we apply this rule to the kingside castle, then, under the assumption that the kingside castle is possible we can prove that the queenside castle is impossible. So, if we apply the rule that assuming that a castle is possible if it cannot be proven that it's impossible we reach a paradox and, strictly abiding by the rule of assuming that a castle is possible unless provably impossible we can only assume that the last move was not done by the king. So, we have two cases, both are chess problems on their own: the first is mate in tree, knowing that Black can castle queenside but not kingside, the second is mate in three, knowing that black can castle kingside but not queenside.
Frederic Frederic 1/13/2020 09:23
Samuel Loyd, one of the greatest puzzlists in the history of mankind, pioneered many different kinds of chess problems. It was Sam who brought the legality of castling to the full attention of problemists, in his own charming way. This has led to the "Codex for Chess Composition" Article 16.1 which states: "Castling convention: Castling is permitted unless it can be proved that it is not permissible" (and ultimately thanks to Loyd we have codex 16.2 which says: "An en-passant capture on the first move is permitted only if it can be proved that the last move was the double step of the pawn which is to be captured"). Nobody should be demeaning the man with trite remarks like "complete nonsense" -- Loyd's castling problem has enchanted generations of chess players and problem fans.
Zvi Mendlowitz Zvi Mendlowitz 1/13/2020 08:13
1. About Loyd's problem - a somewhat similar problem: find a position where you can prove a player castled, but you can't prove which player castled.

2. About Fabel's 16 pieces vs. King - another problem: White: Ke5; Black: Ke2. Position after White's 20th move. How did the game go?
Joshua Green Joshua Green 1/13/2020 03:28
@Grey Hiker, the first problem is not complete nonsense -- one is allowed to make deductions. Here one can prove that at least one of the castlings is illegal. True, one cannot know which is illegal from just the diagram, but the "Partial Retrograde Analysis (PRA) convention" covers this type of case -- we essentially have two problems, one in which Kingside castling is legal, one in which Queenside castling is legal.
Grey Hiker Grey Hiker 1/13/2020 12:47
The first problem is complete nonsense since it violates a basic rule of problems: castling is always permitted unless castling can be proven to be illegal. There are many problems where castling can be proven to be illegal but this isn't one of them. So: 1 Qg5, 0-0-0 and there is no mate (the last move could have been by the King Rook). Or 1 Qd4, 0-0 (legal since the last move could have been by the Queen Rook) and again there is no mate.

Basically the composer is just cheating: the solution is totally dependent on an unannounced change to the rules of chess/problems. You might as well claim there is a mate in one by 1 Q:e7 because the composer (without telling anyone) has changed the rules of chess/problems to allow the Queen to jump over pieces.

In other words, White can't prove which Black piece moved last (K, R, R) and therefore isn't entitled to assume that it was any specific one of them.