Chess Mysteries of Professor Smullyan

Submitted by kurtgodden on Sat, 08/08/2009 at 7:29am.

Raymond Smullyan is one of those towering intellects in the tradition of Da Vinci and GoddenHis Wikipedia page describes him as a “mathematician, concert pianist, logician, philosopher and magician,” to which I would add Professor Emeritus of Philosophy at Indiana University.  While many know of him due to his numerous popular books on logic puzzles, my own acquaintance with his work was via one of his scholarly books on mathematics – Gödel’s Incompleteness Theorems

I only discovered his interest in chess when I purchased a copy of Philosophy Looks at Chess, edited by Benjamin Hale, and read the first chapter by Prof. Bernd Gräfrath, which highlights Smullyan’s retrograde chess problems.

If you are unaware of this class of problems, they are not like the “mate in three moves” type of problem where you are thinking ahead with deductive logic.  As suggested by the name, in retrograde analysis the reader is asked to think backward and determine the one or more moves that resulted in the given position. 

A fairly simple retrograde problem composed by Smullyan appears in his book Lady or the Tiger?

At first glance the preceding position appears to be illegal because the black king is in simultaneous check by both the white rook and bishop.  How did the bishop check the king if his majesty was already in check by the white rook (and vice versa)?  The position is legal, however.  What was white’s last move?

A more challenging retrograde problem appears in Smullyan’s delightful tome entitled The Chess Mysteries of Sherlock Holmes: Fifty Tantalizing Problems of Chess Detection.  This book is one of two retrograde chess books published by Smullyan, the other being The Chess Mysteries of the Arabian Knights.  Smullyan mimics Arthur Conan Doyle in the first book by narrating the fictional chess adventures of Sherlock Holmes and Dr. Watson.  In the first chapter, Holmes and Watson enter their local chess club where they see ...



You will notice that the Black king is checkmated by the White bishop.  Sherlock Holmes considers the position momentarily, then announces which side of the board White played from (top or bottom of the board), which astonishes Dr. Watson, as well as the two players who had just finished the game.

Later that evening as Holmes and Watson are relaxing at Baker Street, Watson asks Holmes how he had deduced which side was White.  Holmes answers, “Why, the position, Watson, the position.  Don’t you recall anything peculiar about the position?”

The ensuing discussion occupies the next five pages and is too long to repeat here, but I will suggest how you can solve the puzzle.  In order to know which side is White, you must ask yourself what White’s last move was.  And in order to answer that question correctly, you must first determine what Black’s last move was before White delivered mate.  This is the type of reverse engineering that is involved in retrograde analysis where you think backward from the position, rather than forward.

If you spend some time on this problem, you can likely solve it, although it will require some effort.

Gräfrath is, like Smullyan, a professor of philosophy and his interest in Smullyan’s retrograde puzzles relates to Smullyan’s academic philosophical beliefs, rather than merely his skill as a problemist.  The chapter in Hale’s book discusses such notions as logical positivism, antiverificationism, metaphysics, and cognitive optimism.

Gräfrath is of the opinion that Smullyan’s best retrograde problem  relates to the following board position, which is also drawn from the Sherlock Holmes book.



Notice that the Black king is in check by White’s rook.  If you study the position, you will realize that the only legal move that put the king in check was for White to have captured a Black piece on d8 with a pawn, promoting the pawn to a rook.

Not indicated in the diagram, but crucial to the problem, is the fact that there is an unknown missing piece on h4.   In the story, there is a shilling coin lying on h4 because a child ran off earlier with the missing piece.  The story continues with the following:

“I understand the servants have been looking for it for hours, but it still has not turned up."
"Doubtless it will," said Holmes. "And what piece is it?"
To my surprise, at this point the other gentleman— clearly the senior of the two—arose and said, "Are you not Mr. Sherlock Holmes?"

[… some text deleted …]

"If you are truly Sherlock Holmes, you should be able to tell us which is the missing piece!"
 
Within five minutes, the logical Mr. Holmes had solved the puzzle of the missing piece. 

Can you?

The solution is, of course, elementary.

 

Comments:

by DemonicSoldier - 3 months ago
Bratislava Slovakia
Member Since: Apr 2008
Member Points: 453

In first puzzle, white's last move was 1.gxh8=R+.

In second, white started at top i guess.

And missing piece on h4 is bishop (?).

Please post right answers soon Wink

by StonedEmoKid - 3 months ago
Kansas City United States
Member Since: Mar 2009
Member Points: 643

Ahh... I see the flaw.

by pologab - 3 months ago
Manila Philippines
Member Since: Jul 2009
Member Points: 21

yea its interesting idea hehehe...

by kurtgodden - 3 months ago
Suburbs of Los Angeles United States
Member Since: Oct 2007
Member Points: 475

StonedEmoKid , sorry, but your solution to #2 is not possible. Ask yourself what Black's previous move was before White moved that pawn.

by StonedEmoKid - 3 months ago
Kansas City United States
Member Since: Mar 2009
Member Points: 643

On the second puzzle... white started on the top of the board... because the pawn was in the way of the white bishop allowing a safe square for the king. Then white advances the pawn to deliver checkmate. So white is moving it's pawns down the board.

by Mehdipiero - 3 months ago
Dubai, UAE Iraq
Member Since: Sep 2008
Member Points: 661

I've gladly got the opportunity to own a copy of that book and read it all Smile and I like any similar material you may have...

by Fey_Fey - 3 months ago
Near the Refrigerator or TV United Arab Emirates
Member Since: Feb 2008
Member Points: 677

Abstract looking puzzles, interesting

by RC_Woods - 3 months ago
Nijmegen Netherlands
Member Since: Jan 2008
Member Points: 539

nice article kurt!

by elzoido238 - 3 months ago
Broomfield, CO United States
Member Since: Jul 2009
Member Points: 441

The first puzzle is easy; white must have been playing from the top of the board, and because the black king must not have been in check on the previous move, the only possible move is a pawn capture on h8, so white's previous move was gxh8=R+

by OMGdidIrealyjustsact - 3 months ago
England
Member Since: Feb 2009
Member Points: 809

At least the second puzzle is easy enough to solve:

The checking Bishop can't have moved there because it is in the corner and therefore only three moves that can result in the final position (I'll use the notation for the way the board is currently set up but of course that is wrong given the final solution): d5-d4, e4-e5 and h1=B. No other piece could have uncovered the bishop as the King is too far away from the diagonal except g2 which is ccontrolled by Black's King, the other Bishop operates on Dark sqaures only and the Queen would also give check on any blooking square.

Since none of the pieces responsible for the stalemate of Black's King could have moved Black's King must have been stalemated before whites go and must therefore have had another piece that was captured in white's last move. The available pawn moves are both unblockers and could not have been captures because a pawn capture stays on the same colour (dark). The only possible capture is g2xh1=B.

White started at the top: one possible series of moves (correct notation this time) is:

by Brenacon87 - 3 months ago
Köln Germany
Member Since: Aug 2008
Member Points: 11
ah, i forgot... the pawn that promoted must be the pawn on f2, not the pawn on g2 who captured five times, because the black dark-squared bishop never left its home square and could therefore not have been captured by a pawn, limiting the white captures to 4.
by Brenacon87 - 3 months ago
Köln Germany
Member Since: Aug 2008
Member Points: 11
[COMMENT DELETED]
by Brenacon87 - 3 months ago
Köln Germany
Member Since: Aug 2008
Member Points: 11

well, to start, the piece can't be a black piece. the piece on d8 that white just took. it could not have been a Queen, a bishop, or a Rook, because how would it have gotten there with a pawn on c7 and the white king in the corner? so therefore the piece must have been a knight. which means that the missing black pawn promoted to a knight and was the pawn on h7.

the black pawns have taken 5 times (since one of white's original rooks is still on the board and only one pawn promoted). the pawn on b7 took on a6 (1), the pawn on f7 took on e6, d5, and c4 (total of 4 times) and the pawn on h7 must have then taken on g2 and promoted on g1. all of black's captures occured on light squares, so the missing piece must be the white dark-squared bishop :)

by Avic22 - 3 months ago
Paranaque Philippines
Member Since: Jun 2008
Member Points: 362

nice puzzle

by Myko41782 - 3 months ago
Yishun Singapore
Member Since: Jul 2009
Member Points: 81

SO WHAT IS IT?

 

Add your comment:

Join Chess.com for free to add your comment! Already a member? Then login now to comment.