Find the game that led to...

Originally posted by Silent Knight
This beautiful problem reminds me of a book by Raymond Smullyan, "The chess mysteries of Sherlock Holmes". I haven't seen it mentioned before on RHP. In it the great detective solves several mysteries using what he calls "retrograde analysis". Here's a fairly easy example:

[fen]k1K5/8/8/8/8/8/7P/6B1[/fen]

"As you see, Watson, neither side is ma ...[text shortened]... hite's last move?"

Even Watson gets it fairly quickly. Other problems are trickier.

Originally posted by cmsMaster
Black, Kb7-a8, white, Kd7-c8

Originally posted by cmsMaster
Black, Kb7-a8, white, Kd7-c8

Originally posted by Mephisto2
Not quite. The black king could not have come to a7 this way, allowing him to make Ka7-a8 as last move (the bishop on g1 would prevent him from doing that).

Originally posted by cmsMaster
The black king was never on a7. Hence I said king on b7 to a8, which seems to work..?

Black moved last so it must have been Ka7-a8 and therefore white must have put him in check with his previous move. The bishop can't move to give check so the only alternative is a discovered check. No piece remains and the king can't have moved to give the discovered check so the piece must have moved to a8 and been taken.

So:

0. Nb6-a8+ Ka7xa8

A simple example once you know the trick to retrograde puzzles and a good introduction to the kind of thought that goes into them.

Originally posted by XanthosNZ
Black moved last so it must have been Ka7-a8 and therefore white must have put him in check with his previous move. The bishop can't move to give check so the only alternative is a discovered check. No piece remains and the king can't have moved to give the discovered check so the piece must have moved to a8 and been taken.

So:

0. Nb6-a8+ Ka7xa8

A ...[text shortened]... rick to retrograde puzzles and a good introduction to the kind of thought that goes into them.

Well evidently the black King moved into the corner in the last move, and since the black field he moved from is covered by a bishop, which can't be moved where it stands, a whitw pawn has been transformed to a bishop.

so g2g1B+, Ka1

so we can go without the Knight 🙂

edit: I omitted the check(+)...

Originally posted by Ponderable
Well evidently the black King moved into the corner in the last move, and since the black field he moved from is covered by a bishop, which can't be moved where it stands, a whitw pawn has been transformed to a bishop.

so g2g1B+, Ka1

so we can go without the Knight 🙂

edit: I omitted the check(+)...

ok. Spelling mistak. It should have been g7g8+

And I have been unaware of the convention in relation to the FEN...

To the best of my knowledge, FENs did not exist at the time of Holmes and Watson. But notice that in his description of the problem, Holmes says, "And we are given that your side is White". I.e., we are looking at the board from White's point of view.

After solving the problem, Watson asks if it was really necessary to know which side was White. Holmes explains that, as noted, if it is not specified there is another solution.

White retracts his last move and mates in 2.

What if white retracts his last move, and has mate in two with the same move as first move? I guess that completely fulfills the request, no?

Originally posted by Mephisto2
What if white retracts his last move, and has mate in two with the same move as first move? I guess that completely fulfills the request, no?

Referring to the very first post and example. . .Why can't the pawn becoming a bishop on the last rank have put the black king in check? Thus, move #1 could be Kh8 and move #2 could be Kf8?