[Retros] Retrograde analysis in other forms

[Noam Elkies]

Gianchess at aol.com writes:


> Apparently this is the missing diagram:

>  

> http://www.chess.com/forum/view/more-puzzles/an-old-puzzle2


Also shown in the stackexchange thread that Andrew already linked to:

http://chess.stackexchange.com/questions/178/is-there-fiction-dealing-with-retrograde-chess-analysis

But is that position correct?  White is missing the a-pawn and
both Rooks, though only one of them could have been captured
far from its starting corner; and Black is missing two Knights,
which must have been captured on c3 and e5 because Black has
not enough spared captures for the d- and e-pawns to reach
d6 and e5 except by moving vertically (and because Black has
all his pawns so the captures cannot be of promoted pieces).
The intention is for White to promote on a8 to replace a piece
captured on b6, which must be the wQ because a Bishop or Knight
couldn't escape from a8.  But why couldn't Black have captured
the original wRa1 on b6, and taken the a-pawn somewhere (possibly
after it promoted)?  I don't see any evidence that the wRa1 is
needed for some other role.  Are we told something else about
the position that would eliminate that possibility?

NDE

P.S. unlike the typical Smullyan puzzler this one has somewhat
interesting forward play as well.