Is Retro Zugzwang possible?

Question

What is just a zugzwang is clear. This is when, in a normal good position, a side is forced to make a move, and any move made leads to a loss. Accordingly, in Retro moves are made not forward, but backward. That is, the same thing, but now with the return of the move. Of course, a situation where black does not have the last code that preserves castling can be called a zugzwang. They return the move - lose the right to castling - lose. But it's too easy. And no one, of course, will call this retro zugzwang.

Therefore, let's complicate the conditions of the task:

1. The initial position should not be lost for black.
2. But the return of any move makes it lost.
3. At the same time, the return of the move by black does not lead to any direct consequences. Black does not lose the right to castling, for example.

Is it possible to create such a position?

Answer 1

As an answer, I propose this position.

2B1k2r/Rp1p1p2/prpP1P2/1P2p3/7N/7p/2P3PN/4K2R w - - 1 13

Checkmate in 4 moves

Pure zugzwang. Given the hypothetical possibility of not returning Black's last move, white does not win. But the move must be returned. And any possible return of the pawn move (e7-e5, e6-e5, g4xh3) leads to a lost position for black, even though none of these returned moves leads to the loss of any rights by black. The right to castling is preserved here in all cases. As a result, we have a pure retro zugzwang at the output.

Until the return of the last move, black does not lose, after the return of the move, a checkmate in 4 moves is inevitable. (I'll clarify just in case. Physically, Black's last move does not return. This is not a retractor. We return it only mentally. And the return of any move by black proves that we can checkmate in 4 moves. Without considering the question "What was the last move made by black?" there is no way to checkmate in 4 moves).

Here, white has one overoptimistic attempt by white.

1. Nf5 0-0 (after which checkmate in 4 moves is inevitable). However, it is easily refuted: 1 ... Rxb5 (hxg2).

Black's rejection of the "saving" castling.