Mate in 4 (Petrovic)

Originally posted by BigDoggProblem
In response to point 2:
Castling is always assumed legal in all chess problems conforming to the FIDE Codex. There are no conditions which need to be met.

That's fair enough. No problems. So then 2. 0-0-0 is perfectly allowed.

But isn't there a difference between legality and obligation?

Just because 2. 0-0-0 is legal, why does that mean you have to do it?

If this were my position in a game (as you have said, it's a legal position), my logic would be: "I can castle here, I'm allowed to. Or I can move to d7 and checkmate. I think I'll go with the latter."

Just because I chose not to castle, why does that reverse its legality in the first place? I can't see why it would, hence it doesn't reverse the legality of the en-passant in move 1?

Originally posted by Ian68
Now you are contradicting yourself.

Not so. I don't have to mention the 'unless provable otherwise' exception every single time that I mention the correct starting assumption about castling. It's been mentioned over and over in this thread already - if it has not sunk in by now, then I can't help you. Sorry!

Originally posted by Zeddicus
That's fair enough. No problems. So then 2. 0-0-0 is perfectly allowed.

But isn't there a difference between legality and obligation?

Just because 2. 0-0-0 is legal, why does that mean you have to do it?

If this were my position in a game (as you have said, it's a legal position), my logic would be: "I can castle here, I'm allowed to. Or I can mov ...[text shortened]... 't see why it would, hence it doesn't reverse the legality of the en-passant in move 1?

You are not obliged to prove castling by playing 0-0-0. You play 0-0-0 to prove that the en passant capture on the previous move was legal. You are obligated to prove the ep capture.

The convention of requiring proof in order to play en passant legally is the most natural convention to have. En Passant opportunities in a real chess game are few and far between - in most positions, you do not have any ep rights, and when you do, they only last for a single move. Castling, on the other hand, can easily remain legal for many moves in a row, sometimes the whole game. The conventions fit the natural availability of these two rights in a game.

You are correct that 2.0-0-0 is not a practical move in a real chess game. However, in a real game, I have the scoresheet and my memory and witnesses as proof of ep and castling rights. In a problem, we do not always know what the preceeding moves were. We have to rely on retro-analysis and/or conventions to guide us.