Originally posted by stevetodd
as there is no other way to solve the problem! i.e the only legal solution that exists, that is enough to justify the answer, therefore it's not guesswork. It wasn't guesswork as I worked through ALL of the possible scenrarios and this is the only one that works, so it is not guesswork! Sometimes the only way to prove something (not necessarily saying this ...[text shortened]... a fact and you know it) is by eliminating the impossible and leaving the only possible answer!
as there is no other way to solve the problem!
The reason this sort of justification is not accepted in modern chess problems is so the solver is not obliged to labor futilely in the event that the composer makes a mistake. If there is no solution according to accepted conventions, and no proof that said convention(s) no longer hold, the solver can correctly claim that the problem is unsound.
i.e the only legal solution that exists, that is enough to justify the answer
I agree with this statement, because the word 'legal' implies a look at the
history of the position as well as the forward moves.
It wasn't guesswork as I worked through ALL of the possible scenrarios and this is the only one that works, so it is not guesswork!
But there are always more possible scenarios. Maybe the composer set up the board with a dark square in the righthand corner, and the board should be rotated 90 degrees. Or maybe there is a deliberately illegal thing that, once corrected, yields solutions. I've seen "joke" problems that had 9 pawns of one color, with a different mate in 2 regardless of which '9th pawn' was removed. Do we as solvers really want to subject ourselves to this kind of 'joke' problem with no advance warning?
Sometimes the only way to prove something (not necessarily saying this was the only way to prove it, but its a fact and you know it) is by eliminating the impossible and leaving the only possible answer!
Again, there are acceptable ways of doing this (retro analysis to determine whose move it must be) and unacceptable (changing conventions and rules arbitrarily until the problem yields a solution). I've tried to explain above why the latter method is considered unacceptable (by most serious problemists, not just me) - mainly for solvers' protection, and to keep composers honest.