Re: MATH : Fermat's Last Theorem: did Fermat have proof or did he guess?

This is definitely not my field of expertise, but I am pretty sure that it is the consensus of the experts that Fermat was probably mistaken. First, the problem has received an enormous amount of attention over the centuries from the very best mathematicians. The fact that none of them found a short proof (supposing Fermat's proof was not much longer than the margin would hold -- this is suggested by Fermat's description of it as "truly wonderful"), nor indeed any proof at all, makes one wonder. Second, it seems clear that if Fermat had a proof, it must have been quite different than the one we have today. For the prerequisites, the mathematical concepts used in the proof, on which Wiles was able to build, had not even been developed in Fermat's time. Third, there are some shorter partial proofs, and I think even some short, seductive, but mistaken ones, that have been discovered over the years, some by good mathematicians. Possibly Fermat's proof was one of these.

Mathematicians do often guess, but of course a guess is never a proof! The inspired guess leads the way, motivates and guides the hard struggle to construct a rigorous proof, but no honest mathematician would ever knowingly say he had proved something that he had only guessed. But a strong hunch can lead you to believe a conjecture is true, and then it is not too uncommon to overlook subtle logical flaws in the proof constructed to establish the guess beyond all doubt. Wiles himself at first fell victim to such an error, which, fortunately, he was able to repair.

But it is hard to see how we can ever know for sure. Perhaps someone will discover a short proof next year that uses only prerequisite concepts that Fermat had or might reasonably have developed on his own. If so, it seems that it would have to be so original and unexpected that it might well lead to further insights and progress in the whole field of mathematics.