[thorel]

Agreed, thanks for the clarifications. Another result worth mentioning, which also shows that you cannot hope to uniquely characterize a structure by "reasonable" axioms is the Löwenheim–Skolem theorem which predates Godel's incompleteness (although the history of these results is somewhat convoluted).

There, the obstacle is in some sense of a simplest nature: if your set of axioms admits a countable model, then it admits models of all infinite cardinalities. In other words, it shows that there is something fundamentally impossible in trying to capture an infinite structure (like numbers) by finite means (e.g. recursively axiomatizable).

[ginnungagap]

> In other words, it shows that there is something fundamentally impossible in trying to capture an infinite structure (like numbers) by finite means (e.g. recursively axiomatizable).

Let me just point out for other readers that Löwenheim-Skolem applies to ANY first order theory (in a countable language, or also in an uncountable language if stated in the form that a theory with an infinite model with cardinality at least that of the language has infinite models in all cardinalities at least as big as that of the language), it doesn't care about how complex the axioms are from a computability point of view

[Tainnor]

You're correct, but I think the spirit of what GP wrote is still true.

You can't capture an infinite structure fully by "finitary" methods, either you use FOL and then you run into L-S, or you use higher-order logic (for which L-S doesn't apply) but then you don't have a complete proof system anymore.

To tie it all together, L-S and incompleteness are about different flavours of "not being able to capture something". L-S is about models of different cardinalities. These models do still all satisfy exactly the same sentences. Incompleteness is about different models actually satisfying different sentences.

[Tainnor]

Löwenheim-Skolem is such a mind-blowing result. It's a shame it's not talked about more often.