The map to the orbifold base need not be an orbifold map

Abstract

We give an explicit example of a fibration $f:X\to Y$ between smooth projective varieties whose "orbifold base" $\hat{I}{”}_f$ in the sense of Campana has the property that the induced morphism $X\to (Y,\hat{I}{”}_f)$ is not a morphism of C-pairs (i.e., it is not an "orbifold morphism"). We however also show that this cannot happen if $f$ is "neat" and $(Y,\hat{I}{”}_f)$ is sufficiently well-behaved. Finally, we discuss the implications of this statement towards conjectures of Campana aiming to give algebro-geometric characterizations of those varieties which either admit a dense entire curve or a potentially dense set of integral points.

13 pages, comments welcome!