Usually one must assign to the journal a more or less restrictive copyright to one's paper before being able to publish it. For math, this usually includes a s

Usually one must assign to the journal a more or less restrictive copyright to one's paper before being able to publish it. For math, this usually includes a stipulation that it be published nowhere else (except that many specifically allow posting to the arXiv, because—hurrah for the big guy being in the right!—it seems that no journal is big enough to fight the community consensus behind the arXiv).