A major factor supporting evolution over big up-front design is the drift in system requirements over time. Even on large military like projects, apparently there's "discovery"--and the more years that pass, the more requirements change.

This isn't my experience. Requirements tend to settle over time (unless they're stupidly written). Users tend to like things to stay the same, with perhaps some improvement to performance here and there.

But if anything, all development is the search for the search for the requirements. Some just value writing them down.

Even if the requirements are indeed fixed your understanding of the problem domain evolves.