Tweening is performed before lighting and clipping. The resulting vertex position (normal) is computed as follows:
POSITION = POSITION1 * (1.0 - TWEENFACTOR) + POSITION2 * TWEENFACTOR, and likewise for normals.