This puts explanations on goals generated by some transformations with arguments. Ideally, no goals should be without explanations.
